Saturday, April 5, 2014

Panic stations for anti-independence campaign as Panelbase poll confirms swing to Yes

When a polling organisation produces a poll showing something out of the ordinary, as Panelbase did last month by showing the No campaign ahead by just 53% to 47%, the most important question is whether the next poll will confirm the findings, or show that normal service has been resumed.  To the horror of Mr McDougall's merry mob, the answer in this case is the former.  Here are tonight's numbers with Don't Knows excluded -

Should Scotland be an independent country?

Yes 47% (n/c)
No 53% (n/c)

*  *  *

1:05am : RevStu has now released some more details of the poll on Wings Over Scotland.  The headline numbers are as follows -

Yes 41% (+1)
No 46% (+1)

The percentage changes above assume that the headline numbers are directly comparable with last month's from Newsnet Scotland.  However, I do have a slight question mark in my mind over that, because in the last Wings poll RevStu headlined the voting intention figures from the whole sample, as opposed to the figures filtered by likelihood to vote (which are the ones that Panelbase normally go with).  I'm not quite sure what the position is this time, and it also remains to be seen which preamble was used.  Hopefully Panelbase will have stuck with their new, more neutral preamble, but if by any chance they've gone back to the old one, the last comparable Panelbase poll would in fact be the one in the Sunday Times that showed No ahead by a full 12%.

In one sense, though, it makes no difference which numbers are being headlined, because either way it's the highest Yes vote that Panelbase or any other pollster have shown since the famous poll commissioned last August by the SNP that had Yes in an outright lead.  If we exclude the August poll due to the methodological criticisms that were made of it, then tonight sees Yes reach its highest level of support in any poll conducted by any polling organisation at any point in the campaign so far.  That's how big this moment is.  It's also a finding that is very much in line with the most recent polls from ICM, Survation and YouGov, each of which showed the Yes vote at a record-breaking level for the pollster in question.

The No lead remains on 5% for a second Panelbase poll in a row, meaning that it is still significantly below the firm's traditional range of 8-13% (and that normal range would probably be a touch higher if you factor in the methodological change that was made towards the end of last year).  It's highly unlikely that there have been two rogue polls in succession, so we can now dispense with any caveats and say with complete confidence that Panelbase have belatedly joined the polling consensus showing a drop in the No lead.

A crucial detail that has yet to be revealed (although it may be in the paywall-protected Sunday Times piece) is the fieldwork dates.  If those dates even partly precede the leak in the Guardian confirming that Osborne's speech on the currency was a bluff, then that would be highly significant, for the simple reason that most respondents to online polls answer the questions as soon as the email invitation is sent out.  In other words, there would be a very plausible scenario you could paint of the Yes vote rising even further since this poll was conducted.  But we'll have to wait and see - obviously that wouldn't apply if the fieldwork took place entirely after the Guardian story broke.

* * *


I seem to have been saying this rather a lot recently, but the Yes vote has just risen to yet another new high watermark on the headline average of this blog's Poll of Polls. And on the mean average with Don't Knows excluded, the Yes vote remains at the high watermark of 42.5% it reached in the last update.

MEAN AVERAGE (not excluding Don't Knows) :

Yes 35.7% (+0.1)
No 48.3% (+0.2)

MEAN AVERAGE (excluding Don't Knows) :

Yes 42.5% (n/c)
No 57.5% (n/c)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 42.0% (n/c)
No 58.0% (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, and that adhere to British Polling Council rules. At present, there are seven - YouGov, TNS-BMRB, Angus Reid, 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.)

Here are the updated long-term trend figures -

The No campaign's lead in the Poll of Polls headline figures :

Sep 2013 - 20.2%
Sep 2013 - 20.0%
Sep 2013 - 18.4%
Oct 2013 - 17.9%
Oct 2013 - 17.5%
Oct 2013 - 17.4%
Nov 2013 - 17.5%
Dec 2013 - 17.1%
Dec 2013 - 16.3%
Dec 2013 - 16.2%
Dec 2013 - 15.8%
Jan 2014 - 14.2%
Jan 2014 - 14.8%
Feb 2014 - 14.8%
Feb 2014 - 14.7%
Feb 2014 - 15.1%
Feb 2014 - 13.6%
Feb 2014 - 14.0%
Mar 2014 - 14.0%
Mar 2014 - 14.3%
Mar 2014 - 14.3%
Mar 2014 - 13.6%
Mar 2014 - 12.9%
Mar 2014 - 13.0%
Mar 2014 - 12.5%
Apr 2014 - 12.6%

In case you're wondering, the apparent 0.1% increase in the No lead in this update isn't real - it's just a quirk of the rounding.

As Panelbase are an online pollster, it's also time for an update in the averages for the four online pollsters that have reported so far this year (the other three are YouGov, ICM and Survation).

MEAN AVERAGE OF ONLINE POLLSTERS (not excluding Don't Knows) :

Yes 39.0% (+0.2)
No 48.0% (+0.2)


Yes 44.8% (n/c)
No 55.2% (n/c)


Yes 45.4% (n/c)
No 54.6% (n/c)


  1. Stu said poll was partly done before "senior cabinet minister gate".

    He also said Yes was on 41% with the don't knows included. Which is fairly awesome for now.

  2. Thanks, Rolfe, I already had the 41% figure in the post, but the news about the fieldwork dates is very encouraging.

    Presumably we'll find out whether the headline numbers were filtered by likelihood to vote (and what the preamble was) when the datasets are released.

  3. I see the Unionists are droning on today about how Panelbase is infiltrated. Yes, Panelbase is the most YES friendly pollster and a little bit of an outlier, but their figures aren't a million miles away from ICM or Survation. Panelbase aren't anywhere near as big of an outlier as Ipsos MORI.

  4. Absolutely, this poll is almost identical to the last ICM poll (the No vote exactly the same, the Yes vote a couple of points higher). For all we know YouGov would be showing broadly similar figures if they used the same weighting procedure.

    The conspiracy theory about Panelbase was debunked months ago when they used a sample from another pollster's panel, and produced exactly the same results.

  5. The indy question tables are up here
    The results have been filtered by the 8-10 likelihood to vote, so are comparable with the NNS poll.