The BBC's Norman Smith, he of the cringe-inducing "Sassenachs" monologue, continues in his fearless one-man quest to bring the standard of London media reporting of the independence referendum down to an even lower level. He tweeted this about an hour ago -
"Poll in today's Aberdeen Press and Journal gives huge lead for No campaign (65%) versus just 17% for independence"
Of course those numbers are mind-bogglingly implausible in the wake of four Scotland-wide polls by established pollsters in recent days showing very modest No leads of between 9% and 13% (and one of which put the No vote as low as 42%). So to say I smelled a rat would be to put it mildly, but to begin with I couldn't find any trace of the poll online. All I could do was post this reply -
"Norman, come on man : basic journalism. Who was it a poll of? Scotland-wide? Obviously not. Don't treat us like idiots."
Sure enough, it turns out to be a poll of "the north and north-east" only. But the numbers still aren't remotely credible even on that basis - there's no way a pollster like ICM would be finding Yes on 37% across Scotland if the figure in the north and north-east was even vaguely close to 17%. The Press & Journal give very little information about the poll's methodology on their website, but are billing it as the second poll of its type, following on from an earlier one in the spring of last year. That in itself should be enough to set alarm bells ringing about its credibility, because Professor Curtice and other analysts were unequivocal in treating the January 2014 poll conducted by ComRes in the Borders and Dumfries & Galloway as the first regional poll of this campaign, which implies that any poll in the north last spring was not regarded as serious. The P&J insist that their new poll is "an independent snapshot of public opinion", but in the absence of any hard information my best guess is that it's one of those "half-breed" polls - not quite a self-selecting voodoo poll, but not scientifically rigorous either. A rough equivalent would be the Scotpulse poll the other day that showed a dead heat between Yes and No across the country. You have to ask yourself this - why is Norman Smith trying to hoodwink people into thinking the P&J poll is something it's not, when it would presumably never even occur to him to give the Scotpulse poll the time of day?