A curious thing happened this last week.
Angus Robertson’s new company, Progress Scotland (‘commissions public opinion polling … to better understand how people’s views are changing in Scotland’) published its first research. If you missed it, Progress Scotland ‘aims to help prepare the case for Scotland to progress towards independence, keeping pace with the views of the people who make their lives here’.
A series of posts on their web site by their ‘Independent polling expert’ Mark Diffley has highlighted some of the main results of a survey carried out for them by Survation of over 2,000 members of that company’s panel of people living in Scotland. The article includes comment on survey results covering – the role views on the EU might play in any future independence referendum; the impact of Brexit on Scottish public opinion; voters who are undecided or open-minded about independence (sic – because they could equally be open-minded about the UK); and belief in whether independence will ever happen. The link above will take you to the articles.
The curious thing is that the full results of the survey seem to have been published only briefly on the Survation web site before being taken down. They have yet to appear on the Progress Scotland site. As any good pollster will tell you, only access to the full results of a survey allow you to judge whether any comment or interpretation is both correct and honest.
Luckily, the eagle-eyed newshounds at the Red Robin web site (‘We provide serious news from a left perspective’) spotted the full results of the survey and downloaded them before they disappeared. You can find them linked here and all due credit to Red Robin for being quick off the mark. As they say,
Data obtained by ex-SNP deputy leader Angus Robertson’s pro-independence polling outfit has shown low support for independence [my emphasis]. Robertson’s official write up failed to mention the findings, and the full tables were removed by Survation shortly after they were published.
Assiduous Tweeter @SteveSayersOne has put out a series of tweets over the last few days looking at aspects of the data obtained by Red Robin. The rest of this article examines one issue arising from the survey in more depth.
Notwithstanding all Progress Scotland’s finessing about the impact of EU and Brexit on people’s views about independence, the fundamental question on the subject is people’s propensity to support independence. As monitored by Professor John Curtice on his What Scotland Thinks web site, this is usually couched in terms of ‘How would you vote in a Scottish independence referendum if held now?’ It is a subject missing from Progress Scotland’s analysis of their own survey.
The Red Robin web site shows that not only did Progress Scotland collect data on the subject they did it in a more detailed and sophisticated way than the basic ‘How would you vote’ question of most surveys. The fact that they interviewed over 2,000 people also means their results will be somewhat more accurate than the typical 1,000 interviewees of most polls on the subject.
The sophistication of the Progress Scotland question is that they asked people to rank their commitment to independence on an 11 point scale, from ‘I completely support Scotland becoming independent’ (0) to ‘I completely support Scotland staying part of the UK’ (10). Here is a graph showing the results (it excludes 3% who said they didn’t know the answer to the question):
It’s impossible to know what went through people’s minds if they gave the question a score of, say, 3 or 7. But we can be pretty sure that at the extremes of the range anyone scoring 0 or 10 is pretty damn well fixed in their opinions (‘A recalcitrant nat/yoon’ as their opponents might say). We can also speculate that those at the mid-point of the range of possible answers, 5, might be most amenable to changing their opinion in either direction.
This is where the detail of the graph gets interesting.
First, note that 40% of people in the survey completely support Scotland staying part of the UK, compared with only 24% completely supporting independence, a much bigger gap than the 55:45 found time and again in other surveys.
Second, at each point as you move away from what I describe as the mid-point of the range, there are more people inclined to support staying in the UK than to support independence, for example, towards the two extremes, 5% at point 9 compared with 2% at point 1.
Third, the net result is that 58% of people lie on the UK side of that mid-point, compared with 35% on the independence side.
Another way of looking at the data is that since 64% completely support either staying in the UK or independence, only the remaining 36% are amenable to change, and some of them (for example, the 1s and 9s, maybe even the 2s and 8s) are also not likely to change. So if there were to be another independence referendum (not in the light of these figures very likely, at least in the short term), any pro-independence campaign would have a heck of a lot more persuasion to do of the 36% than a pro-UK campaign.
You might want to dip into the detailed tables to see what impact views on the EU and Brexit might have on public opinion. But bear in mind that the pro-indy/pro-UK figures here come despite 62% of Scots voting to remain in the EU referendum and despite three years of arguable muddle, incompetence and most recently crisis in negotiating the UK’s exit from the European Union, what one SNP MP rather unhappily described as a ‘cluster bourach’.
You might now understand why you have to search for the full results of Progress Scotland’s survey and why they did not want them to appear in public. Their obfuscation does not bode well for any reputation they might hope to build. But while there is never any reason for complacency on the pro-UK side of the argument, the survey also does not bode well for those advocating another early referendum on separation.
Note. The figures from the survey quoted above have been rounded to the nearest whole percentage point. The graph has been prepared from the more detailed numbers in Table 1 of the full survey results and so may vary marginally from figures in the text.