Monday, February 6, 2012

Herald loses the plot in its reporting of TNS-BMRB poll

First things first, here are the results of the latest TNS-BMRB poll on independence -

Do you agree that the Scottish Government should negotiate with the UK so that Scotland becomes an independent state?

Yes 35% (-4)
No 44% (+6)

Now you'll notice straight away that this poll isn't directly comparable with many recent polls conducted by other companies, because it doesn't use the actual referendum question that has been proposed. Perfectly reasonably, the Herald notes that it wasn't possible to use the proposed question because it hadn't been published at the time fieldwork started, but instead of conceding that this renders the poll somewhat inferior to others, they risibly hint that asking a question that won't actually be asked in the referendum makes their poll superior, and even more incredibly that the results of the poll somehow cast doubt on the legitimacy of the real question! As a final flourish to this rather fantastical line of reasoning, they loftily note that -

"The question has already been criticised by opponents and by many polling experts as both leading and loaded."

Really? Was that before or after they welcomed it as fair, simple and clear? Here is the Scotsman's useful summary of the reaction to the question the day after it was published -

"Mr Salmond’s single question on independence was supported by constitutional experts last night. The UK government also welcomed the clarity of the question he proposes."

I think what irritates me most about this little turn of sophistry from the Herald is that they use it as thin cover to do what they've done many times before without any excuses - take a single (often rather questionable) poll they've commissioned themselves and report it as if it exists in a total vacuum. Remember those absurd MRUK polls during the 2007 election campaign which allowed the Herald to earnestly report on "Labour sailing towards a third term", as if the solid SNP leads being reported by the more credible polling companies were just a figment of our collective imagination?

And so we're breathlessly informed today that this poll is cast-iron proof that the pro-independence side has "lost" the first skirmish of the referendum battle - in spite of the fact that pretty much every other recent poll, regardless of whether it used the SNP's preferred question or not, has shown an increase in support for independence.

A note on TNS itself. If you look at the percentage change figures for this poll, you'll notice something rather important - that independence was actually in the lead in the last poll. And yet we've been informed by the London media and British nationalist politicians alike that "all polls" have shown a clear rejection of independence. In other words, TNS have so little credibility in their eyes that the last poll simply didn't register - it barely even existed. So I wonder if we'll see a consistency of approach this time - or will TNS miraculously be rehabilitated as a credible pollster now that their figures are suddenly 'on-message'?

The Brit Nats had better be careful here. A nine-point lead for No is pretty slender compared to the leads in the Cochrane-esque worded polls about "separation" that they're more used to crowing about.

1 comment:

  1. It's not a very good question... and as you point out it is the wrong question.

    I read somewhere that everybody welcomed the First Minster's question, until someone from a Business Studies department of a university somewhere in Arizona popped up with his theory of bias.

    Then, of course, all the noodle heads here jumped on the bandwagon. Suddenly it was unfair, biased, designed to sway opinion, terrible, offensive...

    This is starting to get real old, and I can only hope that we're not going to have to suffer another 2 1/2 years of moronic commentary from the unionist press.