Monday, July 28, 2014

Panelbase announce huge methodological changes - the new poll is not directly comparable with anything that has gone before

Well, this is a bit startling.  It turns out that the exchange/argument/debate a number of us had on the previous thread was based on an entirely false premise, and indeed on a false premise twice over.  It was certainly the case that in their last poll for the Sunday Times back in May, Panelbase continued to use a slightly different question from the one they use for their other clients - although the full question didn't appear in the datasets, we know it was different because Calum Findlay took part in the poll and quoted it in full.  But if the datasets for the new Sunday Times poll are to be believed, the question has finally been brought into line with other Panelbase polls.  So in that sense the poll is directly comparable to the last Panelbase poll for Yes Scotland - but that fact has been totally eclipsed by the revelation that two other huge methodological changes have been made, meaning that no direct comparison with any previous Panelbase poll is possible.

The first change is one that I find extremely troubling, because it could be interpreted as the first sign that Peter Kellner's attempts to browbeat his fellow pollsters into adopting a more No-friendly methodology have borne fruit.  Basically Panelbase have decided to weight their results by a mixture of how people recall voting in the European elections in May, and how they recall voting in the Holyrood election in 2011.  Because too many people in the Panelbase sample recall voting SNP in May, the effect of adding a European weighting is to increase the No lead.  The problem with this approach was covered in Survation's response to Kellner's notorious article - they pointed out that far too many people recall voting in the European elections, full stop. So either the samples in online polls are hopelessly unrepresentative of the general population (in which case no amount of weighting can be sure of correcting the problem), or else a large number of people are saying they voted in May when they didn't.  If by any chance the latter is the case, downweighting respondents who claim to have voted SNP in May could be a monumental error, because it's perfectly possible that people who falsely say they voted are disproportionately likely to also say that they voted SNP.

I can only hope that no other pollsters follow suit.  This change would only be justified if respondents' recall of whether or not they voted in May bore at least some resemblance to the actual turnout.  That is clearly not even close to being the case.

On a more positive note, the second change is one that is long-overdue across the entire polling industry - Panelbase have started weighting by country of birth.  For some reason most pollsters seem to end up with too many English-born respondents in their sample, which probably goes some way towards explaining the disconnect between canvassing returns and published polls (although it certainly can't explain all of the disparity).  So in the new poll English-born respondents have been heavily downweighted from 162 to 94, while Scottish-born respondents have been upweighted from 794 to 864.  It's possible that Panelbase are the first company to introduce this form of weighting - the only one I'm not sure about is Ipsos-Mori, who routinely ask for people's country of birth as a "demographic" question (which implies that they weight by it), and yet they still seem to end up with too many English-born people in their final results.

Panelbase imply that the two changes have effectively cancelled each other out, although I'm slightly sceptical about that.  The precise wording used is "the net effect of these two new weights is statistically insignificant", and yet it's later suggested that changes between other recent Panelbase polls have also been statistically insignificant. So it seems this definition of statistical insignificance can encompass changes of 1% or greater - which in turn implies that it's perfectly possible the No lead would not be 7% in the new poll if the old methodology was still being used. So the claims of the No campaign that their lead has increased by 4% since the last Panelbase poll should be taken with a lorry-load of salt.

I'm very grateful to Ivor Knox of Panelbase for copying me in on an email that was sent to all of the firm's recent referendum clients (presumably the Sunday Times, Yes Scotland, the SNP, Wings Over Scotland and Newsnet Scotland). The main aim of the email seems to be to head off any "conspiracy theory" that Yes would have taken the lead in the new poll if the methodology hadn't been changed - "the old weighting would also show a No lead" is the emphatic message. However, there's no additional information about whether that lead would have been as big as 7%, which takes us back to the point I made above.

What is revealed, however, is that if weighting by past vote recall had only factored in the European elections (as opposed to both the European and Holyrood elections), and if there had been no weighting by country of birth, the Yes vote after Don't Knows are excluded would have been in the region of 42-43%. That's fascinating, because it more or less explains the difference between YouGov and the other online pollsters - if Panelbase make the sort of adjustment to their past vote weighting that Peter Kellner would approve of, it's enough to take their numbers down to YouGov-type levels, whereas by the same token it's presumably the case that all YouGov would have to do to produce Panelbase/Survation-type numbers is ditch the artificial "Kellner Correction" and introduce weighting by country of birth.

Another important question is what would have happened if the only methodological change Panelbase had made was the introduction of weighting by country of birth. It clearly wouldn't have been quite enough to push Yes above 50%, but the gap would certainly have been narrower.

51 comments:

  1. If the changes are "statistically insignificant" than, erm, why bother making them?

    Weighting by EU voting just seems wrong. With such a low turnout, surely it's just asking for people to make up who they voted for, rather than admit they didn't vote?

    ReplyDelete
  2. Blah, blah, blaaaaaaah....

    ReplyDelete
  3. Anon : Oh go on, tell us which one you are.

    ReplyDelete
  4. Anonymous: Are you a YES voter saying you don't believe any of the polls or, are you a NO troll trying again to diss what James has written? We should be told.

    ReplyDelete
  5. "So it seems this definition of statistical insignificance can encompass changes of 1% or greater - which in turn implies that it's perfectly possible the No lead would not be 7% in the new poll if the old methodology was still being used."

    A VERY important point.

    They (and other pollsters) do seem to be scratching around trying to find a methodology which will justify their and their clients' preconceived ideas of the result they want.

    What faith can there be in their results when methodologies are changed on a whim?

    ReplyDelete
  6. So glad you and your blog are here, James, much appreciated.

    It says a lot about this campaign that every time the data sets are released they invariably show that the polls aren't as favourable to no campaign as the original Better Together and MSM headlines suggest.

    It happens every time. Anyone would think the intention is to mislead...

    ReplyDelete
  7. Well. Looking down this list, I am not surprised by anything they do.

    http://www.britishpollingcouncil.org/officers.html

    ReplyDelete
  8. What exactly could be the justification for weighting by Euros participation? Simply because the Euros are more recent than the last Holyrood election? Or is there some attempt, as with YouGov, to create a distinction between 'passionate Nats' and 'passing Nats' - since SNP voters in the Euros will be more likely to be in the 'passionate' camp?

    ReplyDelete
  9. I said it before and I'm repeating it.

    I have doubts about the polls, because they rely on the majority of their business from the Westminster parties - the NO camp. Independence or not, that will not change and there are no consequences to getting it wrong (as can be seen from the 2011 Holyrood results compared to the predictions).

    The pollsters are apparently still puzzled as to why the 2011 poll predictions were wrong. In the USA, the pollsters who get it wrong are shredded on air and in the newspapers, and their previous failings would be raised every time they come out with a new poll.

    Here we have an MSM in cahoots with Westminster and they hope their influence will save the day for the Unionists (which would keep the MSM and Westminster in a joined-at-the-hip position of power).

    ReplyDelete
  10. I think we should be aware that their methodologies are being stress tested to destruction this time. None of the polling houses will have a directly comparable example from the past by which we can judge their potential performance this time.

    Even with all the tricks, strange leading questions and partisan organisations preloading the outcome of polls, Yes is within touching distance. Let's be clear, even if these numbers are right, Yes only has to mobilise a few % more of their support on the day- if turnout is 80% or more, its not unreasonable to suggest that's possible (maybe even likely!).

    ReplyDelete
  11. "that Peter Kellner's attempts to browbeat his fellow pollsters into adopting a more No-friendly methodology have borne fruit"

    or maybe they realised a professional pollster, who runs a company that most closely predicted the recent euro elections, was right?

    And the rest of you, stop looking for ever flimsier reasons why no surely can't be winning. They are, trying to convince yourself they're not and YES are secretly in the lead, surely just stops the yes side from trying as hard as they should?

    If you think people lie to professional neutral pollsters, what do they say to someone rocking up at their door covered in "YES" badges? i know I've often agreed with door knockers and said "I'll think about it definitely" just to get them off my property ASAP.

    ReplyDelete
  12. James says, "or else a large number of people are saying they voted in May when they didn't. "
    For me, this is the crux of the whole poll debate;
    The methodologies employed are all guesswork and given there is no precedent for this referendum, they are even more unreliable.
    Lest we forget what is at stake here, the UK no less!
    It follows that ANY change in method to negate any real poll lead for YES will be employed, regardless of the client. Not surprising that Kellner is at the forefront, and other are following suit.
    The only poll that matters is the real one. The people of Scotland are not going to get fooled this time.These manipulations are being shared and the game is up. The pollsters will get it wrong and they will come up with a variety of lame excuses.

    ReplyDelete
  13. As it's a individual vote,I take any poll with a huge pinch of salt.

    I think it will be a shock result,but not sure which side

    ReplyDelete
  14. And the rest of you, stop looking for ever flimsier reasons why no surely can't be winning. They are, trying to convince yourself they're not

    Why do you sound so worried if No are really in the lead? I can almost see the sweat running down your brow. Can you direct us to some No canvassing returns? Surely there must be loads of these showing a stonking no win?

    Or maybe not because polls show <30% are set on voting No? When you have 40%+ planning Yes and the remainder undecided (inc those saying no but saying they are actually undecided as they may yet change their minds) that's rather worrying for the union...

    No are facing a disaster even if they win is the problem for the unionists.

    >70% No = matter settled
    60-69% No = matter possibly settled but what happens when there's no devo max?
    50-59% No = very bad for the union. Utter disaster. Jeez, 40% + of the electorate voting for complete independence. Man oh man. What an embarrassment for the UK. A win, but at the same time a massive loss. No devo max and we'll be looking at another referendum soon enough.
    50% + 1 vote for Yes = independence. All over.

    I'd be bricking too if I supported the union.

    As Darling has said before, anything but a big No win is a loss. A 40%+ Yes would be just like 2007; one more step on the 70+ year road back to independence as the remnants of the British empire continues its retreat back to where it started.

    How do save something that's been dying for 70 years? Britishness is literally dying as polls / the census shows us. The young today are the least British of any generation at >70% Scottish only while the older generations who support the union/feel british pass away. Such long-term changes are so difficult to stop. Nearly impossible. The referendum is just the next manifestation of a dying Britain. It had it's time and that time is ending. 1949 covenant, 1979, 1997, 2007, 2011. Each wave stronger than the last and closer to ending the UK. Maybe not in September, but the end is coming.

    ReplyDelete
  15. Anonymous No....?
    "And the rest of you, stop looking for ever flimsier reasons why no surely can't be winning. They are, trying to convince yourself they're not and YES are secretly in the lead,"

    YES is NOT 'secretly' in the lead. By all other methods apart from dodgy polls they are quite openly in the lead. And YouGov "closely predicted" the result of the Euro election? I DON'T THINK SO!

    Results in percentages:

    Actual SNP 29 LAB 25 Con 17 UKIP 10 Grn 8 LD 7

    YouGov SNP 26 LAB 28 CON 15 UKIP 13 GRN 11 LDEM 6.

    +10% error in LAB's YouGov result
    -12% error in SNP's YouGov result.

    More than enough to put YES in lead in Referendum, eg, Polls' results 47 Y 53 N go to 53 Y N 48 if same errors are used to correct poll results.

    ReplyDelete
  16. So is it safe to say then, that based on Panelbase changing to suit Kellner, that Yes will probably never be in the lead in the run up to the referendum?

    I think it's probably something to do with what happened in Quebec, did the Canadian dollar not collapse when the Quebec lot was in the lead?

    Unless Survation stick to their guns and put two fingers up at Kellner of course, but then BT will say it's an outlier and only one polling organisation.

    Yes SS, I agree entirely, but I am now coming round to what you've been saying all along, it will be over 55%......I was hearing as well from friends in the know that were speaking to a particular pollster, that they expect there to be 800,000 undecided people on the day of the referendum.

    The positive will really win and all their negativity and the memories of that will come back to the Scot's and they will vote against it and for Scotland. I've never actually been more confident of winning.

    This blog is tremendous and I hope more Yesser's read it and share it as this is all part of the Establishments efforts to keep down the Yes vote.


    ReplyDelete
  17. @anon "or maybe they realised a professional pollster, who runs a company that most closely predicted the recent euro elections, was right?"

    Eh? On the night of the Euro's Kellner was assuring everyone that Labour would win the overall vote in Scotland.

    He predicted a Labour win...Labour Lost!!!

    You say he was closer than anyone else, so every polling company was predicting a Labour victory?

    That's interesting if true, because it seems to suggest every single pollster is wrong about Scottish voting intentions, and every one of their polling results fails to pick up the popularity of the SNP.

    If however some did predict that the SNP would win, then I'm afraid you are spouting bullshit!

    So.. what one is it?

    Bullshit or just plain wrong?

    Glad to help :-)

    ReplyDelete
  18. James Coleman: "Actual SNP 29 LAB 25 Con 17 UKIP 10 Grn 8 LD 7

    YouGov SNP 26 LAB 28 CON 15 UKIP 13 GRN 11 LDEM 6.

    +10% error in LAB's YouGov result
    -12% error in SNP's YouGov result."

    The % error was 3% in both cases, not 10%, and within the margin of error. Trying to pretend the 3% is actually 10% by taking it as a percentage of a percentage (the 28% / 25%) is mathematical nonsense, but yet another incident of trying to massage the data.

    i know i'm widdling in the wind in this hive of groupthink, but people are telling you what they think you want to hear.

    http://en.wikipedia.org/wiki/List_of_cognitive_biases

    i just hope you accept this gracefully once the results are apparent come september 19th

    ReplyDelete
  19. Scottish skier writes

    the end is coming.


    I love it, you sound like an Old testament preacher. But that tone is something that bears discussion.,

    Firstly I would point out I have contributed before as Expat and admit I am No, but I still like the site very much and, if anything would persuade me, it's is the fact that James offers serious analysis, puts forward really interesting points and does NOT indulge in Old Testament triumphalism. Neither side should do that,

    So I can well understand that James is worried (as he mentioned once) by the spectre of Michael Foot. Foot, a good if sometimes misguided man, and many followers genuinely believed they would win the 1983 UK General Election because of the massive discrepancy between the exultation and sheer passion of their supporters (including the canvas returns) and the opinion polls. The polls were right and he lost. It is all too easy to be caught up in a cause and not see the overall picture.

    And how much more is that true here where one side yes is like a religious revivalist movement with all its joy and hope and emotional conversions and the other is subdued and settled and just wants it all over. To my amazement some people on this site talk of shy Yes, yet nobody has really supplied any reason why they should be shy beyond massively paranoid stuff that MI5 will come after them or something. It is the other side which wants a quiet life and prefers to keep its head down to avoid love bombing and patriotic exuberance (even if not abuse).

    And in this respect let me point out the utter nonsense of saying it is impossible for someone to turn from Yes to No. A few have written here they have never encountered or heard of such a thing.. But of course they haven't, any more than a Wesleyite evangelical in the midst of a massive religious craze would hear of a follower suddenly breaking with the faith. Such a thing would only happen utterly privately, nobody is going to advertise religious doubts to an excited and joyful mob? They just wouldn't.

    I believe Cameron was reckless to allow the question to be posed in such a way, indeed the Nats seem to have got everything they could possibly want from the timing and the question. The words could easily have been 'Do you want to remain a part of the United Kingdom' and then the (current) No camp would be happily beating a more a positive drum

    ReplyDelete
  20. Scottish skier writes

    the end is coming.


    I love it, you sound like an Old testament preacher. But that tone is something that bears discussion.,

    Firstly I would point out I have contributed before as Expat and admit I am No, but I still like the site very much and, if anything would persuade me, it's is the fact that James offers serious analysis, puts forward really interesting points and does NOT indulge in Old Testament triumphalism. Neither side should do that,

    So I can well understand that James is worried (as he mentioned once) by the spectre of Michael Foot. Foot, a good if sometimes misguided man, and many followers genuinely believed they would win the 1983 UK General Election because of the massive discrepancy between the exultation and sheer passion of their supporters (including the canvas returns) and the opinion polls. The polls were right and he lost. It is all too easy to be caught up in a cause and not see the overall picture.

    And how much more is that true here where one side yes is like a religious revivalist movement with all its joy and hope and emotional conversions and the other is subdued and settled and just wants it all over. To my amazement some people on this site talk of shy Yes, yet nobody has really supplied any reason why they should be shy beyond massively paranoid stuff that MI5 will come after them or something. It is the other side which wants a quiet life and prefers to keep its head down to avoid love bombing and patriotic exuberance (even if not abuse).

    And in this respect let me point out the utter nonsense of saying it is impossible for someone to turn from Yes to No. A few have written here they have never encountered or heard of such a thing.. But of course they haven't, any more than a Wesleyite evangelical in the midst of a massive religious craze would hear of a follower suddenly breaking with the faith. Such a thing would only happen utterly privately, nobody is going to advertise religious doubts to an excited and joyful mob? They just wouldn't.

    I believe Cameron was reckless to allow the question to be posed in such a way, indeed the Nats seem to have got everything they could possibly want from the timing and the question. The words could easily have been 'Do you want to remain a part of the United Kingdom' and then the (current) No camp would be happily beating a more a positive drum

    ReplyDelete
  21. Expat part 2

    So is it possible the sheer fervour will push yes over 50%? I think it is just, though you would need to see a shift very very soon in the next week or so (as was recorded fairly accurately by the polls in the Scottish 2007 election contrary to what so many say here) or time will have run out. Note also in that election, so often rehearsed here , polls as much as three months earlier had the Nats ahead for a time so a lead was no big deal. Very different from now.

    But a Yes win is, for many of us, a slightly terrifying scenario because it will all be emotion and no real hard thought. I think, as I would, it would be very tough indeed after that for everyone. Scotland would have to stand on its own, there would be no official pound (an unofficial one would be there for the taking I agree) and in any crisis it would be the IMF to help out, not the Bank of England. There is absolutely no bluff here and Yes should welcome it not duck it and say England is bluffing. . Overwhelmingly the polls in the rest of the UK show up to 80% rule out letting the Bank of England stand as Banker of last resort for Scotland (I.e. a joint currency), The rest of the UK won't give Scotland a free ride and why would you ask such a thing (I am sure the best Yes voters don't want it)

    But don't get me wrong I an not saying in ythe event if YES Scotland will go bust or even there will be a crisis, I am emphatically not an Old testament -style preacher of crisis. I think endless slog and some hardship for both nations would follow, both economies weakened, endless reorganisation, and endless cash having to be spent on reorganisation. All in the interests of an emotional idea which is creditable in a way but like all nationalism, can be so destructive.

    I am afraid I prefer living in a secular state.



    ReplyDelete
  22. Well said Expat. A dispassionate analysis of the facts. There is a distinct air of desperation coming from the Yes camp and the recent rise in comments on this blog merely confirms this.

    ReplyDelete
  23. The PsyOps squad is on its game today. :-)

    ReplyDelete
  24. Lots of new no voices appearing here, trying to convince us why the No-backed polls with all their experimental re-adjustments (which co-incidentally favour no) are right and why canvassing, social media and personal experience is all completely wrong... They must be getting very rattled. If you're so confident why would you care what we think?

    As for the theory that yes voters do move to no but just won't admit it due to fears of reactions from religious zealots, LOL, what complete drivel, is that the best you can come up with? Pure and utter fiction. If this was a real issue then everyone would be saying they vote yes, no one in the whole of Scotland would be admitting to no.

    I know plenty people voting no, I know no-one who has moved from yes to no. You are trying to tell us that loads of the population are confident about expressing their no intentions yet others, for no particular reason, are terrified? The answer is you're talking pish.

    The fact is that once people understand the arguments for voting yes (like their future will mean living in a social democracy instead of a neo-liberal dystopia), they don't change their minds, why would they? Vote no for foodbanks, inequality, vast fortunes frittered on WMDs, being governed by public school English aristocratic toffs? Nah, nae chance. No-one moves from yes to no because they know that they would be turkeys voting for Christmas.

    So dream on with your drivel, the fact you are on here trying to persuade us to believe this guff just proves to us that you are getting desperate.

    ReplyDelete
  25. Anon says: "There is a distinct air of desperation coming from the Yes camp and the recent rise in comments on this blog merely confirms this."

    Haha, the rise in comments is all you "Anonymous" no voters who have turned up here all of a sudden out of nowhere. Wonder why that is?

    ReplyDelete
  26. Anon - You'll find that many of these No voters you speak of, haven't engaged in the process yet, indeed, many won't until 2 weeks until the referendum, if that!

    Some won't even think about it until the week of the ref, they are default no, as was everyone until they looked into what an independent scotland will look like.

    I'd be extremely worried if I was you and try and lock away the 1 million people that haven't looked at both sides of the arguments.

    Squeaky bum time.

    ReplyDelete
  27. It is clear from the language of some of these anonymous no voters that they aren't even Scottish and don't have a vote, refering to Scotland as "you."

    Expat especially, doesn't have a clue what they are talking about; a yes vote would be the results of emotion and no hard thought? Christ, have we not moved on from this type of debate two years ago? And as for no currency union? Away back to the Scotsman with your scaremongery nonsense, we're more sophisticated than that on here.

    Why are you here again?

    ReplyDelete
  28. Anonymous....
    Was it the school of make believe mathematics you attended?

    Percentages are just a way of expressing a ratio and of course you can take %s of %s. Your example of 25%/28% which you claim is nonsense can be re-written as 25/100/28/100 which becomes 25/28 as the 100s cancel out.

    The % error to be applied to YouGov's results is in fact +-3% at 95% confidence level, so eg, the SNP result would likely lie somewhere between 29% and 23%.
    but I suppose in your universe you add +3% to the SNP result and deduct -3% from the Lab to come up with your desired claim that YouGov's results were very close to the actual result.

    Oh and for your and expat's information the following page

    https://twitter.com/JamesCo77979225/status/490489074021851137/photo/1

    shows a very clear massive surge to YES during the last 10 days of the 1997 Referendum. So we don't need to be ahead in the polls until then but I expect we will be. And I have shown a likely scenario for this Referendum beside it.

    ReplyDelete
  29. I see the "I'm a yes voter but" guys have disappeared now, to be replaced by the usual no voting doom-mongers. Or are they one and the same??

    And the pattern where they work in tandem and re-inforce each other is here too : "Well said, Expat..." Or maybe that's just the one person too.

    Is that you, John??

    Bunch of amateurs.

    ReplyDelete
  30. "or maybe they realised a professional pollster, who runs a company that most closely predicted the recent euro elections, was right?"

    I rarely agree with Mike Smithson, but I think he summed it up best when Kellner first published his diatribe - YouGov have got a brass neck talking about the inaccuracy of other pollsters when they themselves got it so badly wrong at both the Holyrood election and the AV referendum.

    "If you think people lie to professional neutral pollsters, what do they say to someone rocking up at their door covered in "YES" badges? i know I've often agreed with door knockers and said "I'll think about it definitely" just to get them off my property ASAP."

    That's a very interesting point, because we do have some limited information on what No canvassers are finding. Unless Better Together are lying to us, they found ridiculously high No votes when they were canvassing for the three most recent parliamentary by-elections - way out of line with the published polls. So, yes, people are lying to at least one set of canvassers (possibly both) on an industrial scale in this campaign, and I think the prudent lesson to learn is that it would be extremely foolish to assume that everyone is telling the God's truth to professional pollsters.

    You talk about professional "neutral" pollsters. Yes, most of them are neutral, but I'm not sure that can fully extend to YouGov anymore. They're now in a kind of no-man's-land somewhere in between neutrality and out-and-out bias. Both Peter Kellner and Laurence Janta-Lipinski have been using language very much in line with the No campaign's propaganda machine.

    ReplyDelete
  31. Just to clarify, I should have said Labour rather than Better Together when referring to canvassing at the by-elections.

    "i know i'm widdling in the wind in this hive of groupthink, but people are telling you what they think you want to hear."

    I think any fair-minded person reading both this thread and the previous one could only conclude that there is no such thing as groupthink on this blog. However, groupthink is certainly a danger for pollsters, and that's the problem with Kellner trying to spook his counterparts into line with YouGov, when the reality is that nobody has a clue which methodology is going to prove to be most accurate.

    ReplyDelete
  32. Expat : "So I can well understand that James is worried (as he mentioned once) by the spectre of Michael Foot."

    That's a complete misrepresentation of what I said. I was pointing out that the polling evidence (with the possible exception of the two most No-friendly pollsters) is clearly showing that Yes have a good chance of winning, and that it's therefore totally daft for Yes supporters to use the sort of language ("the only poll that matters is on September the 18th!") that Foot used when he knew he was heading for certain defeat.

    Anon : "There is a distinct air of desperation coming from the Yes camp and the recent rise in comments on this blog merely confirms this."

    If there is any recent increase in comments on this blog, as far as I can see it's mainly caused by the sudden eagerness of No supporters to post here, when they previously didn't bother. So by your logic, it must be the No side who are getting desperate?

    ReplyDelete
  33. sama anon here again (i should register shouldn't I?) ...What ARE you on about james?

    "Your example of 25%/28% which you claim is nonsense can be re-written as 25/100/28/100 which becomes 25/28 as the 100s cancel out."

    ??? That doesn't make sense in english, let alone maths.

    Yougov's predicted euro election was within the margin of error of the correct result (+ or -3%) for both snp and labour. They said snp 26, snp got 29. They said labour 28, labour got 25. The 3% error margin applies to the 100% total,

    There are no 10%'s or 12%'s in there.. think about it. If they said the BNP would get 1%, and the BNP got 0%, would that mean they were wrong by infinity per cent?

    Percentages of a percentage are not percentages themselves. End of.

    Now getting back to the referendum. I don't mind which side wins, frankly as someone who is a net tax contributor I'm worth more to scotland and whoever wins than a whole street of benefit claimants and the politics will be angled that way.

    PS To the other poster who was muttering about english tories and that, well done, with scotland being about 10% english-born and 17% voting tory you're putting off a lot of potential YES supporters aren't you?

    ReplyDelete
  34. I'm reminded of a tactically-run 10,000 metres race here. Coe and Ovett, maybe? One takes the lead at an early stage, and holds on to it, but can't shake the pursuer. The leader continually hopes to outdistance his rival by a significant margin, but every time he glances back he sees the other man matching him stride for stride, only a short way behind.

    The positions remain like this until both are coming off the bend into the finishing straight. At that point the runner in second place kicks for home, sprinting past the leader and getting to the tape in front.

    We KNOW Yes has finishing speed. 1997 and 2011 both prove that. I remember both election nights vividly. On both occasions I was on tenterhooks until the first result came in. Then look. The people had spoken, the darlings, and they'd produced a bloody landslide.

    In 1997 I didn't have any particular sense of doom, possibly because I was expecting at worst a narrow Yes to the first question even if a No to the second. Nevertheless the sheer scale of the victory completely took my breath away.

    In 2011 I had a very definite, niggling sense of doom in store if we lost. That doom was called SLAB, and the prospect of another five years of that lot and Iain Gray as FM was enough to bring me out in a cold sweat. I still believe a lot of that victory was down to people who took a good hard look at that prospect in the late stages of the campaign and recoiled in horror.

    As this campaign has progressed I have been increasingly haunted by an even more pervasive sense of doom in the event of a No vote. The more time has passed, the worse it looks. Food banks, ATOS benefit sanctions, 60% of the cuts still to come, England's NHS being privatised, fracking, Trident, Carmichael's threats to repress us so severely we'll never be able to break free again, it just goes on and on. That's why I get the jitters from time to time.

    I know it's irrational, but I also have a very strong sense of destiny surrounding this vote. Of history coming together to ensure that this, at last, is our time. I think Yes is building up for the same sort of finish as we saw in 1997 and 2011. But it's a hell of a nerve-racking ride in the mean time.

    ReplyDelete
  35. The no to Scotland campaign and their paid astroturfers have always been desperate liars when it comes to opinion polling.

    A favourite meme is to state that no poll has EVER had a majority for independence. I know it's a lie, you know it's a lie and they know it's a lie but up it pops again and again.

    Even if the rigged polls were telling the truth it would only take a couple of percent higher turnout for Yes to beat no. Hence all the "we've lost so don't bother voting" comments that have appeared en-masse.

    ReplyDelete
  36. "sama anon here again (i should register shouldn't I?)"

    You don't have to register - just select the "Name/URL" option and enter your name - you can leave the URL section blank.

    ReplyDelete
  37. "A favourite meme is to state that no poll has EVER had a majority for independence. I know it's a lie, you know it's a lie and they know it's a lie but up it pops again and again."

    Yes, that's one of my absolute favourites - in reality YouGov, ICM and TNS-BMRB have all published polls over the years showing a lead for independence. In the case of TNS, the most recent one was less than three years ago.

    ReplyDelete
  38. Responses are getting a bit wild and woolly here. First off I am NOT anonymous, I have been posting as Expat (and as a NO) on this site since I came across it . So the idea I suddenly popped up out of panic is as daft as the idea I am part of some sudden psyops conspiracy. For heaven's sake get back to arguing about facts and opinions, and stop abuse which makes you sound more panicked than I actually believe you are or should be. The site is good, the major posts thoughtful (if not I wouldn''t bother) and it deserves better frankly.

    Oh and just saw James's last post. I am sorry if I misrepresented what you said about Foot. It was only a passing reference and you semed to be saying it was not a path you wanted to go down (?). Meaning false confidence. I agree with your point about the polls in any case.

    ReplyDelete
  39. So Expat, are you an expat with or without a vote?

    No offence intended but I come across many No's on social media who don't actually have a vote.

    ReplyDelete
  40. James - any thoughts on the British Election Study poll results reported in the Guardian today?? Encouraging shift to yes but it would be good to have your views on the methodology.

    ReplyDelete
  41. Colin, I've added a new post on the BES.

    ReplyDelete
  42. This is the last time I will reply to you because you are clearly mathematically challenged and maybe even inclined a bit to illiteracy.

    "??? That doesn't make sense in english, let alone maths."

    25/100/28/100 is standard simple mathematical notation which can also be written (25/100)/(28/100). And even with your apparent lack of mathematical ability you should be able to work out that the divisions simplify to 25/28 as the 100s cancel out.

    The rest of your post like all your others is mainly long winded prattle.

    ReplyDelete
  43. James Coleman - The margin of error on these polls is typically ±3%, which doesn't refer to the a relative error of the ratio of the polled value and the true value at the election, instead it is an absolute error on the polled values.

    If we took your approach, it would be saying that somehow polling values get more accurate as their absolute value gets higher:

    78/75 = 4% relative error
    28/25 = 12 % relative error

    Which is obviously wrong.

    ReplyDelete
  44. TheAnonYouStartedArguingWithJuly 29, 2014 at 5:04 PM

    "25/100/28/100 is standard simple mathematical notation which can also be written (25/100)/(28/100). "

    no it can't, the brackets make it a totally different sum james, but who am i to say with my lack of mathematical ability?

    Anyways, that detail dealt with, the fact remains that trying to do a percentage of a percentage on a poll like that produces tosh, as can be seen by my infinity per cent example. Any response to that?

    Ps That's not my reply above, just someone else thinks you're wrong too

    ReplyDelete
  45. "The margin of error on these polls is typically ±3%, which doesn't refer to the a relative error of the ratio of the polled value and the true value at the election, instead it is an absolute error on the polled values."

    Note this is precision error, not accuracy.

    If I run the same poll / method twice on the same day using an identical (demographically) but different sample, the results should be within +/- 3%. They could be 20% out though on the true value.

    I could measure the boiling point of water as 970, 1000 and 1030 C with a very precise but totally inaccurate thermometer. My precision error would be +/-3%, my accuracy an order of magnitude out...

    This is why pollsters can get results well outside the MoE when compared with each other. We can't know which pollster is closer to the actual, absolute value, if any.

    ReplyDelete
  46. "But a Yes win is, for many of us, a slightly terrifying scenario because it will all be emotion and no real hard thought."

    Why terrifying? Do residents of all countries which have become independent walk around in terror?

    Doesn't seem that to me e.g. watching the commonwealth games.

    You are correct emotion is the primary reason for the existence of the majority of countries. Or at least national identity and the socio-economic / geographical reasons this develops are.

    The union is in trouble because Scots increasingly no longer feel British; brutishness peaked in those born in 1944 (2011 census) and has been in decline since. There are a myriad of historical and recent reasons for this, but in the end, that's what it comes down to.

    Scotland won't vote Yes (in September or at some point after) because it is Scottish and not British. That's not something that most people will consciously think about. However, they wouldn't be voting Yes / giving indy serious thought if they were not Scottish / did not see Scotland as their country ahead of Britain. That applies to English Scots, French Scots (Mrs SS), Pakistani Scots....

    It's just how these things work.

    ReplyDelete
  47. "But a Yes win is, for many of us, a slightly terrifying scenario because it will all be emotion and no real hard thought."

    Why terrifying? Do residents of all countries which have become independent walk around in terror?

    Doesn't seem that to me e.g. watching the commonwealth games.

    You are correct emotion is the primary reason for the existence of the majority of countries. Or at least national identity and the socio-economic / geographical reasons this develops are.

    The union is in trouble because Scots increasingly no longer feel British; brutishness peaked in those born in 1944 (2011 census) and has been in decline since. There are a myriad of historical and recent reasons for this, but in the end, that's what it comes down to.

    Scotland won't vote Yes (in September or at some point after) because it is Scottish and not British. That's not something that most people will consciously think about. However, they wouldn't be voting Yes / giving indy serious thought if they were not Scottish / did not see Scotland as their country ahead of Britain. That applies to English Scots, French Scots (Mrs SS), Pakistani Scots....

    It's just how these things work.

    ReplyDelete
  48. "They are, trying to convince yourself they're not and YES are secretly in the lead, surely just stops the yes side from trying as hard as they should?"

    It's motivating to believe that you're winning. It's disillusioning to believe that you're consistently behind and that that isn't changing. So, no, falsely saying that Yes is ahead despite the polls probably doesn't suppress effort from Yes campaigners.

    ReplyDelete
  49. "25/100/28/100 is standard simple mathematical notation which can also be written (25/100)/(28/100)."

    Nope.

    ReplyDelete
  50. Actually (25/100)/(28/100) is exactly the same as 25/100/28/100 since these operations are divisions and in normal integer or real arithmetic the order doesn't matter. The brackets would only matter if addition, subtraction, modulus it powers were involved.

    As a maths geek,I just had to chip in ;-)

    ReplyDelete
  51. wHAT??? Are you all on crack?

    25/100 is a quarter

    28/100 is near enough another quarter

    Divide a quarter by a quarter and you have near enough one.

    25 divided by 100 then divided by 28 then divided by 100 gets smaller and smaller, try it on a calculator

    The brackets make a difference!


    See, in a way, this is why I'd just agree with a Yes Sir! who came to the door. It's easier to avoid the long argument and let them go off and think they've got you round to their way of thinking... so in that spirit, yes you're right, the brackets make no difference, and I'll definitely think about it

    ReplyDelete