Exploding the myth that Alex Salmond "doubled independence support when he was First Minister"

One of the most obvious signs of the contempt that Stuart "Stew" Campbell has for his own readers was the way he kept using his fraudulent "independence support is flatlining" graph for years on end, even after it had been debunked umpteen times.  Basically he had cherry-picked a tiny number of polls from the hundreds of independence polls that had been conducted since 2014 to make it look as if support for Yes had remained absolutely static at 47%, thus completely misrepresenting the fact that an average of all polls showed a steady year-on-year increase for Yes until 2020, and that even in 2021 and 2022 support remained higher than in any year up to and including 2019.  

At some point, however, he finally got bored with that graph and replaced it with a new one, which has now had several outings.  I can't even call the new one misleading - it's a downright lie in most respects.  It claims the following -

* That support for independence stood at around 25% at roughly the time Alex Salmond became First Minister in May 2007.  (No exact number is given, but it appears to be halfway between 20% and 30%.)

* That support had doubled to 50% by the time Salmond was replaced by Nicola Sturgeon in November 2014.

* That the Yes vote had dipped slightly to the high 40s when Sturgeon was replaced by Humza Yousaf in May 2023.

* That the Yes vote remained unchanged in the high 40s when Yousaf was replaced by John Swinney in May 2024.

* That the Yes vote remained unchanged in the high 40s in May of this year.

The idea, of course, is supposed to be that Alex Salmond dramatically increased support for independence but that all of his successors have failed to build on that golden legacy.  And it's not hard to see why Wings readers find that narrative so seductive, but there's just one little snag - there's not actually a shred of truth in it.

The claim about May 2025 is the easiest to deal with because it's so recent.  There were exactly two independence polls in that month: one from Survation that had Yes on 49% and one from Norstat showing Yes on 54%.  So what has Stew done to produce his high 40s figure for the month?  He certainly hasn't used an average of the two polls, because that would have got him to around 51% or 52%.  So has he just used one and ignored the other?  If so, what possible justification does he have for doing that?  Before anyone suggests that maybe he's been sticking to Survation polls throughout the graph for the sake of consistency - nope, Survation didn't even exist in 2007.

And it's that 2007 figure which is by far the most problematical.  Unsurprisingly Stew doesn't give any source for it at all, but unlike the figures for the other years it clearly doesn't come from a straight Yes/No poll on independence with Don't Knows excluded, because that would imply the numbers were around Yes 25%, No 75%.  No poll even remotely like that has been published at any point in the 21st century.  By far the most likely explanation is that he is instead using the Scottish Social Attitudes Survey for that year, which had a complex multi-option format and did not exclude Don't Knows.  To say that he's made an apples-and-oranges comparison does not adequately convey the absurdity and fraudulence of what he's done - it's more like an apples-with-tractors comparison.

So what were the directly comparable Yes/No polls on independence showing in 2007?  There weren't very many independence polls being conducted back then, and most that did take place were conducted by TNS / System Three (last heard of under the branding Kantar).  It looks like there were two polls from the firm in 2007 - one showed Yes and No level on 50% apiece, and the other had No ahead by around 57% to 43% if Don't Knows were excluded.  So if we're ultra-generous and use the more favourable poll for No as the baseline, Alex Salmond increased support for independence by around seven percentage points during his tenure as First Minister - light-years short of the 25-point increase implied by Stew.  If we're not generous, and if we use the more favourable poll for Yes as the baseline, Salmond as FM did not increase support for independence at all.

Some people may be genuinely astonished to learn of this, because the mythology of Salmond doubling independence support has been so deeply ingrained into them.  But that's mainly because Salmond himself was such an effective propagandist, and it wasn't in his own interests to draw attention to the existence of several polls showing an outright Yes lead before he even became First Minister.  During the period of Labour-Liberal Democrat coalition rule from 1999 to 2007, it was in fact reasonably common for TNS / System Three to show a majority for independence.

What actually happened was that Yes support remained high until it became clear that the referendum was really going to happen, and then it was as if reality hit home for a lot of people and the Yes vote dropped like a stone, falling to as low as 33% at one point in 2012-13 (although never going close to Stew's fictional 25% mark).  Over the course of the referendum campaign there was an impressive recovery, but that essentially just got us back to where we started.  The best that can be said is that the 50% Yes vote in late 2014 had a lot more depth and substance to it than the 50% Yes vote of 2007, because people had properly thought about the issues by then.

*  *  *

The running total in the Scot Goes Pop 2025 fundraiser currently stands at £3100, meaning it is 46% of the way towards the target figure of £6800.  If you'd like to help the blog keep going, donations by card are welcome HERE, or alternatively you can cut out fees altogether (depending on which option you select from the menu) by making a direct donation via PayPal.  My PayPal email address is:  jkellysta@yahoo.co.uk