Sunday, July 6, 2014

New TNS-BMRB poll shows 2% increase in support for independence

Well, the plot thickens.  As we were waiting for a poll in the Sunday Post that No campaign staffers seemed to be crowing about on Twitter, a post quietly appeared on John Curtice's blog revealing figures from a new TNS-BMRB poll commissioned by Tom Hunter, showing a marked increase in support for both Yes and No.

Should Scotland be an independent country?

Yes 32% (+2)
No 46% (+4)

Don't be fooled by the fact that support for No appears to have increased more than support for Yes.  As the numbers of Don't Knows decrease, the side in the lead actually needs a bigger increase just to stand still on the underlying figures.  And stand still is exactly what the No campaign have done.  Here are the figures with Don't Knows stripped out...

Yes 41% (n/c)
No 59% (n/c)

There have now been several TNS-BMRB polls in succession showing exactly the same 41/59 split.  That's a double-edged sword - although it fails to provide any corroboration for the Panelbase and Survation polls last month that saw the No lead slumping to an all-time low, it also fails to back up the contrary suggestions in the two most recent YouGov polls (both conducted later than the Panelbase and Survation polls) that Yes have gone into reverse.  According to Professor Curtice, the fieldwork for this poll was conducted over two weeks in mid-June, placing it earlier than the YouGov poll that was published on Monday evening, but around the same time as the previous YouGov poll.  If we're being invited by the No campaign to take the two YouGov polls as a package, which between them point to a new pro-No swing, then the fieldwork for TNS should have come late enough to detect the same trend - but fortunately it's detected no such thing.  That increases the likelihood that what we've been seeing from YouGov has simply been meaningless margin-of-error 'noise'.

However, the jury must remain out on that question until we see what the Sunday Post have for us in the morning.  I suppose there's an outside chance that Hunter may have given the Post permission to use his figures, but there's certainly no mention of any such arrangement in Curtice's post.  So my best guess is that there must be another poll, possibly from Survation. (UPDATE : I'm not quite so sure now.  McDougall has just excitedly tweeted the TNS numbers, which might conceivably indicate that's the poll he was so pumped up about earlier.  Is a poll showing a 2% increase in support for independence really the best they've got? Time will tell.)

There's also no mention in Curtice's post of what the TNS numbers are after turnout filters have been applied - in the last couple of polls those figures were markedly better for Yes than the headline numbers, and were more in line with what we're used to seeing from Yes-friendly pollsters such as Panelbase.  We may have to wait for the publication of the datasets to discover whether that is still the case.

*  *  *

SCOT GOES POP POLL OF POLLS

Although the No lead has crept up very fractionally in this update of the Poll of Polls, I suspect that may well be an illusion caused by the effect of rounding.  We'll have to wait for the datasets to find out, but a rough calculation suggests it's perfectly possible that the unrounded Yes vote in the TNS poll is identical to last time around after Don't Knows are excluded, or has even increased very slightly.

The No lead in this update remains lower than at any time prior to last month.

MEAN AVERAGE (excluding Don't Knows) :

Yes 43.5% (-0.1)
No 56.5% (+0.1)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 36.8% (+0.3)
No 47.8% (+0.6)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 43.3% (-0.3)
No 56.7% (+0.3)

(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.)

Here are the long-term trend figures, with updates prior to Easter recalculated to remove the inactive pollster Angus Reid ...

The No campaign's lead in the Poll of Polls mean average (not excluding Don't Knows) :

Sep 2013 - 21.6%
Sep 2013 - 21.4%
Sep 2013 - 19.4%
Oct 2013 - 18.8%
Oct 2013 - 18.4%
Oct 2013 - 18.2%
Nov 2013 - 18.4%
Nov 2013 - 18.0%
Dec 2013 - 17.0%
Dec 2013 - 16.8%
Dec 2013 - 16.4%
Jan 2014 - 14.4%
Jan 2014 - 14.2%
Jan 2014 - 14.2%
Jan 2014 - 15.2%
Feb 2014 - 15.0%
Feb 2014 - 15.5%
Feb 2014 - 15.5%
Feb 2014 - 13.7%
Feb 2014 - 13.3%
Feb 2014 - 14.2%
Mar 2014 - 14.2%
Mar 2014 - 14.5%
Mar 2014 - 14.5%
Mar 2014 - 14.7%
Mar 2014 - 13.8%
Mar 2014 - 13.0%
Mar 2014 - 12.5%
Apr 2014 - 12.5%
Apr 2014 - 12.7%
Apr 2014 - 12.7%
Apr 2014 - 12.3%
Apr 2014 - 11.4%
May 2014 - 11.2%
May 2014 - 11.2%
May 2014 - 11.5%
May 2014 - 13.3%
Jun 2014 - 12.1%
Jun 2014 - 12.1%
Jun 2014 - 11.3%
Jun 2014 - 9.9%
Jun 2014 - 10.3%
Jun 2014 - 10.7%
Jul 2014 - 11.0%

11 comments:

  1. Fieldwork in mid-june?

    Why are TNS taking so long to release?

    Imagine if that had been done ahead of 2011. We'd have 44 Lab / 32 SNP at the same point in time.

    The gap only closed to parity by t-36.

    ReplyDelete
  2. Current time = t-73
    Current TNS fieldwork = t-87+

    TNS-BMRB face to face
    2011 t-64

    Lab 44%
    SNP 29%
    Con 12%
    Lib 11%

    So even 64 days out TNS still had the reverse 2011 result in terms of SNP and Lab. This new Y/N poll is 84+ days out and we're 73 days out right now.

    Of course there's no way we can say the swing will happen again, but the SNP didn't take a tiny lead until t-38 in 2011 with TNS the most off at this point.

    Final polls had a Const + Regional average gap of 10 points for SNP over labour. The actual gap was 17 points.

    ReplyDelete
  3. Hi, I haven't paid much attention to this before, why do we want independance? Will we be better off seperate?

    ReplyDelete
  4. Maggie

    Yes we will.

    Why not try the Yes Scotland site and look up all the info there.

    ReplyDelete
  5. On a different topic, if the trend I see in my own UK VI poll monitoring continues, then the Labour lead over the Tories will fall to just 2 points at most this month, greatly increasing the probability of a Tory majority or as largest party again in 2015.

    This has been the steady pattern since early 2013 when Labour were a comfortable 11 points ahead.

    ReplyDelete
  6. Hi, I have a question if any one can enlighten?
    Do these poll companies often poll the same folk over and over? I've heard of one case where someone was polled by the same company 3 times since turn of year.
    And if this is common practice, what effect does this have on the validity if the poll?
    Thanks in advance.

    ReplyDelete
  7. Not sure about TNS but I have been polled 4 times by YouGov this year

    ReplyDelete
  8. Four of the six pollsters use volunteer online panels, so they end up interviewing the same people over and over again. The exceptions are Ipsos-Mori, who use telephone fieldwork, and TNS-BMRB, who use face-to-face fieldwork. So those two firms should have a completely new sample every time, barring coincidences.

    ReplyDelete
  9. Thanks James. I'm sure the guy I'm talking about was polled several times by Ipsos Mori over the phone. I'll ask him and get back to you.

    ReplyDelete
  10. I have never been polled and I don't know anyone who has been polled. That doesn't mean anything but who are they asking?

    ReplyDelete
  11. Karen: As mentioned above, four of the six pollsters use volunteer online panels, so if you want to take part in polls all you have to join a panel (probably YouGov, because Survation aren't accepting new members, and Panelbase don't use new members for referendum polls). Your statistical chances of being randomly contacted by Ipsos-Mori or TNS are pretty slim.

    ReplyDelete