Basically SiU commissioned an independence poll from Survation, but insisted that a non-standard question be used in the full knowledge that it was likely to give the misleading impression of a more favourable position for No than is usually the case. We don't need to cast our minds back very far to recall just how susceptible poll respondents are to changes in question wording - there have been polls in the last few months that misleadingly showed a "Yes vote" in the low 50s, but that was only because respondents were invited to assume that Brexit would go ahead. OK, it's highly likely that Brexit will indeed go ahead, but if respondents are asked to make any sort of assumption, it's likely that at least some of them will subconsciously believe that it is "supposed" to change their view.
If one sort of leading question can produce an illusory gain for Yes, it's not at all surprising that a different sort of leading question can produce the opposite, and that's exactly what has happened in the SiU poll. We know that in standard Survation online polling, the Yes vote stands in the mid-to-high 40s, but by instead asking the non-standard question "Should Scotland remain in the United Kingdom or leave the United Kingdom?", SiU were able to produce the false impression that Yes support has slumped to 40%. In a disgraceful betrayal of basic journalistic ethics, The Scotsman has played along with this propaganda stunt, and has cynically left its readers with the sense that the poll figures represent a real change in public opinion.
Why would respondents react so differently to a "remaining in/leaving the UK" question? We don't know for sure, but what we do know is that this is nothing new - there were polls in the run up to the first indyref that showed exactly the same effect. One possibility is that a minority of respondents may be unsure of what "the United Kingdom" actually is, and may wrongly assume that they are being asked about their views on Scotland retaining the monarchy.
Amusingly, though, the poll has backfired on SiU in a couple of ways. Firstly, it shows that, if Don't Knows are removed, a clear majority of respondents want there to be a second independence referendum before May 2026, ie. within the next seven-and-a-half years. It's hard to overstate just what a staggering finding that is for a poll commissioned by Scotland in Union - it essentially proves that one of their central claims about public opinion is the polar opposite of the truth. Here are the figures with Don't Knows removed -
When should another referendum on Scotland leaving the UK be held if at all?
Total for before May 2026: 54.0%
Total for after May 2026 or never: 46.0%
Incidentally, even with the leading wording about "leaving the UK", a very substantial minority of 46.7% want a second referendum before May 2021, ie. within just two-and-a-half years.
The other amusing detail is that there is a statistical tie on the question of whether Scotland should "leave" the UK in the event of a no-deal Brexit, ie. the result is within the standard margin of error of the poll, meaning that it's impossible to know which side is really in the lead.
If the UK leaves the EU WITHOUT a deal in place, and there is a subsequent referendum on Scotland leaving the UK, how would you vote?
For Scotland to remain as part of the UK: 52.3%
For Scotland to leave the UK: 47.7%
There are also Westminster voting intention figures, which continue to show a handsome double-digit lead for the SNP over both the Tories and Labour.
Scottish voting intentions for next UK general election:
SNP 39% (-1)
Conservatives 26% (-1)
Labour 24% (+1)
Liberal Democrats 8% (+1)
The percentage changes listed above are measured from the Scottish portion of the recent GB-wide Survation mega-poll for Channel 4. If the last full-scale Scottish poll from Survation for the Record is used as the baseline instead, the SNP are actually up 3 points. Either way, though, these new figures put the SNP on course for substantial seat gains - mainly from Labour, who would be almost wiped out once again.
* * *