Help Talking-up Scotland keep going

Crowdfunder at: https://www.crowdfunder.co.uk/p/help-talking-up-scotland-keep-going

Or by direct bank transfer to: JOHN WATSON ROBERTSON Sort Code – 08-91-04 Account – 12266421

There’s a pattern there, above. YouGov don’t seem to find any Yes leads and a massive 10% No lead, in the most recent polling. I’ll come back to Ipsos. Take these out and we’ve have some of the kind of settled will evidence the new PM seems to need.

It’s not just the last three YouGov polls, going back years they almost never get a Yes lead. YouGov is of course, Tory-owned. So how might they be doing this. Have a look at the way they have adjusted their initial sample to make a No result more likely.

They have reduced the size of the declared SNP voters from 659 to 601, a greater reduction than for the Unionists in the initial unweighted sample and increased the Reform UK sample.

Despite the collapse of Labour in the 2026 Holyrood election, they have ignored that and reduced the SNP sample dramatically on the basis of the 2024 GE.

On declared current support, they have reduced the NO and increased the Yes sample, but by only 7 in the latter.

Finally, bizarrely still using the 2014 result and most obviously biasing the result, they have kept a No lead over Yes by 934 to 901 despite all of the other polls, which they are aware of, giving a Yes lead.

Help Talking-up Scotland keep going

Crowdfunder at: https://www.crowdfunder.co.uk/p/help-talking-up-scotland-keep-going

Or by direct bank transfer to: JOHN WATSON ROBERTSON Sort Code – 08-91-04 Account – 12266421

It was essentially the same in the first part of the year up to March.

A headline in this morning’s Times is inaccurate and they have issued this correction: “YouGov is not making changes to its methodology or to any previously published results. We are happy to make this clear.”

YouGov is not changing our methodology or results, which accurately…

— YouGov (@YouGov) March 16, 2026

YouGov responds above, after these claims in the Times:

YouGov to publish more data after Nigel Farage row – The Reform UK leader had accused the pollster of being ‘deceptive’ and underplaying the popularity of his party in its voting intention surveys

I couldn’t help but smile after our years of noting YouGov‘s complete inability to ever find a Yes lead and to always find a smaller SNP lead, than in all the polls around them. See:

I’ve written before about YouGov‘s unerring and almost unique (Survation too) ability to pluck a No lead out of a trend dominated by Yes leads, found by various other pollsters. It’s been a long time since they found a Yes lead. They seem to prevent it by adjusting (weighting) their sample to have 55% No voters and 45% Yes voters as in 2014 and thus get the same result ten years later.

Help Talking-up Scotland keep going

Crowdfunder at: https://www.crowdfunder.co.uk/p/help-talking-up-scotland-keep-going

Or by direct bank transfer to: JOHN WATSON ROBERTSON Sort Code – 08-91-04 Account – 12266421

James Kelly of Scot Goes Pop has already reported on this at: https://scotgoespop.blogspot.com/2026/02/why-is-yougov-under-representing-yes.html but I’d like to see the numbers.

You see above, instantly, the anomaly of the small Glasgow sample? They have actually been doing it for at least a year in all their polls and gradually reducing the Glasgow sample. In March 2025, it was 116. then 104, then 102, then 109 and most recently, 90! Notice also that only the Glasgow sample suggests a Yes lead even after they have weighted it according to the 2014 result.

I’m really annoyed to have not noticed this before.

Here are the populations of 8 electoral regions in Scotland from which the list MSPs are selected, the sample for each used by YouGov and the consequent ratios:

These are shocking differences between the Glasgow and other regions and inexplicable other than in an intentional attempt to distort the actual level of support for independence.

In, for example, in an area where they find a No lead comparable to that in 2014, such as the North-east, they sample around 1 in 4 000 but in an area where the Yes lead is large, Glasgow, they only sample 1 in nearly 8 000. To be representative the Glasgow sample should be twice as big but that would shift their overall No lead and they don’t want us thinking that do they?

The last FIVE YouGov polls on Scottish independence uniquely, due to skewing their sample to match the 2014 result, find No leads among a sequence of Yes leads by other pollsters. They presumably do something similar for the Holyrood polls.

The small letters in the right-hand column refer to irregular practice such as not including 16/17 year-olds.

Help Talking-up Scotland keep going

Crowdfunder at: https://www.crowdfunder.co.uk/p/help-talking-up-scotland-keep-going

Or by direct bank transfer to: JOHN WATSON ROBERTSON Sort Code – 08-91-04 Account – 12266421