Wednesday, February 5, 2020
Tuesday, February 4, 2020
Scot Goes Pop / Panelbase poll: By decisive margins, the Scottish public say the election result *does* give the SNP a mandate to hold an independence referendum, and that Brexit *is* a big enough change of circumstances to justify another indyref
I have two more results for you tonight from the Panelbase poll this blog commissioned with your help. The first question is an extremely simple and obvious one that no-one seems to have asked before. The UK government insist that the SNP's landslide win in the general election is neither here nor there, and that it's certainly not a mandate to hold a second independence referendum (even though a referendum this year was specifically promised in the SNP's manifesto). But do the public agree with the UK government? Let's find out...
In last month's general election, the SNP won 48 of the 59 Scottish seats in the House of Commons. Do you believe this result gives the SNP a mandate from the Scottish people to hold a second independence referendum?
Yes 48%
No 42%
With Don't Knows excluded, that works out as roughly 53% who say there is a mandate for an indyref, and 47% who say there is not. In BBC lingo, that's a decisive margin, although the main reason it's a relatively close result is the near-unanimity of opinion among Conservative voters. No fewer than 95% of respondents who voted Tory in December deny that the SNP have a mandate, which I must say is rather ironic given that they're the ones who voted for a party that claimed the holding of an indyref was on the ballot paper. It takes a fair bit of cognitive dissonance to go down that road and then tell yourself that the very clear outcome of the election is irrelevant, but that's what seems to have happened.
There's more of a mixture of views among the supporters of other unionist parties, though - 24% of Liberal Democrat voters accept the SNP's mandate, and so do a very healthy 39% of Labour voters. (But remember that over one-third of Labour-voting respondents support independence anyway.)
The next question I decided to ask was another very obvious one. I'm sure something similar must have been asked in previous polls, but maybe not in quite these precise terms.
When Scotland voted against independence in the 2014 referendum, it was not known that Britain would leave the EU a few years later. Do you think Brexit is a big enough change of circumstances to justify holding a second independence referendum?
Yes 52%
No 41%
If Don't Knows are excluded, approximately 56% of those who expressed a view think that Brexit justifies another indyref, and 44% disagree. I must admit I was quite surprised by the size of that majority, and it suggests to me that a significant minority of anti-independence voters must understand and agree with the argument that the No vote in 2014 was at least partly won on a false prospectus, ie. that it would preserve the EU citizenship of Scottish residents. What's the fairest thing to do in those circumstances? The blindingly obvious answer is "hold a re-run".
Incidentally, no single group of respondents is particularly responsible for the bigger majority on the second question - there's just a smidgeon more support in lots of different places. Even Tory voters were slightly more likely to answer Yes to the second question - although admittedly the difference is merely between 2% on the first question and 4% on the second!
* * *
There are more questions from the poll still to be released, and in my opinion the most significant result of all is still to come. If you'd like to be the first to know when the remaining results are published, you can follow me on Twitter HERE.
In last month's general election, the SNP won 48 of the 59 Scottish seats in the House of Commons. Do you believe this result gives the SNP a mandate from the Scottish people to hold a second independence referendum?
Yes 48%
No 42%
With Don't Knows excluded, that works out as roughly 53% who say there is a mandate for an indyref, and 47% who say there is not. In BBC lingo, that's a decisive margin, although the main reason it's a relatively close result is the near-unanimity of opinion among Conservative voters. No fewer than 95% of respondents who voted Tory in December deny that the SNP have a mandate, which I must say is rather ironic given that they're the ones who voted for a party that claimed the holding of an indyref was on the ballot paper. It takes a fair bit of cognitive dissonance to go down that road and then tell yourself that the very clear outcome of the election is irrelevant, but that's what seems to have happened.
There's more of a mixture of views among the supporters of other unionist parties, though - 24% of Liberal Democrat voters accept the SNP's mandate, and so do a very healthy 39% of Labour voters. (But remember that over one-third of Labour-voting respondents support independence anyway.)
The next question I decided to ask was another very obvious one. I'm sure something similar must have been asked in previous polls, but maybe not in quite these precise terms.
When Scotland voted against independence in the 2014 referendum, it was not known that Britain would leave the EU a few years later. Do you think Brexit is a big enough change of circumstances to justify holding a second independence referendum?
Yes 52%
No 41%
If Don't Knows are excluded, approximately 56% of those who expressed a view think that Brexit justifies another indyref, and 44% disagree. I must admit I was quite surprised by the size of that majority, and it suggests to me that a significant minority of anti-independence voters must understand and agree with the argument that the No vote in 2014 was at least partly won on a false prospectus, ie. that it would preserve the EU citizenship of Scottish residents. What's the fairest thing to do in those circumstances? The blindingly obvious answer is "hold a re-run".
Incidentally, no single group of respondents is particularly responsible for the bigger majority on the second question - there's just a smidgeon more support in lots of different places. Even Tory voters were slightly more likely to answer Yes to the second question - although admittedly the difference is merely between 2% on the first question and 4% on the second!
* * *
There are more questions from the poll still to be released, and in my opinion the most significant result of all is still to come. If you'd like to be the first to know when the remaining results are published, you can follow me on Twitter HERE.
Monday, February 3, 2020
Scot Goes Pop / Panelbase poll on independence: Yes vote surges to 52% - the highest figure for three-and-a-half years. And pro-indy parties are on course to take 57% of the seats at Holyrood 2021.
I can now reveal the first results of the Panelbase poll commissioned by this blog, with the very generous help of readers via the crowdfunder. First up are the headline independence numbers, which show a clear Yes majority.
Should Scotland be an independent country? (Panelbase poll for Scot Goes Pop, 28th-31st January 2020):
Yes 52% (+5)
No 48% (-5)
(Note: Before the exclusion of Don't Knows, the figures are Yes 49%, No 46%, Don't Know 6%)
That’s the best result for Yes in any poll conducted by a member firm of the British Polling Council since the immediate aftermath of the EU referendum in the summer of 2016. It’s a particularly remarkable result for a poll conducted by Panelbase, which was one of two pollsters (the other was YouGov) that in 2017 and 2018 frequently reported a Yes vote that was a little below the 45% achieved in the 2014 referendum. It’s possible that part of the reason for the turnaround is that Panelbase have now (in common, presumably, with most other firms) replaced weighting by 2017 past vote with weighting by 2019 past vote.
That means the two political weightings now applied by Panelbase are 2014 indyref vote, and 2019 general election vote. The former is still leading to a very sharp downweighting of respondents who voted Yes in the indyref – the 469 Yes voters in the unweighted sample have been scaled down to 405. There is actually quite a significant downweighting of SNP voters from the 2019 election as well, but it’s nowhere near as big an adjustment as in the last Panelbase poll that used 2017 past vote weighting. So I would guess (and it can only be a guess) that part of the swing to Yes in the poll reflects a genuine change in public opinion, and part of it can be explained by the methodological revision, which will have corrected what may have been a small but significant underestimate of the Yes vote in polls published last year.
The fieldwork was conducted entirely in the days leading up to Brexit – so it remains to be seen what impact the reality of leaving the EU will have. It could be that it will push Yes higher, although the counter-argument is that the minimal change during the transitional period could lead to a false sense of security and people saying “ah, that’s not so bad, then”.
This poll differs from YouGov’s numbers last week in suggesting that support for independence is somewhat higher among men than among women. It also suggests a significantly higher Yes support among Labour voters than YouGov reported – and if it’s true that more than one-third of the rump Labour vote in the 2019 election want Scotland to be an independent country, Richard Leonard’s headache has just got even worse. It means that Labour could have even further to fall in Scotland if they follow the ultra-unionist path advocated by certain candidates in the party’s leadership and deputy leadership elections.
Respondents who voted Labour in December 2019:
Yes 35%
No 54%
Don't Know 11%
And now to the Scottish Parliament voting intention numbers, which if anything are even more sensational.
Scottish Parliament constituency ballot:
SNP 50% (+7)
Conservatives 26% (n/c)
Labour 14% (-5)
Liberal Democrats 7% (-1)
Greens 3% (+1)
Scottish Parliament regional list ballot:
SNP 47% (+9)
Conservatives 25% (-1)
Labour 14% (-4)
Greens 7% (+1)
Liberal Democrats 7% (-2)
In recent years it’s been pretty common for Panelbase polls to suggest that the pro-independence parties might fall slightly short of a majority of seats in the 2021 election. But this poll could scarcely be more different, because the figures would almost certainly translate into a single-party overall majority for the SNP. One projection model suggests the SNP would have 67 seats, with all other parties in combination on 62. The SNP and Greens combined would push pro-indy representation up to 74 seats, leaving the unionist parties with just 55. And of course one thing that leaps out straight away is that pro-independence parties have an absolute majority of the popular vote on both ballots, which in theory leaves open the possibility of using the 2021 election to seek an outright mandate for independence itself – an idea that many people in the SNP have touted, but which so far Nicola Sturgeon doesn’t appear attracted to.
Full seats projection: SNP 67 (+4), Conservatives 32 (+1), Labour 17 (-7), Greens 7 (+1), Liberal Democrats 6 (+1)
* * *
Now that there have been three post-election polls on independence (from Survation, YouGov and Panelbase respectively), it's possible to produce a rough-and-ready 'poll of polls', and of course it's a very simple calculation. The three average out as: Yes 51%, No 49%.
There are several more questions to come from the Panelbase poll, and some of the results are pretty incredible. I had fairly modest expectations about one of the questions in particular, and my jaw dropped to the floor when I saw the outcome this morning. The remaining results will be released gradually over the coming days - if you'd like to be the first to know about them, you can follow me on Twitter HERE.
Should Scotland be an independent country? (Panelbase poll for Scot Goes Pop, 28th-31st January 2020):
Yes 52% (+5)
No 48% (-5)
(Note: Before the exclusion of Don't Knows, the figures are Yes 49%, No 46%, Don't Know 6%)
That’s the best result for Yes in any poll conducted by a member firm of the British Polling Council since the immediate aftermath of the EU referendum in the summer of 2016. It’s a particularly remarkable result for a poll conducted by Panelbase, which was one of two pollsters (the other was YouGov) that in 2017 and 2018 frequently reported a Yes vote that was a little below the 45% achieved in the 2014 referendum. It’s possible that part of the reason for the turnaround is that Panelbase have now (in common, presumably, with most other firms) replaced weighting by 2017 past vote with weighting by 2019 past vote.
That means the two political weightings now applied by Panelbase are 2014 indyref vote, and 2019 general election vote. The former is still leading to a very sharp downweighting of respondents who voted Yes in the indyref – the 469 Yes voters in the unweighted sample have been scaled down to 405. There is actually quite a significant downweighting of SNP voters from the 2019 election as well, but it’s nowhere near as big an adjustment as in the last Panelbase poll that used 2017 past vote weighting. So I would guess (and it can only be a guess) that part of the swing to Yes in the poll reflects a genuine change in public opinion, and part of it can be explained by the methodological revision, which will have corrected what may have been a small but significant underestimate of the Yes vote in polls published last year.
The fieldwork was conducted entirely in the days leading up to Brexit – so it remains to be seen what impact the reality of leaving the EU will have. It could be that it will push Yes higher, although the counter-argument is that the minimal change during the transitional period could lead to a false sense of security and people saying “ah, that’s not so bad, then”.
This poll differs from YouGov’s numbers last week in suggesting that support for independence is somewhat higher among men than among women. It also suggests a significantly higher Yes support among Labour voters than YouGov reported – and if it’s true that more than one-third of the rump Labour vote in the 2019 election want Scotland to be an independent country, Richard Leonard’s headache has just got even worse. It means that Labour could have even further to fall in Scotland if they follow the ultra-unionist path advocated by certain candidates in the party’s leadership and deputy leadership elections.
Respondents who voted Labour in December 2019:
Yes 35%
No 54%
Don't Know 11%
And now to the Scottish Parliament voting intention numbers, which if anything are even more sensational.
Scottish Parliament constituency ballot:
SNP 50% (+7)
Conservatives 26% (n/c)
Labour 14% (-5)
Liberal Democrats 7% (-1)
Greens 3% (+1)
Scottish Parliament regional list ballot:
SNP 47% (+9)
Conservatives 25% (-1)
Labour 14% (-4)
Greens 7% (+1)
Liberal Democrats 7% (-2)
In recent years it’s been pretty common for Panelbase polls to suggest that the pro-independence parties might fall slightly short of a majority of seats in the 2021 election. But this poll could scarcely be more different, because the figures would almost certainly translate into a single-party overall majority for the SNP. One projection model suggests the SNP would have 67 seats, with all other parties in combination on 62. The SNP and Greens combined would push pro-indy representation up to 74 seats, leaving the unionist parties with just 55. And of course one thing that leaps out straight away is that pro-independence parties have an absolute majority of the popular vote on both ballots, which in theory leaves open the possibility of using the 2021 election to seek an outright mandate for independence itself – an idea that many people in the SNP have touted, but which so far Nicola Sturgeon doesn’t appear attracted to.
Full seats projection: SNP 67 (+4), Conservatives 32 (+1), Labour 17 (-7), Greens 7 (+1), Liberal Democrats 6 (+1)
* * *
Now that there have been three post-election polls on independence (from Survation, YouGov and Panelbase respectively), it's possible to produce a rough-and-ready 'poll of polls', and of course it's a very simple calculation. The three average out as: Yes 51%, No 49%.
There are several more questions to come from the Panelbase poll, and some of the results are pretty incredible. I had fairly modest expectations about one of the questions in particular, and my jaw dropped to the floor when I saw the outcome this morning. The remaining results will be released gradually over the coming days - if you'd like to be the first to know about them, you can follow me on Twitter HERE.
Super Survation survey sees Yes soar into slight lead
For the avoidance of doubt, this is not the Scot Goes Pop-commissioned poll, which was conducted by another firm and will be published this evening. It looks like Survation have been sitting on this for a little while, because the questions were asked as part of the Progress Scotland poll conducted between the 20th and 22nd of January.
Should Scotland be an independent country?
Yes 50.2%
No 49.8%
Scottish Parliament constituency voting intentions:
SNP 51%
Conservatives 23%
Labour 17%
Liberal Democrats 7%
Scottish Parliament regional list voting intentions:
SNP 38%
Conservatives 21%
Labour 19%
Greens 9%
Liberal Democrats 9%
As I always point out, Survation's regional list figures should be taken with a heavy dose of salt, because they generally seem to be markedly worse for the SNP and markedly better for the Greens than other firms suggest. I strongly suspect the problem lies in the way the question is posed, which may lead some respondents to wrongly suppose that they're being asked for a second preference vote.
Should Scotland be an independent country?
Yes 50.2%
No 49.8%
Scottish Parliament constituency voting intentions:
SNP 51%
Conservatives 23%
Labour 17%
Liberal Democrats 7%
Scottish Parliament regional list voting intentions:
SNP 38%
Conservatives 21%
Labour 19%
Greens 9%
Liberal Democrats 9%
As I always point out, Survation's regional list figures should be taken with a heavy dose of salt, because they generally seem to be markedly worse for the SNP and markedly better for the Greens than other firms suggest. I strongly suspect the problem lies in the way the question is posed, which may lead some respondents to wrongly suppose that they're being asked for a second preference vote.
Sunday, February 2, 2020
Al Jazeera film on Brexit Day
Yesterday morning I made my way to a swanky coffee shop in the west end of Glasgow to be filmed speaking about Brexit Day for an Al Jazeera social media feature, along with Remain-supporting Australian Tahnee Conn, Leave-supporting Tom Walker, and the no-introductions-required Gerry Hassan. You can watch it via the tweet below.
"Absolutely gutted depressed. Hate it."
— Al Jazeera English (@AJEnglish) February 2, 2020
What does it mean for the UK to leave the EU after 47 years of membership? We speak to Remainers and Brexiters. pic.twitter.com/79E5jcCY4m
Saturday, February 1, 2020
Scot Goes Pop poll update
Just another quick update for the people who contributed to the crowdfunder for this blog's forthcoming poll on independence. The fieldwork has now been completed, and I've been given a tentative indication of the likely result on the main question. But I'm not going to drop any hints at all, because it's provisional and may change. I should have the full results by Monday at the latest, and then I'll probably drip-feed them to you over a few days.
Meanwhile, I have an article in today's edition of The National expressing my scepticism about the ultra-cautious approach set out by Nicola Sturgeon yesterday. You can read it HERE.
Meanwhile, I have an article in today's edition of The National expressing my scepticism about the ultra-cautious approach set out by Nicola Sturgeon yesterday. You can read it HERE.
Friday, January 31, 2020
It gives me no pleasure to say this, but a major strategic error has just been made
I only caught the last few seconds of that, but it sounded like Nicola Sturgeon has firmly chosen the Andrew Wilson magical thinking approach of "persuade and the obstacle will somehow vanish by some unspecified means"? If so, I have to be honest and say that's a mistake.— James Kelly (@JamesKelly) January 31, 2020
Rule No 1: It is impossible, literally *impossible* to be sure of a win before a referendum is called. The really big changes in public opinion (in whatever direction they occur) will happen once the campaign is underway.— James Kelly (@JamesKelly) January 31, 2020
Compared to the 2014 campaign, 51-49 is in "beyond wildest dreams" territory as a starting point. Are we really waiting for unachievable levels of support (like 60% or 67%) before doing anything? If so, the leadership have given up on independence. They just don't know it yet.— James Kelly (@JamesKelly) January 31, 2020
You're missing the point. If we're "learning the lesson from David Cameron" that polls can be wrong, that's an argument for saying we shouldn't call a referendum even if we're 20, 30, 40 points ahead. In other words we can never do it.— James Kelly (@JamesKelly) January 31, 2020
Oh for heaven's sake. There's nothing "illegitimate" about legislating for a referendum and then allowing the Supreme Court to decide whether it can take place.— James Kelly (@JamesKelly) January 31, 2020
Her tone was entirely off caution is fine (up to a point) if you sound convincingly like trying to push things forward. Nicola missed the target today.— Chris ward (@rampitchfork) January 31, 2020
* * *
I have more analysis of yesterday's sensational YouGov poll in an article at The National - you can read it HERE.
Thursday, January 30, 2020
Drama as YouGov poll shows a pro-independence majority
The first post-election poll on independence has been published, and it suggests that the long-anticipated (but highly elusive) swing to Yes by Remainers who voted No in 2014 may have finally taken place on a really telling scale.
Should Scotland be an independent country? (YouGov, 22nd-27th January 2020):
Should Scotland be an independent country? (YouGov, 22nd-27th January 2020):
Yes 51% (+7)
No 49% (-7)
Bear in mind that YouGov have in recent years been one of the most No-friendly firms, so it shouldn't be assumed that 51% is as good as it's going to get for Yes - it's conceivable that other firms will show something a little better, but we'll have to wait and see. On the other hand, the 7% swing to Yes may be a touch exaggerated, because the previous YouGov poll looks very much like an outlier in retrospect - it showed a sharp dip in the Yes vote that wasn't picked up by other pollsters. But it seems highly probable that YouGov's new numbers are reflecting some sort of genuine swing on the ground - 49% was the highest Yes vote they reported last year, and they haven't shown an outright Yes lead since 2015.
An intriguing supplementary question attempts to measure the depth of both support for and opposition to independence, and finds that absolutely definite Yes and No voters are tied on 35% apiece. For my money that's a highly encouraging finding, because there's always a danger that a sudden swing might be based on rather shaky foundations. That doesn't appear to be the case here - there's no longer any underlying 'natural majority' for No.
The unionist parties will of course ignore the headline result on independence and point to the finding that a majority are opposed to holding an independence referendum this year. But the snag is that 44% are in favour of holding a referendum within the next five years (ie. within the next Holyrood parliamentary term) and only 39% are opposed. With Don't Knows excluded, that's a pro-referendum majority of around 53% to 47%. Given that the Tory government take the extremist and ludicrous line that no referendum can be held until after Nicola Sturgeon's death (which could potentially mean another fifty years of delay!), it's safe to say that the public are a lot, lot closer to the SNP's view on referendum timing than they are to the Tories' view.
Thoughtful Scottish Tory MSPs (if such a thing exists) will be deeply concerned that this poll shows they are about to make an error of historic proportions by confirming Jackson Carlaw as their official leader. It looks like he can't be absolved of blame for the election disaster last month - he has a dismal net personal rating of -29, which is 29 points worse than Nicola Sturgeon's and almost as bad as Richard Leonard's. (Yes, it's that bad.)
From a personal point of view, I breathed a sign of relief when I read through the list of questions that YouGov asked. As you know, this blog will be publishing a poll in the very near future, and I was slightly concerned that there might be a significant amount of overlap between YouGov's questions and the ones that I asked. But as it turns out, there's no real duplication at all, except on the independence question itself. (Unless of course YouGov have a second batch of questions that they haven't released yet, but time will tell.)
Bear in mind that YouGov have in recent years been one of the most No-friendly firms, so it shouldn't be assumed that 51% is as good as it's going to get for Yes - it's conceivable that other firms will show something a little better, but we'll have to wait and see. On the other hand, the 7% swing to Yes may be a touch exaggerated, because the previous YouGov poll looks very much like an outlier in retrospect - it showed a sharp dip in the Yes vote that wasn't picked up by other pollsters. But it seems highly probable that YouGov's new numbers are reflecting some sort of genuine swing on the ground - 49% was the highest Yes vote they reported last year, and they haven't shown an outright Yes lead since 2015.
An intriguing supplementary question attempts to measure the depth of both support for and opposition to independence, and finds that absolutely definite Yes and No voters are tied on 35% apiece. For my money that's a highly encouraging finding, because there's always a danger that a sudden swing might be based on rather shaky foundations. That doesn't appear to be the case here - there's no longer any underlying 'natural majority' for No.
The unionist parties will of course ignore the headline result on independence and point to the finding that a majority are opposed to holding an independence referendum this year. But the snag is that 44% are in favour of holding a referendum within the next five years (ie. within the next Holyrood parliamentary term) and only 39% are opposed. With Don't Knows excluded, that's a pro-referendum majority of around 53% to 47%. Given that the Tory government take the extremist and ludicrous line that no referendum can be held until after Nicola Sturgeon's death (which could potentially mean another fifty years of delay!), it's safe to say that the public are a lot, lot closer to the SNP's view on referendum timing than they are to the Tories' view.
Thoughtful Scottish Tory MSPs (if such a thing exists) will be deeply concerned that this poll shows they are about to make an error of historic proportions by confirming Jackson Carlaw as their official leader. It looks like he can't be absolved of blame for the election disaster last month - he has a dismal net personal rating of -29, which is 29 points worse than Nicola Sturgeon's and almost as bad as Richard Leonard's. (Yes, it's that bad.)
From a personal point of view, I breathed a sign of relief when I read through the list of questions that YouGov asked. As you know, this blog will be publishing a poll in the very near future, and I was slightly concerned that there might be a significant amount of overlap between YouGov's questions and the ones that I asked. But as it turns out, there's no real duplication at all, except on the independence question itself. (Unless of course YouGov have a second batch of questions that they haven't released yet, but time will tell.)
Subscribe to:
Posts (Atom)