A brief note on opinion polls
A reader comment earlier today sent us off to do a little research. Specifically, we were interested in the results of opinion polling before the last referendum concerning the Scottish constitution – the 1997 vote on devolution. The results were fascinating.
In the days leading up to the referendum, two polls with standard sample sizes were conducted by System 3 for the Herald. They showed very similar results, averaging 61% of respondents in favour of a Scottish Parliament (with 23% opposed and 16% don’t-knows), and 46% in favour of that Parliament having tax-raising powers (31% against, 23% don’t-knows).
The second poll was conducted the day before the referendum. The actual vote, just 24 hours later, was 74-26 for the Parliament and 64-36 for tax-raising powers – overnight swings of 7% and 9% respectively in favour of the two propositions.
(Of the 16% of Don’t Knows on the first question, when it came to the crunch 13% had plumped for Yes compared to just 3% for No. On the tax-raising question, meanwhile, the 23% previously answering as Don’t Knows had divided 17% for Yes, 6% for No.)
This site welcomes both the continued determination of the Unionist parties to bully the Scottish electorate into making a stark choice between hope and fear once again, and also their complacency about the outcome.
It’s a clear sign of unionist desperation when their media puppets distort the poll findings. Not that I think polls are currently that important anyway.
I do however agree with the unionist camp when they argue for a clear one question referendum, but only after they disclose the truth. That is no Westminster government has any intentions of devolving real fiscal powers to The Scottish Parliament. Surely that is honest clarity?
Well, Gordon Brown pretty much disclosed the heck out of that yesterday. But at every interview from now on the “Better Together” campaign needs to be challenged on whether he was voicing their official position, until they give an answer.
I think you need to take turnout into account. If you drop the don’t knows from the opinion poll results, the actual figures come out at 73-yes 27-no to the first question and 60-yes 40-no to the second question. That’s actually damn close to the actual referendum result.
A valid point, but at over 60% turnout was pretty high. It seems reasonable to assume a substantial proportion of Don’t Knows did indeed make their mind up one way or the other.
Don’t knows jumping to the winning side at the last minute – I believe this may be a recognised phenomenon.
Goodanalysis Stu but , much as I wish it weren’t the case, the Fearties would have been a lot more relaxed in ’97? Asking them now to overcome their fears to vote for Independence will I fear be more difficult. However the two elements which will affect this most IMHO are the Unionists continuing to promise only ‘jam tomorrow’ and critically a commitment by the Yes campaign to telling us what will be different in an independent Scotland compared to telling us what will not change? And as I and others have said before a commitment to a written constitution would be something our Unionist friends could not answer!
A valid point, but at over 60% turnout was pretty high. It seems reasonable to assume a substantial proportion of Don’t Knows did indeed make their mind up one way or the other.
Turnout was over 60%, but that still leaves the non-voters (at over 30%) considerably more numerous than the don’t knows in the poll. My instinct with these things is to assume that the don’t knows overwhelmingly split to the “sorry I need to watch some paint dry” group, and then some other people who did say yes or no in the opinion poll will also not bother to show up.
“Goodanalysis Stu but , much as I wish it weren’t the case, the Fearties would have been a lot more relaxed in ’97?”
Perhaps. But you could also argue that in some ways devolution was in fact a bigger step. Independence is just an extension of what we already have – the Scottish Parliament. There’s a case to be made that setting up our own government at all, even a limited one, was in fact a more dramatic act of “separation” from Mummy Westminster.
“My instinct with these things is to assume that the don’t knows overwhelmingly split to the “sorry I need to watch some paint dry” group, and then some other people who did say yes or no in the opinion poll will also not bother to show up.”
Entirely possible, but since there’s no way of knowing either way it’s an endlessly-debatable point.
What does strike me as bizarre is that almost 40% of eligible voters had no opinion, and a reasonably similar number will almost certainly feel the same way about the independence referendum. Even in a cynical world that’s a remarkable level of disconnection for so many people to sustain.
People are more passionate for change so I would expect the result that eventuated in 1997 based on polls.
How much less than 50% gets the Yes result over the line would be a good question. I suspect 40% if they vote is enough, as 40% to match on an 80% turnout will not physically make the effort to vote No.
Interestingly, if you multiply out the above poll figures discounting the “don’t-know” responses, it works out very close to the actual result for the first question, but further off for the tax-raising issue.
Basically, 61/84 (that’s 61 yesses plus the 23 naysayers) is 72.6% in favour of a parliament; 46/77 (46Y + 31N) on the other hand comes out at just 59.7%. Viewed like this, the projection was slightly under two points out in the first question but nearly four points out for the tax powers (both within a reasonable margin of error).
That suggests that the uncertain vote largely stayed at home – a reasonable assumption, given how close the polls were to the actual referendum – but evidently there were some partially undecideds who did show up. These are folk who had a clear idea on the first question but were wavering on the second; having made the effort to vote for the first question, they appear to have been more inclined to go for the second yes too.
Turnout at the actual election, however, was just 60.4%, almost 14 points lower than the 84% who expressed a firm decision in the first question in the eve-of-vote polls. It’s intriguing to note that both sides seem to have held up equally well in the turnout, for the first question at least. I have always held a strong suspicion that the no respondents to independence polls are a comparatively soft vote and less likely to show up on the day, but perhaps I’m mistaken.
I wonder if the pollsters gave a breakdown of the various Yes-Yes / Yes-Dunno / Yes-No / No-Yes / No-Dunno / No-No responses. My gut instinct is that the Yes-Yes and No-No held up, more Yes-Dunno switched to Yes-Yes than the other way, and the Yes-No perhaps most likely to stay at home. I doubt there were many in the No-Yes or No-Dunno camps.
It’s all academic and a long time in the past, but I’m sure the likes of Scottish Skier and Oldnat could draw more lessons from this than my amateur chunnerings.
I think that the most interesting factor about the 1997 vote was the attempted manipulation of the public by adding the second tax-raising question. It was inserted to scare us into voting no to the first question.
As it happens, it backfired with the result showing yes-yes. This was clearly a poke in the eye to those who tried to manufacture a result through fear.
I hope the same happens in 2014.
Thinking back to 1997 and the various devolution campaigns. I can only recall the NO-NO and Yes-Yes groupings. However I was quite interested upon googling to find there was apparently a No-Yes campaign asking for FFA run by some Scottish Conservatives. I presume it had precisely no impact because I’m buggered if I can recall it, despite following events quite closely at the time.
link to bbc.co.uk
Apologies — you’ve all had the conversation while I was busy typing!
But while we’re on the subject of the ’97 referendum… It never ceases to amuse me that the people who throw their arms in the air to complain that the dastardly SNP might hold the independence referendum on the 700th anniversary of Bannockburn (or even just the same year as it) are the very same ones who held the devolution vote on September 11, 1997. I’m sure it’s entirely coincidental that that was the 700th anniversary of the Battle of Stirling Brig.
Tam Dalyell, who for a long time was my own MP, was a prominent No-Yes supporter too. He didn’t want a Parliament at all, but if there was going to be one he wanted it to have tax powers.
If you look at opinion polls for independence at the time of the devolution referendum, then these would suggest if put to the ballot, around 6 in 10 would have likely voted Y.
Certainly, I’d imagine every supporter of independence at that time voted Y-Y which, probably not coincidentally, was why the second Q got ~6 in 10. Many in the UK don’t realise how close things were at that time.
I’m happy for the the unionists to always focus on the Y and not the N in polls and particularly that they keep using Yougov (yougov have been at odds with all other indy polls since 2007 when their weighting methods apparently stopped working for Scotland in this and other areas such as UKGE VI). Really, they should be focussing on the N for it is that which they wish. If you look at long term averages, the N has never had a sustained majority over the past 15 years. Normally, it is around 4 in 10, and therefore similar to the Y.
What is different right now from the time of the devolution referendum is the unsures. Back then, this was only ~1 in 10. Now it is up to ~2 in 10. This happened following the SNP win in 2007. I’d speculate that this was because people now saw that an independence referendum could well be on the cards and it was time to start thinking about it as a very real option.
We are unlikely to see much difference in polls for a while yet, unless something major happens (e.g. coalition collapses and Tories rule as a minority); they will jump up and down within normal (very high) levels of variance based on differing methodologies and question wordings. Assuming devo max is ruled out completely and it becomes clear it will be a straight Y/N, then we should see things happen and quite quickly.
Keep in mind that up to 7 in 10 want major constitutional change. That is consistent. Also, contrary to MSM reports, the fact that there is no consistent majority for N shows that the union is not loved in Scotland. For many in the middle, it is but a practical choice. Give them the confidence in an independent Scotland and they’ll vote yes.
Personally, I’d like to see a poll asking ‘In an ideal world, would you like Scotland to be an independent country?’ I reckon you’d get up to 7 in 10 say yes to that.
And mind there was a 12.5% swing to the SNP in just 3 months ahead of May 11. Events dear boy, events…
you could also argue that in some ways devolution was in fact a bigger step.
No way! Devolution was simply devolving the Scottish Office powers to a democratic body (although perhaps a new development was the tax varying power?). It was supported by Unionist parties so Scots could feel comfortable that they were still doing the ‘sensible’ thing (full unashamed Scottishness being regarded as reckless).
Independence is a massive change in comparison.
I’m sure I read somewhere recently that when it comes to independence referendums that the YES vote on the day is always higher than the opinion polls indicate. I tried to Google it but not much luck. I’ll try again when I’m less busy.
South Sudan Referendum
Result was 98.83 YES to Independence
Quote from Wikipaedia.
“A similar survey carried out one year prior (to the Referendum) by the US-based National Democratic Institute had indicated that 90 percent of voters would vote for secession”
The last Quebec referendum had 67% no at the start of the campaign and only won by a whisker.
Still can’t find the article I read.
Two weeks before the vote in Quebec the Independence vote had risen to 55% for YES from the 33% at the beginning of the campaign. The Federal Govt then held two big “please stay” rallies within 2 or 3 days of the vote and at the first promised to give more power to Qebec if it voted NO. The result was 50.58% NO 49.42% YES. And the Federal Govt didn’t follow through on its promise of more powers for Quebec.
The question was:_
“Do you agree that Quebec should become sovereign after having made a formal offer to Canada for a new economic and political partnership within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12, 1995?”
and many people in Quebec were so confused by the question that they thought that a NO vote meant they would be gaining a form of Independence.
So can you blame AS for not allowing Westminster anywhere near the question?
“you could also argue that in some ways devolution was in fact a bigger step. Independence is just an extension of what we already have”
Pure comedy gold. Hilarious.
Is there anyone here who can tell me whether there the Times has a separate Scottish edition?
“Is there anyone here who can tell me whether there the Times has a separate Scottish edition?”
Yes, apparently it does.
Do feel free to justify your fatuous and worthless response on the other point with some sort of argument.
Oh all right then. Devolution was the creation of a subsidiary authority with certain specified powers under the auspices of our sovereign national parliament.
“Independence” involves secession and separation from the UK, with all the costs, uncertainties and hassles that involves. You know, things like international organisations and agreements, currency, lender of last resort, defence, having to run new and separate Treasury, DWP, MOD and so on.
I’m interested that you say “extension of what WE already have”. Didn’t realise they had devolution in Bath.
“I’m interested that you say “extension of what WE already have”. Didn’t realise they had devolution in Bath.”
Sigh. So your position is that anyone who ever leaves Scotland surrenders their nationality, yes?
Oh, it’s sm753 again – here to win friends and influence people. Good luck with that.
“Independence” involves secession and separation from the UK
Secession and separation!?! You forgot divorce and ‘wrenching away’.
“Sigh. So your position is that anyone who ever leaves Scotland surrenders their nationality, yes?”
No, the good news is that you will keep your British nationality forever.
You just won’t have a vote in the referendum, and the outcome will not affect you significantly. Which makes your intense interest in all of this a bit odd.
You’re not alone in this of course; it seems many of the most shrill online separatists sign themselves off from Los Angeles, Spain, Johannesburg and so on. Curious.
“No, the good news is that you will keep your British nationality forever.”
I’ve never had British nationality. Britain isn’t a nation.
“You’re not alone in this of course; it seems many of the most shrill online separatists sign themselves off from Los Angeles, Spain, Johannesburg and so on. Curious.”
Once again: there’s a fairly clear line between heated debate and disingenuous deliberate trolling aimed solely at causing offence and thereby disruption. Take heed of it.
Other people living in England who show intense interest in our referendum despite not having a vote or not being significantly affected by the outcome include David Cameron and Ed Milliband. Is their interest odd too?
What I find odd is that anyone can think Scotland’s independence will have little significant effect in the remains of the UK.
As UK politicians it is, ipso facto, legitimate for them to take an interest in any issue affecting any part of the UK.
I find all these random punters living in various places outside Scotland taking such an intense interest in Scottish politics somewhat curious. They don’t live here. It doesn’t affect them.
Technically I agree. There is no set definition of “nation” or “country”, the only term which has any legal standing is “state”. And that of course risks confusion with “State”, as in the US, Malaysian or German etc sub-“state” entity.
So our passports may be technically wrong to “state” “Nationality: BRITISH CITIZEN”.
But they do. Including yours. That’s what the rest of the world sees.
And you clearly have been happy to enjoy the benefits of your British Citizenship, residing in Bath, for quite a few years. Good choice of location; I was a Bristolian for 5 years, a Cheltenhamite for a year after that and this is the first summer for ages we’ve not been back to our usual cottage.
Thing is, I don’t have an intense online campaigning interest in Bristol / Somerset / Gloucestershire politics. Cos I don’t live there. Moi loverrrr.
sm753 – presumably you chastise those who have an “intense online campaigning interest” in American politics, but live on this side of the Atlantic? Or if we want to take things a bit closer to home, to all those in Scotland who stick their oar in when it comes to changes to the English health, education and justice systems?
(As for enjoying the benefits of British citizenship, you do realise Stu will still be able to reside in Bath after independence, thanks to the benefits of EU citizenship?)
“But they do. Including yours.”
I don’t have one.
And it’s not like many of us get a choice when we come to order a passport either….
Funny thing I’ve experienced about ‘British’ is that it’s only really used in the UK. I travel regularly with work overseas and rarely have people asked if I’m British. They ask if I’m English, maybe american, when I correct them with ‘Scottish’ they immeadiately realise the faux pas. All friendly of course as chances are if you meet someone speaking very good English abroad they will most likely be English or American.
Mrs SS is French and on the continent I generally always hear the French talking about English, Scottish etc; its rare to find people who don’t know the difference, certainly in my experience of France anyway. The word British just isn’t really used.
I guess that’s because British is not a nationality as Britain is not a country and never has been in the normal sense. That’s why when people from England, Scotland, Wales etc, when travelling, if asked their nationality, they say English, Scottish etc, rarely British.
“And it’s not like many of us get a choice when we come to order a passport either….”
Yes indeed… Michael McKeown, sorry… sm753 (get the right name for this blog) would just love to send off for his “Hello Kitty” design passport with Diamonti Union Jack on the back, but its just not a go…
“I don’t have one.”
Genuine question – why on earth not?
“Genuine question – why on earth not?”
Same reason I don’t have any ashtrays, tampons, goose medicine or helicopter fuel.
He’s a man of cloth and moves in mysterious ways?
sm753 says:
I find all these random punters living in various places outside Scotland taking such an intense interest in Scottish politics somewhat curious. They don’t live here. It doesn’t affect them.
I find the idea that the potential ending of the Act of Union being defined as “Scottish Politics” a bit disingenuous. Regardless of anyone’s point of view on independence, it will affect the remaining parts of the UK in a number of ways – so this issue is not a parochial “Scottish” one and it is perfectly legitimate for people from the rest of the UK to be interested (The volume of “subsidy junky” comments on newspaper websites and the BBC shows that it has captured many imaginations in England ;0 )
Travel broadens the mind, it is said.
Now I know they get up to some hoopy, crazy stuff in Bath… but y’know.
Travelling can also lead to you being fluent in shite in many languages.
@Rev. Stuart Campbell said: Same reason I don’t have any ashtrays, tampons, goose medicine or helicopter fuel.
Yeah, same thing happened to me when my eBay account got suspended.
“Yeah, same thing happened to me when my eBay account got suspended.”
😀
You win today!
[…] box makes. People respond differently to answering polls in comparison to casting real votes. Wings Over Scotland provides excellent analysis on this. The day before the 1997 devolution referendum, polling found […]
If it can be fairly assumed that the abstainers are not moved enough by the yes party to vote for them, then they should be regarded as a collective No.
The fairest way to conduct a referendum is to assume that everyone is satisfied with the status quo unless they actually vote against it. This would mean, in the case of Scots Independence, that only those voting YES need to vote. The result would then be an accurate and honest result rather than a change effected by,perhaps, 26% of the population.