Saturday, March 19, 2016

Doomsday dawns for distraught Dugdale as splendidly sadistic Survation survey extinguishes all hope for Scottish Labour

The latest poll in Survation's monthly series for Cleggy and the Vow-Meisters is now out, and it shows Labour slumping to joint all-time lows for Survation polls in both the constituency and list ballots.

Constituency ballot :

SNP 54% (n/c)
Labour 20% (-1)
Conservatives 16% (n/c)
Liberal Democrats 7% (+2)

Regional list ballot :

SNP 42% (-1)
Conservatives 18% (+4)
Labour 18% (-1)
Greens 10% (+1)
Liberal Democrats 6% (-1)
UKIP 5% (-1)

The Record bills this poll as predicting that Labour will finish third in terms of seats behind the Tories, but that isn't strictly true - even leaving aside the fact that polls don't make predictions, all that the poll itself says is that the Conservatives have drawn level with Labour on the list vote.  Pumping the figures into one particular prediction website may give the Tories one more seat than Labour, but it's debatable whether any prediction model can say anything meaningful about what would happen in the event of a dead heat.  The outcome would mostly depend on the regional distribution of each party's list votes.

There are also two important facts we know to be true - a) the list vote for major parties is generally very strongly correlated to their constituency vote, and b) polling for the constituency ballot tends to be more accurate than polling for the list ballot.  So it could well be that Labour's 4% advantage over the Tories on the constituency ballot tells us more about what is likely to happen on the list than the list figures themselves do.

Predictably, the Greens' spin machine has cranked into gear over their showing on the list, but actually this poll isn't as great for them as it appears.  Survation have consistently been the most favourable pollster for both the Greens and UKIP, and that appears to be due to the inadvisable way in which respondents are asked for their list voting intention.  10% is well within the Greens' recent 'normal range' with Survation - they were as high as 11% in September.

Explanatory note : As a fond tribute to the mainstream media's restrained take on the GERS report, Scot Goes Pop headlines will contain 50% added hysteria for an indefinite period.

Wednesday, March 16, 2016

Drama as Leave take the lead for the first time in a telephone poll

In spite of the lavish coverage of Lynton Crosby's grandiose musings on yesterday's EU referendum poll from ORB, I didn't actually pay a huge amount of attention to the poll itself, because at first glance the numbers looked fairly routine.  But as soon as you realise that it was conducted by telephone and not among a volunteer online polling panel, suddenly it looks a hell of a lot more significant.

Remain 47%
Leave 49%

This isn't quite the game-changer that it might have been if it was a ComRes, Ipsos-Mori or Survation telephone poll putting Leave ahead, because that would be fairly convincing proof that there has been a swing to Leave.  Since ORB haven't previously polled by phone during this campaign, it's possible that there's some kind of house effect that is favouring Leave in this poll, and that if the other telephone firms polled again, they'd produce much more familiar numbers.  But there's no getting away from it - the fact that it's even possible for any sort of telephone poll to put Leave ahead completely transforms our understanding of this contest.  Until now, the telephone leads for Remain have ranged from substantial to overwhelming, and that's been the comfort blanket that has reassured many pro-European commentators that Brexit is highly unlikely.  Suddenly it doesn't look quite so simple.

Probably the biggest health warning that needs to be slapped on this poll is that ORB don't seem to have asked the exact referendum question.  (I'm not 100% sure about that, because the datasets are quite difficult to decipher, but it looks that way.)  We know from Ipsos-Mori's research that the official question tends to produce better numbers from Remain - which is a bit odd, given that the wording is scrupulously neutral.  So perhaps Leave are being flattered slightly by ORB, but even allowing for that, it's hard to deny that this is the Outers' most encouraging phone poll to date by some distance.

Where does this leave us?  Completely in the dark, basically.

* YouGov haven't reported recently, but their most recent batch of polls seemed to suggest a significant swing to Remain among their online panel.  However, a more up-to-date full-scale Scottish poll from YouGov showed a big swing to Leave - it's unclear whether that's a Scotland-only effect, or whether it reflects a very recent shift in Britain-wide opinion that will only be picked up by the next UK poll from the firm.

* ICM are the other most prominent online pollster, and you could make a case that they're either showing no meaningful recent change, or a small swing to Leave.  But there's certainly no sign at all of a YouGov-style swing to Remain.

* Neither ComRes nor Ipsos-Mori have published a telephone poll recently.  The last time they did, ComRes seemed to be suggesting that Leave had made some progress, while Ipsos-Mori showed no change of any significance.  But both firms still showed substantial Remain leads.

* And now we have the first telephone poll to put Leave ahead, but with no baseline figures from ORB to measure from.

It's as clear as mud, but on balance I'd suggest Leave have reason to be more confident than they were a few days ago.


Should the United Kingdom remain a member of the European Union or leave the European Union?


Remain 45.8% (-0.1)
Leave 39.7% (+1.1)


Remain 41.2% (+0.3)
Leave 40.0% (+0.2)


Remain 50.3% (-0.5)
Leave 39.3% (+2.0)

(The Poll of Polls takes account of all polls that were conducted at least partly within the last month. The online average is based on ten polls - five from YouGov, three from ICM, one from ORB and one from BMG. The telephone average is based on four polls - one from ComRes, one from Ipsos-Mori, one from ORB and one from Survation.)

*  *  *

You learn something new every day - according to the Telegraph, members of the House of Lords will be allowed to vote in the EU referendum, but only if they live in Gibraltar.  That sounds eminently sensible.

*  *  *

Many thanks to the 227 people who voted in this blog's latest readers' poll.  It may be somewhat academic after Hillary Clinton's solid results overnight, but no fewer than 214 of you (that's 94.3%) reckoned that Scotland's favourite "centre-right socialist" Stephen Daisley would vote for Donald Trump in a hypothetical contest against Bernie Sanders.  Evidently you think "right" is the operative word, rather than "centre" or "socialist".

Tuesday, March 15, 2016

VOTE : What would Stephen do?

Alas, it's probably unlikely that Bernie Sanders will win the Democratic nomination, although he does still have a genuine fighting chance if he can follow up his sensational Michigan victory with some good results tonight.  For a bit of fun, let's suppose that he does make it through, and that Donald Trump is the Republican nominee.  I'm sure I'm not alone in wondering - what would Stephen Daisley do if he had a vote?  That's the subject of today's readers' poll.

There's not much doubt that Daisley would vote for Hillary Clinton given half a chance - she's like Liz Kendall with charisma, and that's a heady combination for our Stephen.  But could he bring himself to vote for Sanders over a madman like Trump?  Let's consider the evidence...

1) Stephen thought that Jeremy Corbyn becoming Labour leader, on a very similar platform to Sanders, was a catastrophe on a par with the Black Death.

2) In a series of bizarre tweets a couple of weeks ago, Stephen seemed to indicate support for "constitutional originalism".  In other words, he thinks justices on the US Supreme Court should not interpret the constitution as a living document that adapts with the times, but should treat each word and phrase in the precise way that was intended in the 18th century.  That suggests he would prefer the sort of conservative justices that a Republican president might appoint, rather than the liberal justices that even Clinton might appoint.

3) In a jaw-dropping article published on the STV website in the wake of the Paris attacks, Stephen seemed to strongly hint that the torture of terror suspects might be morally acceptable in some circumstances.  Donald Trump has of course expressed his support for waterboarding, and for targeting the families of terrorists.

Using your skill and judgement, I want you to guess what Stephen would do if he had a vote in a hypothetical Sanders v Trump contest.  You'll find the voting form at the top of the sidebar (desktop version of the site only).