When I first received the tables from the new Scot Goes Pop / Panelbase poll, I was tearing my hair out slightly, because the numbers were pretty good for Yes in many ways, but just two respondents swung the balance, ie. if just two people who said they were No voters had been in the Yes column instead, the headline numbers would have been rounded to 50-50. You’ll get the best sense of how close it is by looking at the numbers rounded to one decimal place, which make it Yes 49.3%, No 50.7%. With Don’t Knows left in, even the rounded numbers show only a statistically insignificant one-point lead for No.
All of which tells a far better story for Yes than today’s Savanta ComRes poll, which is left looking like a possible outlier. Bear in mind that the Panelbase and ComRes fieldwork dates overlap to a significant extent – the Scot Goes Pop / Panelbase poll was conducted between Wednesday of last week and Monday of this week, while the ComRes poll was in the field between Friday of last week and Tuesday of this week. It would be a bit of a stretch to claim that the ComRes poll is more up to date and picked up a swing that Panelbase couldn’t. Remember also that recent Survation and Ashcroft polls are bang in line with Panelbase in showing Yes on 49%, and while the last YouGov poll put the figure at 47%, that was only a 2% decrease from a firm that is typically on the No-friendly end of the spectrum. So that’s all fairly consistent with the suspicion that the new ComRes poll may be underestimating Yes due to random sampling variation. For now at least, it looks like public opinion on independence continues to be essentially split down the middle, which is an excellent starting point for a referendum campaign.
Here are the headline numbers from the Scot Goes Pop / Panelbase poll…
Should Scotland be an independent country? (Scot Goes Pop / Panelbase poll, 21st-26th April 2021)
Yes 49%
No 51%
Before Don’t Knows are removed, the numbers are –
Yes 47%
No 48%
Undecided 6%
(1075 respondents were interviewed. The percentages add up to more than 100 due to rounding.)
As ever, the figures for Labour voters are of particular interest – 28% are pro-independence and 72% are anti-independence, which is not quite as good for Yes as some previous Panelbase polls. That partly explains why Yes don’t have an outright lead.
* * *
There’s lots more to come from the poll – several supplementary questions of interest to the independence movement, and also Westminster voting intentions. If you’d like to be the first to know the results, feel free to follow me on Twitter HERE.
