Friday, February 7, 2014

Support for independence rises to its highest level of the campaign in latest YouGov poll

Just when we were feeling mildly disappointed that YouGov had broken the seemingly relentless sequence of polls since the publication of the White Paper to show clear swings to Yes (of the small, medium and huge varieties), along they come right on cue with another poll in which they rejoin the consensus.  At 34%, the pro-independence vote is now at its highest in any YouGov poll since October 2011, and at 18%, the lead for the anti-independence campaign is at its lowest since the same date.  Of course the question used back in 2011 bore no resemblance to the current one, so it would be more meaningful simply to say that the Yes vote is now at its highest level since the campaign got underway.

If there was a referendum tomorrow on Scotland's future and this was the question, how would you vote?

Should Scotland be an independent country?

Yes 34% (+1)
No 52% (n/c)

Unlike several other pollsters, YouGov only post percentages in their datasets, rather than the raw numbers.  If you pump the crude figures 34 and 52 into a calculator, you'll find that the position with Don't Knows excluded is Yes 40%, No 60%.  However, YouGov say that it's actually Yes 39%, No 61%, which means rounding issues must be at play.  With a filter applied for likelihood to vote, though, support for Yes does indeed rise to 40%, and support for No falls to 60%.

I'm glad to say that the information released this time is detailed enough for us to be sure that the notorious Dodgy Preamble was absent from this poll.  However, to the best of my knowledge we're still no clearer about the preamble used in the poll published earlier this week (ie. we can't be sure that the polls are directly comparable).

It may or may not be due to their methodological antics that YouGov are showing a much more confusing picture in respect of the trend than other pollsters.  One thing that is clear is that the No lead has fallen since the late summer/early autumn, in line with the findings of ICM, Ipsos-Mori and TNS-BMRB.  But by how much?  Since YouGov's August poll (the last published YouGov poll that we're 100% sure used the Dodgy Preamble), the No lead has slumped by 12% - an even bigger drop than that recorded by ICM.  But since the September YouGov poll (the first to use the more neutrally-worded preamble) the No lead has fallen by just 2%.  What confuses the issue even further is that the first of the semi-published 'phantom' YouGov polls, conducted in late November/early December, apparently suggested that the No lead had gone back up again by 4%.  That seems highly implausible given what other pollsters were showing at the same time, so perhaps it indicates that the September poll had flattered (in YouGov terms) the Yes position due to margin-of-error issues, which would mean that the No lead has in reality dropped by quite a bit more than 2% since the early autumn.  That theory would obviously be more consistent with the trend found by ICM and TNS-BMRB.

This poll is (at least) the third YouGov poll in a row to show that 18-24 year olds are the most pro-independence age group, so that's yet another nail in the coffin for one of the mainstream media's most beloved narratives!  In fact, the Yes campaign now hold an outright lead among 18-24 year olds, albeit that finding is based on a significantly smaller sample size than the January poll.  YouGov were probably stuck with a more limited number of young respondents in this poll due to time factors - the fieldwork lasted for less than half as long on this occasion.

Incredibly, the poll gives identical voting intentions for higher-income and lower-income voters - in both cases Yes 34%, No 52%.  I find that highly suspicious, because it contradicts the clear evidence from other pollsters that lower-income respondents are significantly more likely to break for Yes.  As I suggested the other day, if YouGov are failing to find a representative sample of lower-income voters, that could well be a potential explanation (admittedly only one of several) for why they continue to paint a less optimistic picture for the pro-independence campaign than the majority of BPC pollsters.

* * *


After the little blip that was artificially generated by the intervention of Survation, I'm delighted to say that normal service has been resumed with yet another swing towards independence in the Poll of Polls.  Of the eight updates since we got underway at the time of the publication of the White Paper, no fewer than six have shown a swing to Yes, with just one showing a swing to No, and one showing an unchanged position.

MEAN AVERAGE (not excluding Don't Knows) :

Yes 34.0% (+0.1)
No 48.7% (n/c)

MEAN AVERAGE (excluding Don't Knows) :

Yes 41.1% (+0.1)
No 58.9% (-0.1)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 40.8% (n/c)
No 59.2% (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, and that adhere to British Polling Council rules. At present, there are seven - Angus Reid, 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.)

At exactly 34%, the Yes campaign have now reached their highest ever level of support on the headline average.  In case you're wondering, the reason why the median average is unchanged is that YouGov remain on the No-friendly end of the spectrum (along with Survation and Ipsos-Mori), and therefore the change in this poll doesn't affect the calculation.  The Yes-friendly end of the spectrum is currently comprised of ICM, Angus Reid and Panelbase, with TNS-BMRB providing the mid-point.

* * *

A small housekeeping announcement - my desktop computer packed up on Tuesday, so blogging is a bit of a struggle at the moment.  I made a pig's ear of linking to this post on Twitter earlier, so apologies for that!


  1. I regularly do the You Gov polls on line they always ask if there was an election how would you vote and then they go on to ask about whether you think the Westminster government is doing a good job or not then on about the leaders of the 3 main parties they never take into account the effects that an snp surge would have especially on the Labour party. Their sample size for Scotland must be small and to them what ever happens to the SNP vote wouldn't make a difference when we know that the loss of say 20 Scottish seats by Labour could make all the difference. You Gov pols should be taken with a pinch of salt as they always underestimate both the SNP and the Yes vote. I think that monitoring internet chatter is a better way of assessing things as the internet noise was way ahead of the polls in 2011 and I see the same this time.

  2. James, thanks again for keeping us poll watchers up to date with the news. As always a fascinating analysis. And v encouraging to see the swing to Yes reflected in the new poll of polls.

  3. When even YouGov show the gap narrowing that's good news.

    Comparing polls of the referendum year with polls from the year to go mark last September:
    (change in percentage points)

    ICM Yes: 37%(+5) No: 44%(-5)
    TNS Yes: 29%(+4) No: 42%(-2)
    YouGov Yes: 34%(+2) No: 52%(NC)

  4. We shall see what poll numbers we get in the next couple of months. Something must be happening when you hear of turnouts at public meetings. 400 at Motherwell - we had over 400 in Dundee a little while ago. Allan Little's critique of BT on yesterday's Newsnight was interesting.

  5. Motherwell might be more fertile territory than you'd think. Remember Ravenscraig. People may think they couldn't do any worse with a complete change of regime.