With methodology and wording from Alba? Odd that the best polls for Alba are the ones commissioned by ScotGoesPop/James Kelly and the Alba Party themselves. 🤔
— indy swim 🏴 (@indy_swim) October 20, 2022
Jesus, man, you’re repeating yourself from yesterday? I’ve already pointed out to you that the client makes no difference to the result. What precisely are you accusing me of doing to rig these polls? Drop the innuendo – let’s have specifics.
— James Kelly (@JamesKelly) October 20, 2022
“He didn’t answer” as Nick Robinson would say, except in this case he really didn’t answer. And it’s no wonder, because the guy hasn’t got a leg to stand on. I really do take exception to this guff, especially as it’s been trotted out twice in the space of 48 hours by someone who isn’t even attempting to substantiate his innuendo. His profile bio gives the game away: “Unintentionally annoying British nationalists since 2019 & Alba supporters since 2021.” It would appear that he gives parity of esteem to his loathing of unionists and his loathing of fellow independence supporters.
Here are the facts about the opinion polls I’ve commissioned. There have been six, but four of those were conducted before Alba even existed (meaning that I was still a member of the SNP at the time). Of the remaining two, only one actually produced favourable results for Alba. That was a Panelbase poll conducted during the 2021 Holyrood election campaign, between the 21st and 26th of April. It showed Alba on 6% – but that was identical to the Alba figure from the previous Panelbase poll, commissioned by Believe in Scotland, and conducted around two weeks earlier. Believe in Scotland, I think it’s fair to say, are broadly supportive of the SNP and the Greens, and have little or no sympathy for Alba. But it doesn’t stop there, because the Panelbase poll before the one for Believe in Scotland also showed Alba on 6% – and that was commissioned by a unionist client, the Sunday Times.
None of this should be a surprise, because as far as voting intention questions are concerned, pollsters do not change their question wording or methodology from one poll to the next to suit a particular client. It’s true that clients have lots of influence on the supplementary questions that are asked in any poll, but that has literally zero effect on the voting intention results because supplementary questions are generally asked later in the question sequence. So no matter whether their client is an Alba member like myself, an SNP-supporting group such as Believe in Scotland, or an anti-independence newspaper like the Sunday Times, Panelbase ask exactly the same voting intention questions and use exactly the same methodology. Little wonder, then, that the results produced in 2021 campaign were so stable.
Of course Panelbase were consistently producing better results for Alba than any other pollster in 2021, so the one way in which an Alba-supporting client could hope to indirectly influence the result of a poll would be by deliberately selecting the firm that they thought would produce the best numbers. Is that what I did in April 2021? Simple answer: no.
When I started commissioning polls in 2020, I didn’t contact Panelbase to begin with, because they’ve conducted many polls for Wings Over Scotland and I thought a touch of pluralism might be no bad thing. But after contacting five or six other firms, I felt like I was hitting a brick wall – one or two didn’t reply at all, one wasn’t able to help within the desired timescale, one suddenly stopped replying midway through the process, one would have charged an astronomical amount, etc, etc. So I then turned to Panelbase and the whole thing was sorted out within 24 hours. That, I think, is the real reason why so many pro-indy clients have generally commissioned Panelbase rather than other firms. Panelbase have experience of dealing with alternative media clients, and it just seems to be a much smoother process. And of course once you’ve had a good experience with a firm, you tend to stick with them.
I did commission a Survation poll in January 2021, but that was because Panelbase weren’t able to help that month for contractual reasons. However, I mentioned to Panelbase at that point that I still hoped to go back to them for a poll during the Holyrood campaign, and they said with a bit of notice they could make sure they fitted it in. So the whole thing was agreed in principle months before I even knew that there was going to be any such thing as an Alba Party. In fact, I think by that point I was starting to give up hope of an Alex Salmond-led list-only party, and I was more or less expecting to be “both votes SNP” myself.
As far as my most recent poll is concerned, that showed Alba on 2% of the first preference vote for the local elections (which, as it turned out, was bang on the money in the wards where Alba stood). No Holyrood question was asked. There was a Westminster question, but Alba weren’t even included as an option. I had no influence over that decision, and frankly I wasn’t remotely bothered by it, because as has been well-rehearsed over recent days, I think Alba would be extraordinarily foolish to split the pro-indy vote by standing in a Westminster election conducted by first-past-the-post, except possibly in the two constituencies where they have the incumbent MPs.
So, in a nutshell, Indy Swim, I don’t know what the hell you think you’re talking about, and to be honest I don’t think you do either.
