Scotsman/Savanta polls: NO if you want it?

When I see the words ‘Scotsman’, ‘Savanta’ and ‘poll’ together, I think ‘NO.’

Today, the Scotsman has the first poll in seven or more with a NO lead. It’s the first from them since the end of September, when they had a 1% lead for NO among a series of such results. They haven’t had a YES lead in 2022.

In January 2022, we could write:

Looking at the polls going back to the beginning of 2021, Savanta Com Res often with the Scotsman stand out.

They report 1 Yes win, 9 No wins and 2 ties after the Ipsos MORI 9% win for Yes in November.

No other pollster is so good at finding NO wins.

In the same period, Panelbase had 4 Yes and 6 NO. Ipsos MORI had 3 Yes, 0 No and 1 Tie.

What’s going on? Sampling biases?

And in September 2021:

There have been 47 Scottish independence polls by major pollsters in 2021. 22 have given a No majority, 5 have been tied and 20 have given a Yes majority.

Savanta ComRes, who only appeared for the first time in August 2020, have done 12 of them with 10 returning a No majority, 1 tied and only 1 with a Yes majority, in January. That’s a very negative set compared to Opinium with 2 Yes and 1 tied or Panelbase with 4 Yes and 3 No, or Ipsos MORI with 2 Yes and 1 No.

Savanta ComRes has attracted concerned comment before. On March 7th, I posted:

James Kelly, the thinking man’s James Kelly, has posted these comments on the Savanta Com Res poll for the Scotsman which purports to show a clear lead for NO, for the first time in more than a year:

“these figures are not weighted for voter turnout, with further polling expected this week to show a clearer impact of the inquiry on Scottish independence voting intention..these figures on Scottish independence are not directly comparable with previous polls on the subject, due to this and the nature of the poll.”

In other words this is not a standard poll, and we have absolutely no idea whether a poll conducted in the normal way would show a Yes lead or a No lead.  Which begs the obvious question: why on earth has this poll been published?

The other two polls giving NO a lead, in the last few days, from Survation/Daily Record and Panelbase/Sunday Times, returned a non-significant 1% lead.

MSM_Monitor has some thoughts on why the poll was published at all:

Why might a failure to weight the figures for voter turnout matter? Well SNP and Conservative expressions of turnout tend to be higher, 82% and 80% in the last Savanta ComRes poll for Westminster voting intentions, while Labour and LibDem turnout tends to be lower, 70% and 59% in the same poll. The same pattern can be found in the poll before.

A pattern suggests something different being done, perhaps in the construction of their sample?

ComRes has been accused of an ‘inbuilt conservative bias‘ in the past:


5 thoughts on “Scotsman/Savanta polls: NO if you want it?

  1. The piper pays the tune. The polls are conductedbybiased individuals to make money for them and hedge funds to manipulate the results. Making £Billions gambling on the result. Especially in tight margins. The pollsters censored and fined so many times. They still get away with it. Any poltry fine is far outweighed by illegal actions and profits.

    Cambridge Analytica etc. Data disorder non accurate polling methods to accommodate the required outcome. People who believe in Independenceneedto get out and vote every election. A highe4 turnout to get rid Iraq f the opposition.
    Vote for Independence every election and take another as well. Independence will come whenpeople vote for it a higher turnout. Instead of people sitting on their hands and not turning out to vote SNP/SNP every election. To defeat the opposition. Pure and simple. Get outto vote for SNP/Independence ever election . If people want it as they claim.Some claimto support Independence then vote for the opposition. Or do not vote. SNP. Vote for Independence. They poll YES but vote No Or do not vote at every election, A polling descrepancy. Leads to misleading polls. Especially in tight margins. The Polls are used to influence the result. Or under estimate support. The bookies use more accurate polling methods. The Pollsters methodology is suspect. Especially in tight margins. Censored and fined many times for corruption. They should not be allowed in the Purdah period. Too suspect methodology for accurate results. Gerrymandering to make illegally monies for corrupt people. The piper pays the tune to manipulate the results.

    People who be,ieve in Independence should get out and vote SNP at every election. Join and donate to achieve it. Stop criticising on the sidelines. If they want Independence. Vote for it.Use the vote or lose it, getting rid of the opposition will bring it sooner.

    Liked by 1 person

  2. Core to unionism in Scotland and NI is the belief that they represent a majority view in their respective communities.
    Take that away and their beliefs crumble.
    Essential for the British state to keep pushing the line that unionism is the settled will of those communities and by trying to demoralise their opponents by saying,even if the majority want independence,they won’t allow it.
    We’ll see.

    Liked by 2 people

  3. As Gordon says, “The piper pays the tune”. It’s not difficult, and would explain why the trend that Sevanta show – just fix the weightings in a particular way. Easy really. Start off from the conclusion and work backwards to the question – normal Unionist strategy

    Liked by 1 person

  4. bringiton
    DECEMBER 22, 2022 AT 10:12 AM
    Core to unionism in Scotland and NI is the belief that they represent a majority view in their respective communities.

    You could have stopped right there if you had highlighted
    “they represent a majority view in their respective communities”
    But not in our country.

    Liked by 1 person

  5. Going back through their record, Savanta in 23 polls have only had 5 for Yes and 2 tied. Even more interesting is that their first poll was in August 2020, which came out for Yes, and after that in their first 6 polls, 5 came out for Yes. Of course, during this time (second half of 2020 and early 21) polls were consistently Yes. But guess who found the first “no”?
    Since then – in 17 subsequent polls – Savanta have come out for No on 15 occasions (2 tied).
    In their last 10 polls, its interesting that while the No vote varies a bit – 46-49. the Yes vote varies much less 44-46. Compare this to Yougov, where the Yes vote in the same period varies between 39 and 47. Or Ipsos Mori where it varies between 44 and 53.
    Do Savanta use a panel of the same (they would claim) “representative” voters? If so, what is interesting is much less the level, but the trend. In that case, what they are showing is that not much is changing, at least among their “panel” as in their last 5 polls the Yes vote has been 44 or 45.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.