Saturday, September 6, 2014

Good evening, Scotland. Yes takes the lead.

Tonight's breathtaking findings from the traditionally No-friendly pollster YouGov...

Should Scotland be an independent country?

Yes 51% (+4)
No 49% (-4)

Most media outlets are reporting the figures which take Don't Knows into account as Yes 47% (+5), No 46% (-2), although the YouGov website says Yes 47% (+5), No 45% (-3).  I would guess the former is most likely to be correct, in which case the lead on the unrounded figures with Don't Knows excluded is probably only just over 1%.

Nevertheless, a Yes lead is a Yes lead, and there was certainly nothing inevitable about us seeing one of those at any point before polling day.  Strictly speaking, this is actually the second time in the long campaign that a polling firm affiliated to the British Polling Council has shown a Yes lead, although it has become part of the orthodoxy to more or less disregard the previous occasion (a Panelbase poll from just over a year ago), because it's assumed that an unusual question sequence distorted the result.  Apart from that, we have to go back to well before the start of the campaign to find a poll from a BPC firm that showed a lead for independence - the most recent one was a TNS-BMRB poll in the late summer of 2011.

So does tonight's poll mean that Yes are 'really' in the lead?  Not necessarily.  Even before you take account of methodological mistakes that a pollster might be making, and also the fact that pollsters can't legislate (or not without difficulty) for respondents sometimes lying to them, there's a standard 3% margin of error that applies to every poll because of normal sampling variation.  So tonight's result is still consistent with No having a modest lead, and because of the almost unbelievable scale of the swing YouGov have reported over the last few weeks, you'd have to assume that's actually the most likely scenario.

But then again, even after the last YouGov poll on Tuesday, the significantly lower swing we were pondering at that point seemed implausible to me, and I was therefore fully expecting some kind of reversion to the mean in tonight's poll.  That hasn't happened, and that's what really matters, regardless of whether Yes are being slightly flattered by sampling variation.  Unless this is an out-and-out rogue poll (only a 5% chance), the worst position for Yes that the margin of error allows is Yes 48%, No 52%, which would still be better than the headline numbers in the previous poll.  So we can now say with almost absolute confidence that what we saw on Tuesday was not a fluke, or a fleeting post-debate "Salmondgasm" - it was a real and very hefty swing to Yes.

There is of course an important piece of supporting evidence tonight which might lead us to suspect that Yes are more likely to be on the lower end of the range allowed by the margin of error, and that's the new Panelbase poll commissioned by Yes Scotland -

Should Scotland be an independent country?

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

With Don't Knows taken into account, the position is -

Yes 44% (+2)
No 48% (+2)

On the unrounded numbers in the last published Panelbase poll (we know there's been an unpublished poll conducted by the firm in the interim), Yes were actually on 47.6% after DKs were excluded, so it's possible that when we see the datasets we'll find that they've gained a little and have narrowed the gap to the lowest level since Panelbase changed their methodology - but even if that's the case the movement will obviously have been very slight.

If this was a normal campaign, you'd look at the two polls tonight and think they were perfectly consistent with each other, both in terms of the headline numbers and the trend - you'd imagine there has maybe been a 2% swing to Yes, and that due to sampling variation Panelbase have underestimated it and YouGov have overestimated it. But the situation looks very different when you bear in mind the huge disparity there has been between various firms throughout this campaign, and the fact that YouGov have until the last few days been firmly on the No-friendly end of the spectrum, and that Panelbase have been firmly on the Yes-friendly end.  So it's an astounding paradox that YouGov have become the first pollster to put Yes in the lead in a credible poll, and that they've done it on the same night that a Panelbase poll still shows No in the lead, albeit only by a very narrow margin.

Let's try to take in the sheer extent of the gulf between the trends shown by Panelbase and YouGov that has been necessary to move us to the unlikely position where YouGov are, as of this moment, the more Yes-friendly of the two.  Just over a year ago, YouGov published a notorious poll which purported to show that Yes were on 33% and No were on 67% - that means there has been an 18% swing to Yes since then.  By contrast, in late August/early September of last year Panelbase were showing Yes on roughly 44% and No on roughly 56%, which would imply a far, far smaller swing of just 4% to Yes between then and tonight's poll.  One month ago, in a poll that partly took place immediately after the first leaders' debate, YouGov put Yes on 39% and No on 61% - meaning there has been a 12% swing to Yes in the space of a few short weeks.  The swing suggested by Panelbase over roughly the same timescale is just 2%.

To be fair, part of this disparity can be explained by methodological changes, and especially by YouGov's imperfect moves over the last year-and-a-bit to put their house in order.  They ditched their Dodgy Preamble, they moved away from weighting by Westminster-centric target figures for party identification (although that dreadful practice was directly succeeded by the No-friendly "Kellner Correction"), and much more recently they introduced weighting by country of birth. All of those changes are likely to have boosted Yes somewhat.  Over the same period, Panelbase have also made some modest methodological changes, the net effect of which may well have been slightly of assistance to No.  But even these moves to promote a degree of convergence can't explain the bulk of the extra swing to Yes reported by YouGov.

So what in the name of Foulkes is going on?  The most likely explanation is that the huge swing to Yes is something that is specific to YouGov, and that won't be fully replicated by other pollsters - or at least not by the other online pollsters.  The reason is that YouGov are effectively measuring something different to the others.  Specifically, there are two factors that set them apart from their three online rivals.  Most obviously, there's the Kellner Correction, which in practical terms leads to a sharp upweighting of the small group of respondents who voted Labour in 2010 but switched to the SNP in 2011.  In Tuesday's poll, that group was upweighted two-fold.  If the pro-Yes swing is concentrated in that group, the Kellner Correction will obviously be magnifying it, and therefore nothing on the same scale will be seen in the findings of the other online pollsters.  The other factor is that YouGov just seem to have far more Labour supporters on their books than the other firms do - that was one of the points Survation made in response to Peter Kellner's infamous diatribe.  So, again, if the swing is particularly significant among Labour supporters (as it seems to be), it would look bigger in YouGov's results.

The fact that Panelbase and Survation have published polls since the second debate which have been very good for Yes but haven't shown anything like such a big shift is consistent with the above theory.  However, there is an alternative possibility - which is that it's Panelbase which is different, and that we can therefore expect the other pollsters to be more in line with the trend shown by YouGov.  On the face of it, it's much harder to make the case that Panelbase are measuring something different to the others, because their methodology and the composition of their panel is much more in line with their online rivals ICM and Survation.  But in fact there is quite a bit of evidence that Panelbase have in the past produced trends that don't tally up with the overall picture.  In retrospect it's absolutely clear that there was a significant swing to Yes over the winter, and yet Panelbase were literally the only BPC pollster that didn't detect that at all.  Even when they did belatedly show movement in the spring, it wasn't on anything like the same scale that ICM and TNS-BMRB had picked up during the winter.  OK, it could be argued that the other firms were just catching up with a strength for Yes that was factored into Panelbase's figures from the word go - but why would that have been the case?  What makes Panelbase's sample so distinctive?  Whatever it is, that could - I only say could - be the explanation for why Panelbase are failing to show further progress for Yes this evening,

One other point intrigues me - why did Yes Scotland actually release this poll, given that it didn't show any further clear-cut breakthrough?  I can think of four possibilities -

1) They felt under pressure to do it, after the conspiracy theories started to mount about why they hadn't published last week's poll.  (And unfortunately we have to take our share of the blame for that.)

2) They were keen to highlight the swing to Yes among women in the poll.

3) They had heard about the YouGov poll, and decided to use the Panelbase poll to dampen down the hype, because being seen as a slight underdog works in their favour.  If there's any truth in that, it would mean Kenny Farquharson missed the point in the most deliciously ironic of ways - when he first heard about the poll results, he wrote a tweet accusing the Yes campaign of poor expectation management.  It could be that the fact that he heard about it in the first place was in itself part of a clever expectation management strategy.

4) They knew that the simple act of releasing the poll would help Yes in the Curtice Poll of Polls, which is based on a crude average of the last six polls to be conducted, regardless of which firms conducted them.  I haven't been able to get onto Professor Curtice's blog so far this evening, but I would expect to find that he'll have Yes at 47% in his new Poll of Polls - a higher figure than in the Scot Goes Pop Poll of Polls, which uses only one poll per firm.

*  *  *


Swing required for 1 out of 6 pollsters to show Yes ahead or level : 0.0%

Swing required for 2 out of 6 pollsters to show Yes ahead or level : 2.0%

Swing required for 3 out of 6 pollsters to show Yes ahead or level : 3.0%

Swing required for 5 out of 6 pollsters to show Yes ahead or level : 4.5%

Swing required for 6 out of 6 pollsters to show Yes ahead or level : 7.0%

* * *


I've got a slight problem in calculating this update of the Poll of Polls, because of the two different versions I've seen of the YouGov numbers with Don't Knows taken into account.  For the time being, I'll assume it's a one point lead for Yes, but obviously if that's wrong I'll have to correct the figures tomorrow.  (UPDATE : It turns out the Yes lead was two points, not one, so I have indeed had to correct the figures below - the average No lead is now 0.2% lower still.)

MEAN AVERAGE (excluding Don't Knows) :

Yes 45.9% (+0.9)
No 54.1% (-0.9)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 40.5% (+1.2)
No 47.8% (-0.2)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 45.7% (n/c)
No 54.3% (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 six - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori 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.)

The Naked Nation

A poem by Robert MacDonald

'We are the last North Britons'
typed the web 2.0 generation
casting off the robes of state
to reveal a naked nation.

A flick, a tug, a pull, a heave
and trickled information
runs free from sources wide to blogs
of soaring circulation

where once well-woven truths revealed
as threadbare barefaced lies:
and people stir and strain and talk
a' loosening their British ties.

Information is the virus now
the ballot-fouling killer pox.
Our keyboards are the carriers now
McCrone is out the box.

The robes of noble bluster stripped
not by conflict, war or strife
but by simple deeds and talking
of how we want to live our life.

"We'll govern our own country,"
said the grassroots conversation,
"our wee blue books are clout enough
to win. A better nation."

Friday, September 5, 2014

A few thoughts on the 'missing million' conundrum

On the last thread, Chalks pointed to the doubts cast by John Curtice on the claims from the Yes campaign that their pursuit of the votes of the "missing million" means they are picking up support which is being missed by the conventional opinion polls.  Curtice bases his scepticism on an analysis of the voting intentions of poll respondents who say they didn't vote in the 2011 election, and who he claims are actually shown to be more likely than others to be No voters.  Straight away, that sets a number of alarm bells ringing in my head, so here is your cut-out-and-keep-guide to why you should at least maintain a healthy scepticism about Curtice's scepticism...

Four out of six of the active pollsters in this campaign conduct their fieldwork among volunteer online panels.  One point I made in the interview for the Phantom Power film a few weeks ago (and which didn't make the final cut) is that to the extent that online pollsters have proved their credibility in recent years, you could easily make the old joke : "I know it works in practice, but does it work in theory?"  Polling among volunteer panels shouldn't work in theory, so the firms in question don't even worry about that - but they do go to great lengths to make sure their results are as accurate as possible in practice by 'working backwards'.  If a general election has just happened and a pollster's findings weren't quite right, they ask themselves how they could, for example, tweak their weighting procedures so that their raw data would fit the actual result.  If they also have evidence that the tweaked methodology would have produced reasonably accurate results in previous elections as well, then bingo, they've got a refined methodology that works in practice and will probably work in future elections as well.

But that's where they may be coming slightly unstuck with this referendum, because they don't have any baseline to work from to test that they are getting their methodology right - there have been no independence referendums before, and we know this contest is going to be radically different from normal elections, partly because of higher turnout and higher levels of voter registration.  That doesn't necessarily mean any given pollster is bound to be getting it wrong, but it does increase the degree of uncertainty (hence the unusual amount of variation between different firms' findings), and the biggest area of uncertainty for any online pollster will be people who usually don't vote, and people who haven't previously been registered to vote.  There simply aren't enough of those people "on the books" of volunteer online panels - in normal circumstances there don't need to be, because online pollsters only need to be right in practice, and in practice the 'missing million' don't count at all in normal elections.

So we should certainly doubt any analysis of the voting intentions of previous non-voters that draws too heavily on the findings of online firms.  There is also room for doubt with Ipsos-Mori, who by definition are only reaching people who are willing or able to answer a landline telephone, meaning there is no way of knowing whether enough of the 'missing million' are being contacted.  The one firm who in principle should be delivering the goods is TNS-BMRB, who actually go out into the real world and knock on people's doors - so they ought to be reaching people in the most deprived communities as easily as they are reaching John McTernan's butler.  And, unfortunately, it's true that TNS-BMRB have tended to be one of the more No-friendly pollsters (albeit usually not quite as No-friendly as Ipsos-Mori or YouGov).  But it also has to be borne in mind that their numbers go through a very unusual weighting procedure.  For example, here's what happened in the TNS poll conducted in June -

Only 226 people were found who said they didn't vote in 2011, but they were upweighted to count as 320 people.

124 people were found who couldn't remember how they voted in 2011, but they were upweighted to count as 173 people.

As far as I can see, the logic for upweighting both groups is the same - TNS seem to think that all of these respondents are representative of non-voters from 2011 (and indeed of previously unregistered voters).  Quite why people who don't recall how they voted should be automatically treated as abstainers is a bit of a mystery, and that's the first red flag we need to raise about the TNS approach, because in most cases it's the upweighting of the "forgetters" that is actually helping No the most.  If we assume for the sake of argument that many of these people did in fact vote in 2011 but genuinely can't remember how, then that significantly changes the picture that the TNS data is providing about the 'missing million'.

But the broader question is why such sharp upweighting needs to be happening at all?  I think there are two factors at play here.  Firstly, there must be a lower response rate to TNS polls among people who don't usually vote (ie. those people either won't answer their door, or will be more likely to turn the interviewer away).  So that introduces at least a degree of the same uncertainty that applies to the other pollsters.  The second factor is one that Professor Curtice hasn't even acknowledged, as far as I can see - it's likely that a significant minority of people are lying, and are telling TNS they voted in 2011 when they didn't, largely out of embarrassment.  Those people will presumably give as their "vote recall" the party they would have voted for if they'd actually made it to the polling station, or perhaps the party that in retrospect they'd like to think they would have voted for.  So there will actually be members of the 'missing million' who are telling TNS they voted SNP in 2011, and who are being weighted accordingly.

None of this is to say that a high turnout and a high level of voter registration is bound to favour Yes - but I do think Curtice's specific objections to that notion are not based on particularly solid ground.

*  *  *

A small correction to yesterday's post : Ivor Knox of Panelbase sent me an email earlier today to clarify that all of his firm's polls for the Sunday Times have been commissioned by the Scottish edition of the paper, whereas the upcoming YouGov poll has been commissioned by the UK edition.  So Panelbase haven't been sidelined, although Mr Knox stressed that he couldn't say whether they'd be conducting any further referendum polls for the Sunday Times.

That still leaves the mystery of who commissioned the Panelbase poll that has been in the field this week - perhaps we'll find out over the next couple of days.

*  *  *

This may not be an entirely original observation coming from me, but Political Betting's editor Mike Smithson hasn't exactly been covering himself in glory of late.  I've just spotted this tweet, which refers to a comment from Ivor Knox about protecting the client's right to confidentiality -

"My reading of @PanelbaseMD Tweet is that he has new IndyRef poll which client doesn't want to publish #BadforYES?"

If Mr Smithson had been paying attention, he'd know it was established beyond reasonable doubt several days ago that there was an unpublished Panelbase poll conducted last week, probably for the Yes campaign.  That may mean that Yes failed to get a significant further boost in that poll, but that's pure speculation, and even if it did happen to be true, the fieldwork ended three days earlier than in the breakthrough YouGov poll.  There has also been a second Panelbase poll in the field this week, but I know of no evidence that it has been withheld - indeed it may not even be completed yet.  I first heard about it on Wednesday.

UPDATE : There's confirmation from Teri in the comments section below that the fieldwork for this week's Panelbase poll didn't conclude until today (Friday), so Smithson's claim about a "new" poll being withheld is complete and utter nonsense.

And a special message for Neil Edward Lovatt (for I know you're reading this) : please stop using this blog as a bogus "source" for your eccentric rumour-mongering on Twitter.  You're a disgrace to "risk assessors" everywhere.

Run that past me again?

I got the latest of several emails from Tory millionaire grass-roots group "Vote No Borders" the other day (in spite of never having signed up for their mailing list!), and it was entitled "Why build another wall?"  Which is a good question, but an even better question is "Why not take that up with your colleagues in the No campaign such as Home Secretary Theresa May, who are the only people to have ever actually proposed anything other than a completely open border between an independent Scotland and the rest of the UK, albeit only as a transparent ruse?"

Oh, and a question I'd quite like to ask "Vote No Borders" myself is : I take it this inspiring vision of a world without borders also includes the border between France and England, which I had to tediously queue up to get through at Calais only a couple of weeks ago?  Admittedly it was quite fun walking up to the stern UK border guard while wearing my little Yes badge, but even so.  Mr Cameron, tear down this wall!

*  *  *

Alex Massie is continuing with his "equal opportunities irritant" approach to referendum blogging, with one or two posts designed to get on the nerves of No supporters being reliably followed by one or two posts designed to get on the nerves of Yes supporters, and so on into infinity.  Which is fair enough, but he gave the game away when in a perhaps unguarded remark at the end of his latest piece, he described himself as a "normally jaded unionist".  Question : why did the BBC allow any sort of "unionist" to take part in a high-profile TV debate a few months ago on the specific basis that he was an "undecided voter" in a referendum on independence?

*  *  *

Stephen Daisley's irreverent articles for the STV website have been a breath of fresh air in recent weeks, but if his piece on the "town hall debate" the other night is anything to go by, let's hope to God he never becomes ITV's political editor.  He seems to be suffering from the Tom Bradby syndrome of projecting his own political instincts onto an imaginary "average voter", and therefore assuming that absolute personal bias constitutes perfect objectivity.  Apparently normal people were all utterly bored by what Patrick Harvie said about the prospect of children being burned to death by the UK's Trident nuclear warheads.  Well, if you like, Stephen, we could always pretend that Trident sings children a lullaby to gently send them to sleep - it would be a lie, but a pretty lie all the same.  The reality is that the Hiroshima bomb, which was many, many magnitudes less powerful than a modern British nuclear weapon, incinerated countless children as they sat in school classrooms.

Ironically, Stephen concludes his article by charging both campaigns with ignoring the big issues of debt and the affordability of pensions.  Forgive me, but isn't the avoidance of a nuclear holocaust, and the virtual extinction of the human race it would entail, a weightier issue still?  Crushingly boring to you, Stephen, I know, but...

Oh, and Stephen also praises Scottish Tory leader Ruth Davidson to the skies for her "moving words" about the military.  So no problem at all with her gaffe about the UK needing nuclear weapons to "stand tall in the world", ie. weapons of mass destruction justified purely on the grounds of national prestige.  Just remind me - what did we say we were invading Iraq for?

*  *  *

The reaction of a Scottish Labour spokesman to the decision of the RMT union membership to back a Yes vote -

"Six trade unions, representing 130,000 people across the public and private sector in Scotland, are in favour of a No vote on September 18."

Yes, those unions may represent 130,000 people, but how many of those 130,000 people were actually balloted before the unions decided their stance?

"Even this ballot did not show a majority in favour of independence - 56% of people who voted didn't support a Yes vote."

'Even' this ballot?  As opposed to all the other ballots that didn't take place, I suppose?  In case you're wondering, the 56% figure refers to the fact that there was also a "no position" option on the RMT ballot - 44% of members voted Yes, 41% voted No, and the remaining 15% voted for no position.  That does of course mean that 59% didn't support a No vote, which I can only presume must be EVEN WORSE than 56% not supporting a Yes vote?

Thursday, September 4, 2014

What it's like going on a day-trip to Arran while wearing a bloody enormous Yes badge

For months now there's been a nagging voice in my head telling me that I should at least make a token effort to volunteer with the local Yes campaign in Cumbernauld.  But deep down I think I've always known that I'm not one of life's natural canvassers (ie. there's a risk I could make matters worse rather than better!).  Early last month, though, I decided to split the difference and do what I should have done ages ago - turn myself into a walking advertisement by ordering some badges from the online Yes Scotland shop.  When they arrived, I realised to my horror that I was going to have to choose between an enormous badge that would make me feel very self-conscious, and a small one that would require you to practically be standing on top of me before you could decipher the word written on it.  I used the small one while I was travelling on the continent, because I knew it wouldn't make much difference anyway - but I gradually resigned myself to the thought that I was going to have to switch to the enormous one when I got home, otherwise the whole exercise would be pointless.

I first took the plunge when I spent the day in Edinburgh last Thursday, and suffice to say the difference was palpable.  It was like I was constantly making a verbal statement, and people felt entitled to 'reply', which is a bit startling when in your own mind you're just minding your own business as per usual.  When I got back to Glasgow late that night I went into a 24-hour shop, and the man at the counter immediately asked me to give him one reason why Scotland should be independent.  It took me completely by surprise, and I had to admit to him that I was very tired (of course three seconds after leaving the shop I thought of all the things I should have said).  Thankfully, he nodded to me anyway and said : "I think you're right".

Yesterday I took things a step further and wore the badge throughout a day-trip to Arran.  It got the following reactions -

* A Caledonian MacBrayne employee asked me where I got the badge, and said "good" when I told him it was from the Yes Scotland online shop.

* An Englishman on the island shouted "Yes!" and gave me a big thumbs up as I passed, which was an absolutely lovely show of solidarity (unless of course he's an Arran resident, in which case he's probably a Yes voter himself).  But again, it took me completely by surprise when I was lost in thought about something else, and I just smiled awkwardly and kept on walking, which I think disappointed him slightly!

* As I was waiting to get back on the ferry as a new batch of foot passengers came off, one of them looked at me as if he'd just seen something utterly revolting (entirely understandable, to be fair) and very loudly said "OH MY GOD VOTE NO", in a Gary: Tank Commander sort of voice.

* When I got back to Glasgow, a slightly drunk young woman mistook me for a campaigner and begged me to give her a Yes badge.  She didn't strike me as the sort of person who would normally be passionate about politics (or even necessarily turn out to vote) so that probably tells you something important.

* I then walked into a very similar 24-hour shop to last week's, and the man at the counter eyed me for a couple of seconds before grinning and saying in a thick Pakistani accent : "YES -  FOR - SCOTLAND!"

The one and only time when I deliberately tucked the badge away was when I reached a spot called Giant's Graves, which is the location of the remains of two Neolithic tombs.  It was obvious even from a distance that something slightly peculiar was going on there, and I had a feeling I was about to have an awkward enough conversation without bringing the referendum into it.  At first I thought it was some kind of hippy ritual, because a woman was lying down in one of the tombs, while a man stared into space intensely.  When I got there, the woman peeked over the edge of the stone and said "Hi!", in a tone of voice that you'd put on if you were trying to sound extremely friendly while wishing to God the other person would go away.  I said "hiya, am I interrupting?"  Her response was "not yet!", which she said in such a mischievously guilty sounding way that I almost started to wonder if they were planning to have sex in the tomb.  But to my relief I then noticed some fancy-looking equipment lying around, so I think it must have been a photo-shoot.  The woman was good-looking and unnecessarily well-groomed for the surroundings, so she was probably a model and the guy was probably the photographer.  It was very irritating, though, because I had walked two miles uphill to get there, and you don't really expect to be in the back of beyond and STILL find yourself in someone's way!

What we know about upcoming polls

People are always asking me "when is the next poll due?", and my answer is usually that I don't have a clue.  In fact, my answer is no different today - but we do have an unusual degree of clarity about a few polls that can be expected to appear over the next week or so.

TNS-BMRB say their next poll will be released in the middle of next week.  Their fieldwork is always more out-of-date by the time of publication than any other firm, but hopefully this timetable means that a decent chunk of the fieldwork will have been done after the second leaders' debate.

YouGov seemingly have two polls coming up in quick succession - one on Sunday for the Sunday Times, and one early next week for the Sun.  (Unless the original plan for a poll in the Sun was scrapped in favour of the Sunday poll.)  I must say I find it profoundly depressing that the Sunday Times have seemingly ditched their previous house pollsters Panelbase, who were presumably too Yes-friendly for them at this crucial stage of the campaign.  Let's hope no sneaky McDougall-esque apple-and-oranges comparisons will be made in Sunday's paper.

Panelbase have had another poll in the field this week.  Our old friend Mike "can't be arsed" Smithson claims to have been told that this one will be published, which on the face of it is inconsistent with the client being either Yes Scotland or the SNP, because they would presumably wait to see the results before deciding whether to publish.  My first thought was that it must be a Sunday Times poll, but that was before I heard the news about the ST switching to YouGov.  So a bit of a mystery.

Meanwhile, a dog's breakfast of a poll has been reported in the Telegraph, apparently conducted by an outfit I've never heard of before called TubeMogul.  Every instinct in my body tells me it's a voodoo poll with no scientific validity (ie. self-selecting respondents), but I can't say that for sure because no meaningful information is given about methodology.  At face value, though, it's an excellent poll for Yes - there's a 50/50 split among women and a 44/56 split among men, which presumably averages out as Yes 47%, No 53% - exactly the same as reported by both YouGov and Survation.  Needless to say, the Telegraph have somehow managed to convert this into a "disaster for Salmond" narrative by pretending that only the tiny subsample of 16 and 17 year olds actually matters!

Brit Nats and egg-tastic distraction tactics

I have a new article at the International Business Times about the eggstraordinary way in which the media lapped up Jim Murphy's distraction tactics a few days ago - you can read it HERE. (It's also on Yahoo HERE.) In case you're wondering, the title I suggested was 'Supporters of independence are just the boy or girl next door'. Credit is due to Mick Pork for (I presume!) originally coining the term "eggpocalypse", and obviously the sentiment in the closing paragraph is inspired by comments made by Scottish Skier.

*  *  *

I got a direct message on Twitter yesterday which confirmed the suggestions that there is (or has been) another Panelbase poll in the field - this time it's not quite so clear to me who the client is.

I can't really shed any light on the claims and counter-claims swirling around about the unpublished Panelbase poll that was conducted last week (probably for Yes Scotland or the SNP), except to note that there was a single tweet on Saturday night, which went strangely unremarked upon, saying that there were rumours that the Sunday Herald were about to publish a poll showing a 50/50 split.  I did wonder for an hour or two if there might be something in that, but obviously when the front page appeared with no mention of a poll, I just concluded it was completely baseless.

Wednesday, September 3, 2014

Wisdom on Wednesday : It's this simple

"For too many years of my life I have lived under Westminster governments who neither I, nor the majority of Scottish people voted for, and I fail to understand how this is acceptable."

The legendary Shetland fiddler Aly Bain.

Monday, September 1, 2014

Scotland stands on the brink of independence as Yes vote surges to 47% in historic poll from traditionally No-friendly firm YouGov

Well, it looks like we can stop wondering whether Alex Salmond's victory in the second TV debate had an impact on public opinion. Here are tonight's sensational findings from YouGov -

Should Scotland be an independent country?

Yes 47% (+4)
No 53% (-4)

And when Don't Knows are taken into account, it's...

Yes 42% (+4)
No 48% (-3)

Let's try to put this into some kind of perspective - although given the scale of what appears to be happening, it's difficult to know where to start. Throughout this campaign, YouGov have consistently vied with Ipsos-Mori for the title of the most No-friendly pollster. At one point, just over a year ago, they showed figures of Yes 33%, No 67% - an almost unique example of the London media's beloved "2-1 margin for No" narrative actually becoming reality. Immediately after the SNP's victory in 2011, the President of YouGov, Peter Kellner, made the daft claim that it was literally impossible for Yes to win a referendum that was still over three years away, and since then it's been murderously difficult to escape the impression that he's been (perhaps subconsciously) moulding his firm's methodology to bolster that ludicrous prediction. And, yes, the No-friendly methodology remains in place in tonight's poll - we already have confirmation that there has been no change in methods since the last YouGov poll. So if Yes are on 47% after the so-called Kellner Correction was applied to the figures, the mind boggles as to what the results would be without it. Based on the traditional differential between the various firms, this poll is perfectly consistent with a narrow Yes lead among the more Yes-friendly pollsters such as Survation and Panelbase.

But that last point offers a clue as to why we should still exercise a little caution, because of course the only other published poll to be conducted since the second debate was from Survation, and although it was very good for Yes (an equalling of their all-time best showing) it didn't put them in an outright lead. So it's always possible that the YouGov poll might prove to be an outlier, and that the swing to Yes isn't quite as big as it appears. However, it's very hard to believe that there hasn't been a swing to Yes of some description, and from the comments that have been attributed to Peter Kellner tonight, it appears that even he agrees with that. He claims to have checked that the movement is real and concluded that it is. Doubtless he'll be careful to base that analysis only on the published results, but to my mind it's hugely significant that we know there was another YouGov referendum poll in the field over the last few days - perhaps a Better Together internal poll, perhaps a poll for an academic study. So Kellner will have lots of information to go on in private, and if none of that is causing him to doubt tonight's swing, the conclusion to draw is obvious.

Let's not forget that there was also a poll in the field over recent days conducted by Panelbase, which appeared to be an internal poll for Yes Scotland. I initially assumed that the reason it wasn't passed to the Sunday Herald for publication was that it didn't show any further progress for Yes, but the events of tonight at least raise the possibility that it was actually withheld because the Yes campaign were keen to maintain their underdog status, which appears to be working so well for them. However, that will have to remain an open question for now.

Tonight's timing, of course, could hardly be any better, as this is the first YouGov poll that is partly an exit poll - a significant minority of respondents will have been telling the firm how they have already voted, rather than merely how they plan to vote. So at worst, if YouGov's No-friendly methodology proves to be accurate, it appears that we no longer have to worry that the fact that voting is already underway might put Yes at a significant disadvantage, and leave them needing well over 50.1% of the votes cast on polling day. At best, if YouGov's methodology is painting too pessimistic a picture for Yes, it could be that it's actually a small advantage for us that the referendum has started.

If this poll doesn't prove to be an outlier, probably the one lingering concern might be that we're merely seeing a temporary post-debate bounce. But my gut feeling (for what it's worth) is that Alex Salmond's performance was the equivalent of what we always say about North Sea Oil - it was just a fantastic bonus, not what we rely upon. The greatest amount of persuasion has been done on the ground.

To give a small amount of credit to YouGov, they're one of only two pollsters that have taken the eminently sensible step of introducing country of birth weighting, to correct for the fact that there always seems to be too many English-born people in their samples. But they're also one of only two pollsters (the other is TNS) that do not filter or weight their headline numbers by likelihood to vote. Quite often (but categorically not always) that practice slightly suppresses the reported Yes vote. We'll have to wait for the datasets to see if that has been the case in tonight's poll - last time around the No lead was 2% lower among definite voters than it was in the published numbers.

Finally, I should just give some comparison figures for people who don't fully appreciate just how No-friendly YouGov have always been. Before tonight, their highest ever figure for Yes after Don't Knows are excluded was 43% - and that was only reached in the last poll. That record has been smashed by an almost unbelievable 4%. The previous record for Yes with Don't Knows taken into account was 38%, which again was only reached in the last poll, and which has also been surpassed by 4%. And for three successive YouGov polls between mid-June and early August, Yes were only hovering at 39-40% after Don't Knows were excluded - they're now 7-8% higher than that.

* * *


Swing required for 1 out of 6 pollsters to show Yes ahead or level : 2.0%

Swing required for 3 out of 6 pollsters to show Yes ahead or level : 3.0%

Swing required for 5 out of 6 pollsters to show Yes ahead or level : 4.5%

Swing required for 6 out of 6 pollsters to show Yes ahead or level : 7.0%

* * *


You won't be surprised to hear that on all counts, tonight's Poll of Polls update sees Yes moving to an all-time high. By far the biggest swing is on the median average, which comes about because YouGov have (albeit almost certainly only temporarily) moved to the Yes-friendly end of the scale.

MEAN AVERAGE (excluding Don't Knows) :

Yes 45.0% (+0.6)
No 55.0% (-0.6)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 39.3% (+0.6)
No 48.0% (-0.5)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 45.7% (+2.0)
No 54.3% (-2.0)

(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 six - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori 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.)

More poll drama tonight as YouGov look set to report yet another record high for Yes

The referendum polling drought will be abruptly broken at 10pm tonight, when new numbers from YouGov (the pollster notorious for adjusting their results with the No-friendly "Kellner Correction") are released. If a tweet from Nick Robinson and a comment in Brian Taylor's blog are anything to go by, they'll show a further narrowing of the gap - and that's absolutely massive news, because the last YouGov poll already showed a record high for Yes of 43% among the whole sample, and of 44% among definite voters. And it gets even better - our old friend Laurence Janta-Lipinski has been dropping broad hints on Twitter that the shift is significant, so there could well be an increase in the Yes vote of more than 1%. If so, we're getting into territory that would be consistent with a dead heat or a small Yes lead with the more Yes-friendly pollsters.

Stay tuned...

Sunday, August 31, 2014

Calm before the storm?

I'm mildly taken aback by the lack of polls this evening, partly because this referendum is rather a big deal and we're less than three weeks away from it, and partly because we know for a fact that there have been YouGov and Panelbase polls in the field over the last week. From the way the Panelbase questions were described to me, that poll bore all the hallmarks of a Yes Scotland internal, and therefore the fact that it hasn't been passed to the Sunday Herald for publication may indicate that the Yes vote hasn't reached another all-time high (ie. 49% or higher). However, we certainly shouldn't jump to the conclusion that it's a bad poll, because even if it had merely equalled the previous record of 48% (which in truth would be a fantastic result after Panelbase's recent methodological change), it would probably have been withheld due to the danger of the No campaign saying : "Yes Scotland's own poll shows that they have failed to gain support since the debate, blah, blah, blah".

Equally surprising is the silence from YouGov, because we know they have been conducting at least two independence polls over recent days. One of them used question wording that is typical of not-for-publication Better Together internal polls, although the epic length led some people to wonder if it might be for an academic study. But the other poll had "for a newspaper" written all over it, and I fully expected to see it tonight. Perhaps it'll still turn up at some point over the next couple of days.

While we have this little lull, I thought I'd amuse myself by conducting an experiment. As you probably know, Panelbase and YouGov recently took the very sensible step of changing their methodology to correct for the fact that they have a disproportionately high number of English-born people in their samples. ICM, Survation and TNS-BMRB have thus far failed to follow suit, but there's something of a mystery (in my mind at least) about Ipsos-Mori's position. They do always ask for their respondents' country of birth, and they give that question the grand classification of "DEM 6", which implies that they weight by it - and yet their weighted numbers don't seem to be in line with the census results. So just out of interest, I did a rough reweighting of Ipsos-Mori's most recent referendum poll, using YouGov's census-derived target figures for country of birth. It had a slightly bigger effect than I expected - it increased the Yes vote (after Don't Knows are excluded) by a full 1%, and thus decreased the overall No lead by a full 2%. That may not be a dramatic transformation, but it's certainly not to be sniffed at.

One other thing that intrigues me is that Ipsos-Mori's small sample of respondents who were born outside the UK and Ireland is relatively favourable for Yes, and produces a No lead that is smaller than exists even among Scottish-born respondents (after the certainty to vote filter is applied). If memory serves me right, that finding is not untypical of previous Ipsos-Mori polls, and yet it's in complete contrast to both Panelbase and YouGov, both of whom show huge No leads among people born outside the UK. Of course, what makes Ipsos-Mori unusual in this campaign is that they're not reliant on volunteer online panels, and given that you could easily imagine that there might be special problems in recruiting non-native-English speakers to those panels, you'd be forgiven for wondering whether the people that Panelbase and YouGov do have are really representative. So that's yet another potential area of uncertainty.

* * *

UPDATE : A couple of people have mentioned on Twitter that they've been interviewed by Ipsos-Mori over the weekend - one of them tried to find out who the client was, but to no avail. Please God let it just be a coincidence that STV have another big debate coming up on Tuesday night. Are they never going to bloody learn that the purpose of a debate is to help people decide how to vote based on the arguments, not to tell them at the outset how they're already voting? (And through the distorting lens of one of the most No-friendly polling firms at that.)