If you want a ‘No’ for independence go to Savanta or Survation for your poll

Image: Seattle Times

In February 2021, Savanta for ITV News reported at 4% lead for Yes. By the end of December 2022 they had reported 15 leads for No, no more for Yes and 2 tied.

On 21 December they broke a run of 7 leads for Yes by other pollsters, with a 2% lead for No, on behalf of the Scotsman.

In another run of Yes leads by other pollsters in mid March to mid April 2021, only they, for the Scotsman, returned a tie.

Survation had 1 for Yes and 6 for No, which, though with a smaller sample, does look like roughly the same ratio.

The contrast with other pollsters is marked.

In the same period, February to December 2021, Panelbase had 5 leads for Yes, 7 for No and 1 tied.

YouGov had 2 leads for Yes and 3 for No.

Ipsos-MORI had 2 for Yes and 1 for No.

The others have not polled often enough for comparison.

In a year of inevitably shifting opinion, Panelbase, YouGov and Ipsos-MORI seem to reflect that while Savanta and Survation seem able to find a No lead regardless.

What could be causing this?

Different sampling with perhaps a bias toward older and rural communities in the Savanta and Survation samples? I’ve a had a look but see no sign of that.

A subtle influence of the group commissioning the poll?

All Scotsman polls were done by Savanta.

All STV polls were done by Ipsos MORI.

Nearly all Sunday Times polls were done by Panelbase.

Times polls were mostly done by YouGov.

Any other ideas?

From Golfnut:

Sorry to bang on about weightings again, but it’s quite an important point. The No vote intention in recent polling is very closely linked to the % of the sample that voted No as per this chart. So, the higher the weighting of 2014 No voters the higher the No vote share. pic.twitter.com/gDYjSepJCv

— Independent Voices 🇺🇦🏴󠁧󠁢󠁳󠁣󠁴󠁿 (@Celebs4indy) October 5, 2022




7 thoughts on “If you want a ‘No’ for independence go to Savanta or Survation for your poll

  1. The pollsters have been censored and fined so many times. Pay a fine and get away with it. Making £Millions out of it. Funding Political Parties. Tax evading political pollsters manipulating the results. To make £Billions for hedge funds. When they are caught out. ‘It wasnae me Gov’.

    YouGov Tory Cabinet minster having to pay back £Millions tax evaded. Denying it.

    Westminster denying fraudsters gerrymandering. They usually get it wrong but profit from wrong doing. Cambridge Analytica etc acting illegally and paying an irrelevant fine. Getting away with the loot. Instead of going to jail. The piper pays the tune to manipulate the results. Gerrymanding influence. The Bookies get it right. Making a shedload. They are better at statistics. Gamblers never win.


  2. Weighting, Celebs for Yes.

    Liked by 1 person

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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.