Never underestimate the importance of the battle of the narratives after a landmark event like the Supreme Court ruling. The dearth of independence polls in the immediate aftermath gave the unionist parties the opportunity to fill up the space with – frankly – lies about what they were finding on the doorsteps, with voters supposedly reacting like obedient slaves to the discovery that their country does not have the legal ability to decide its own future, and saying that their government should just accept that Scotland is in a prison and get on with serving the sentence. Even the appearance of the Redfield & Wilton poll showing outright majority support for independence didn’t thwart the unionist propaganda plan too much, because Labour in particular just studiously ignored the independence numbers and concentrated instead on the finding that the SNP’s lead in Westminster voting intentions had dropped sharply – which they were to some extent able to get away with because, after all, it’s a Westminster election that the SNP are now planning to use as a de facto independence referendum.
Today’s new Ipsos UK poll, however, puts unionists in danger of losing the battle of the narratives – and it’s up to all of us to make sure the numbers become as widely known as possible. (The fact that STV were Ipsos UK’s client should help considerably, although there’s still the challenge of making sure the London-based media don’t ignore the poll.)
Should Scotland be an independent country? (Ipsos UK / STV, 28th November – 5th December 2022)
Yes 56% (+6)
No 44% (-6)
Scottish voting intentions for the next UK general election:
SNP 51% (+7)
Labour 25% (+2)
Conservatives 13% (-6)
Liberal Democrats 6% (-4)
Greens 3% (-)
Seats projection: SNP 58 (+10), Labour 1 (-), Conservatives 0 (-6), Liberal Democrats 0 (-4)
This poll differs from the Redfield & Wilton poll in quite a number of respects. Although both show a pro-independence majority, Ipsos UK shows the Yes vote increasing by a larger amount (six points rather than four), and also shows a Yes lead that isn’t within the standard margin of error. In other words, the Ipsos UK poll unequivocally shows a pro-independence majority, whereas Redfield & Wilton could technically be said to have shown a ‘statistical tie’, to use the American term.
But of course by far the biggest differences are to be found in the Westminster numbers, with the trends reported by the two polls not being even remotely reconcilable with each other. Redfield & Wilton showed the SNP down four points and Labour up twelve, working out at a very substantial sixteen point drop in the SNP’s lead. Although Ipsos UK also show a boost for Labour, it’s a statistically insignificant two points, while the SNP are up seven points and have thus considerably extended their lead. This also moves the SNP above 50%, which is no longer just a psychological barrier – it’s their self-defined target for victory in a plebiscite election. Indeed, in combination with the Greens, the Westminster vote for pro-indy parties stands at 54% – just a touch below the Yes vote on the standard indyref question. That’s another big difference with Redfield & Wilton, who suggested that the combined vote for pro-indy parties was a full nine percentage points lower than the Yes vote.
It obviously matters tremendously which pollster is getting it right and which is getting it wrong, because if Redfield & Wilton are right, a Westminster election used as a plebiscite may not be winnable for the pro-indy camp due to the Labour surge, whereas if Ipsos UK are right, the SNP are shrugging off the Labour surge down south and tightening their grip on Scottish politics. If we (or rather the SNP leadership) read this situation incorrectly due to faith in an incorrect poll, it could have catastrophic consequences if the wrong strategic call is made as a result – ie. sticking with the Westminster plebiscite plan rather than using an early Holyrood election in 2023 instead.
My suspicion is that a gulf is opening up between Ipsos telephone polling and online polling from the other firms. There won’t necessarily be such a big gulf on indyref voting intentions, but on Westminster numbers my guess is that the other online pollsters will be closer to Redfield & Wilton – I say that in part due to the straws in the wind we’re seeing from subsamples. So if the SNP put their faith in the Ipsos numbers, that may mean putting all their eggs in one basket, because it could mean assuming that the other pollsters are all wrong. That would be a big call. Remember that Ipsos have in recent years tended to be on the Yes-friendly end of the spectrum – a complete reversal from the 2014 campaign, where they were just about the most No-friendly firm.
In a nutshell, I would still strongly argue that the most promising strategy for a de facto referendum is to engineer an early Scottish Parliament election next year, possibly in the autumn at around the same time the referendum had originally been planned for.
Related
