Believe it or not, I've still been receiving the odd comment on the gun control posts I wrote before Christmas. This rather extraordinary one was left by Epsilon Given a day or two ago, although to begin with you might be hard-pressed to spot any connection with the subject in hand...
"Just for the record, I'd like to clarify just what "counter-intuitive" means to a mathematician.
Intuition is a funny thing. It generally leads us to truth, and a good intuition helps us to discover new theorems, even entirely new lines of thought.
When something we expect to be false, or nonsensical even, turns out to be true, we call such things "counter-intuitive". As we encounter these things, we sometimes adopt them into our intuition--but there is no shortage of things to surprise us!
Historically, we have encountered many counter-intuitive ideas--ideas that often met fierce resistance every step of the way to acceptance. The square root of -1, for example. Or a number for "nothing". Or the irrationality of the square root of 2.
And sometimes, proving something counter-intuitive can lead to your death. The person who first proved that the square root of 2 was irrational, for example, was drowned at sea by his fellow Pythagorians.
I'll give several more counter-intuitive examples that I'd expect to go over your head without deep explanation, unless you are already acquainted with them. You can cut up a unit sphere into nine pieces, and re-assemble them to make two unit spheres (Banach-Tarski Paradox). You can have two intersecting lines both parallel to a third line (hyperbolic geometry). There are more irrational numbers in the interval between zero and one, than there are integers (Georg Cantor's diagonal argument). When we "generalize" the factorial function, the factorial of 1/2 is 1/pi (Gamma function). e^(i*pi)+1=0 (Euler's Formula).
Heck, I spent many hours trying to trisect an angle using a compass and straightedge, and I spent hours trying to integrate e^(-x^2) in terms of "elementary functions", because I was told that these were impossible--and this conflicted with what I was taught in our culture, that "nothing is impossible". Later, I actually read the proof that demonstrated, once and for all, that the former is impossible, and more recently I learned that the latter proof is done along the same lines, so to me, these are no longer counter-intuitive.
The idea "More guns, less crime" is, like all these other things, counter-intuitive. Our intuition tells us that getting rid of guns will decrease violence. But we can find statistics that show when we ban guns, violence increases. And we can find statistically that, internationally, there is no correlation between gun laws, gun ownership, the murder rate, and the suicide rate, from country to country. As I have examined these things closely, I've come to the conclusion that honest people with guns is a good thing, because they can defend their lives against dishonest people determined to do harm.
I have seen the statistics and the studies. I have thought about the philosophies behind the two positions. In all this, I have come to the conclusion that more guns really do mean less crime! And that, more importantly, if you value liberty and security, then you shouldn't depend on the State for protection--it is your duty to protect yourself and those you love--and nothing can change that. Indeed, the State, more often than not, cannot protect you--and occasionally, the State is even your enemy.
And that those who believe otherwise, might as well believe that the moon is made of cheese, or that imaginary numbers really are imaginary, for all the truth that their position holds."
Words fail me. Epsilon, this is not the first time that I've had to make the elementary point to you that I do - honestly, truly - know what "counter-intuitive" means, and have done all along. Your apparent belief that if I take issue with anything you say it must mean that I've simply misunderstood the terminology you use (and require a detailed explanation to bring me up to speed) is...well, a touch bemusing, to say the least. I'll simply reiterate what I said to you before - some things that are counter-intuitive turn out to be true. Many don't. I note that you confirm that at the outset yourself, so your point is...?
The latter part of your comment essentially boils down to this - you've generously had a little think about the statistical evidence on our behalf, you've selectively culled the bits that suit you, you've declared your side of the argument the winner on that basis, and thus regard yourself entitled to feel morally superior to those of us who don't "do our duty" by "protecting" ourselves with a gun.
Well, let me put this to you. Here is the compelling evidence (as opposed to intuition) that those of us on the other side of the argument in Scotland look towards. It suggests overwhelmingly that our approach has succeeded in protecting the public. Might I suggest therefore that we are equally entitled to think - valuing security and liberty as we do - that it would be a disgraceful dereliction of our collective "duty" if we were to suddenly decide to indulge the wishes of a minority to own handguns, and thus compromise the safety and liberty of everyone?
Saturday, March 5, 2011
Which was less likely - Governor George W Bush of Texas granting clemency to a death row prisoner in 1999, or the unionist parties giving the people a say on their constitutional future in 2007 or 2010?
You have to hand it to the more mindless critics of the SNP and Plaid Cymru - if nothing else, they have their tedious little repertoire well-drilled. I popped over to Political Betting yesterday afternoon simply to express the unwelcome view that the Welsh referendum result was a "fantastic day for Wales", and yet another four-hour marathon ensued. First of all I was asked if I was even being serious - surely my comment could only be intended as sarcasm? No, I explained, it was indeed a fantastic day for Wales, as it was hard to think of a single good reason why the Welsh people weren't capable of running their own affairs on exactly the same basis as people in Scotland or Northern Ireland. Aha, came the predictable retort, it was absolutely fine if those countries governed themselves, just so long as they paid for it themselves. "As an English taxpayer," one commenter added, "I'm sick of subsidising the Scots". I told him in that case he should rejoice - because he doesn't. I pointed him in the direction of Professor Andrew Hughes Hallet's analysis -
"“The usual perception is that Scotland spends about 20% on public services more per head than the UK average...
“Those numbers are very misleading mainly because the spending in that part is what’s spent on behalf of Scotland but not necessarily in Scotland.
“The estimate for Scotland’s share, that’s contributions to defence, is 2.8 billion but it’s roughly 2.0 billion are actually paid out in Scotland
“So there’s an implicit subsidy going south in that sense and you can think of lots of other examples ...”
Hughes Hallett added:
“At the moment, on the current account, there’s a subsidy going to London, which is helping London.
“When you get down to it, on the current account for the last five years at least, maybe longer, Scotland has had a current account surplus, which is currently according to the national accounts in Scotland £1.3 billion.”
Asked whether Scotland would definitely be better off, Prof Hallett replied: “You can definitely say that it [Scotland] would be better off in terms of the revenue.”
Prof Hughes Hallett pointed to ‘missing’ income that is generated in Scotland but is actually attributed to London, giving the Crown Estate as an example saying: “The Crown agents who take fees for electricity generation and give it to the Treasury...”
Professor Hughes Hallett also destroyed one of the myths surrounding the bail out of HBOS and RBS claiming that their dealings in England would have meant that England would have shouldered a significant part of their liabilities."
Well, naturally the Nat-bashing hordes weren't best pleased about having one of their most cherished articles of faith totally demolished by a renowned economist, so after a series of fairly pathetic attempts to dismiss Hughes Hallett as a "no-name economist from a second-rate US university", we then moved on to the next phase of the standard repertoire - random (and rather desperate) muck-flinging. Yep, you've guessed it, it was the familiar heady blend of wild and long-since-disproved assertions that the SNP 'did a deal' over freeing Megrahi, and suggestions that they had bottled it on the independence referendum (or "acted dishonourably", as one commenter sniffed airily) by not bothering to spend an afternoon going through the motions of putting something to the vote in the Scottish Parliament that everyone knew was going to be defeated by 78 votes to 50, because the three unionist parties were hell-bent on voting it down.
It really is quite comical. If the SNP had taken the opposite course, we all know what the mantra would be by now - Alex Salmond would have "wasted parliamentary time and money on something no-one gives two hoots about" (the last bit is © Tavish "Two Hoots" Scott). As it is, they synthetically claim to be outraged that the SNP "weren't even trying to deliver independence". What are the SNP for, they plaintively cry.
That line of argument is, I'd suggest, the rough equivalent of claiming that a death row prisoner in Texas in 1999 wasn't really "trying to stay alive" because he dispensed with that all-important last-minute plea for clemency to Governor George W Bush.
An exotic variant on the traditional line came from PB's deputy editor David Herdson, who insisted that the SNP had 'squandered their golden chance' to call a referendum immediately after being elected in 2007. If they had tabled a parliamentary vote at that stage, he earnestly claimed, they would have got it through on the basis that the opposition parties would have recognised that the SNP had "won the election" and thus had the moral right to do it.
I believe the phrase "aye, right" was invented for moments like this...
"“The usual perception is that Scotland spends about 20% on public services more per head than the UK average...
“Those numbers are very misleading mainly because the spending in that part is what’s spent on behalf of Scotland but not necessarily in Scotland.
“The estimate for Scotland’s share, that’s contributions to defence, is 2.8 billion but it’s roughly 2.0 billion are actually paid out in Scotland
“So there’s an implicit subsidy going south in that sense and you can think of lots of other examples ...”
Hughes Hallett added:
“At the moment, on the current account, there’s a subsidy going to London, which is helping London.
“When you get down to it, on the current account for the last five years at least, maybe longer, Scotland has had a current account surplus, which is currently according to the national accounts in Scotland £1.3 billion.”
Asked whether Scotland would definitely be better off, Prof Hallett replied: “You can definitely say that it [Scotland] would be better off in terms of the revenue.”
Prof Hughes Hallett pointed to ‘missing’ income that is generated in Scotland but is actually attributed to London, giving the Crown Estate as an example saying: “The Crown agents who take fees for electricity generation and give it to the Treasury...”
Professor Hughes Hallett also destroyed one of the myths surrounding the bail out of HBOS and RBS claiming that their dealings in England would have meant that England would have shouldered a significant part of their liabilities."
Well, naturally the Nat-bashing hordes weren't best pleased about having one of their most cherished articles of faith totally demolished by a renowned economist, so after a series of fairly pathetic attempts to dismiss Hughes Hallett as a "no-name economist from a second-rate US university", we then moved on to the next phase of the standard repertoire - random (and rather desperate) muck-flinging. Yep, you've guessed it, it was the familiar heady blend of wild and long-since-disproved assertions that the SNP 'did a deal' over freeing Megrahi, and suggestions that they had bottled it on the independence referendum (or "acted dishonourably", as one commenter sniffed airily) by not bothering to spend an afternoon going through the motions of putting something to the vote in the Scottish Parliament that everyone knew was going to be defeated by 78 votes to 50, because the three unionist parties were hell-bent on voting it down.
It really is quite comical. If the SNP had taken the opposite course, we all know what the mantra would be by now - Alex Salmond would have "wasted parliamentary time and money on something no-one gives two hoots about" (the last bit is © Tavish "Two Hoots" Scott). As it is, they synthetically claim to be outraged that the SNP "weren't even trying to deliver independence". What are the SNP for, they plaintively cry.
That line of argument is, I'd suggest, the rough equivalent of claiming that a death row prisoner in Texas in 1999 wasn't really "trying to stay alive" because he dispensed with that all-important last-minute plea for clemency to Governor George W Bush.
An exotic variant on the traditional line came from PB's deputy editor David Herdson, who insisted that the SNP had 'squandered their golden chance' to call a referendum immediately after being elected in 2007. If they had tabled a parliamentary vote at that stage, he earnestly claimed, they would have got it through on the basis that the opposition parties would have recognised that the SNP had "won the election" and thus had the moral right to do it.
I believe the phrase "aye, right" was invented for moments like this...
Friday, March 4, 2011
An election that might tell us a little something
At the start of the year, I noted that the Oldham East and Saddleworth by-election was a contest that told us almost nothing about the parties' prospects for Holyrood, partly because the circumstances were so unusual, but mainly because Scotland has of late seemed totally cut adrift from UK-wide trends as relating to Labour and the Tories. However, the Lib Dems might just be a different matter, and their utter humiliation in Barnsley Central this evening offers further reason to suspect that their Westminster alliance with the Tories stands to cost them a number of Scottish Parliament seats in May.
As I've noted before, a poor performance for the Lib Dems is ultimately bad news for Labour. Not only might it harm Iain "the Snarl" Gray's chances of becoming First Minister, it also reduces the likelihood of any government he leads being stable. If he finds himself forced by the arithmetic to look beyond the Lib Dem (and possibly Galloway one-man-band) ranks, his options for reliable allies over the four - or five? - years ahead will be distinctly limited.
UPDATE : Stuart Dickson alerted me to this little gem. Note the commendable dedication to accuracy in the visual representation of a general election result in Barnsley Central that had Labour on 47.3% of the vote, and the Lib Dems on 17.3%. Note also the refreshingly open acknowledgement that there was a virtual dead heat for second place, with the Tories just six votes behind the Lib Dems. Those of us who support lesser parties can but feel humbled in the face of yet another example of the Lib Dems' legendary honesty and moral rectitude.
Which begs only one question - do they have a bar-chart template for elections in which they're starting from sixth place? I'm sure they must have all eventualities covered...
As I've noted before, a poor performance for the Lib Dems is ultimately bad news for Labour. Not only might it harm Iain "the Snarl" Gray's chances of becoming First Minister, it also reduces the likelihood of any government he leads being stable. If he finds himself forced by the arithmetic to look beyond the Lib Dem (and possibly Galloway one-man-band) ranks, his options for reliable allies over the four - or five? - years ahead will be distinctly limited.
UPDATE : Stuart Dickson alerted me to this little gem. Note the commendable dedication to accuracy in the visual representation of a general election result in Barnsley Central that had Labour on 47.3% of the vote, and the Lib Dems on 17.3%. Note also the refreshingly open acknowledgement that there was a virtual dead heat for second place, with the Tories just six votes behind the Lib Dems. Those of us who support lesser parties can but feel humbled in the face of yet another example of the Lib Dems' legendary honesty and moral rectitude.
Which begs only one question - do they have a bar-chart template for elections in which they're starting from sixth place? I'm sure they must have all eventualities covered...
Thursday, March 3, 2011
Decision day in Wales
It's great to see Malc at Better Nation devote a post to the referendum in Wales today - it's received so little coverage in the "national" mainstream media that you'd be forgiven for not being aware that it's even taking place. On one point I'd quibble with Malc, though -
"To be clear, this isn’t a referendum to extend devolution or bestow more powers on the National Assembly for Wales. They already have the opportunity to get the powers which will be delivered in the event of a Yes vote in today’s referendum – they were bestowed on the NAW by the Government of Wales Act (2008).
This is more about speed of delivery – rather than having to apply to Westminster for individual powers in each of 20 fields specified in the Act using a lengthy process known as Legislative Competence Orders (LCOs)..."
To my mind, the vote plainly is about the extension of devolved powers. The whole point of having to apply for a power, surely, is that the application can be refused - or to put it another way, the power isn't actually 'bestowed' on the Assembly until an application is accepted.
Malc also reveals the entire lengthy preamble to today's referendum question, which includes this rather startling sentence -
"The Assembly cannot make laws on subject areas such as defence, tax or welfare benefits, whatever the result of this vote."
Which is as good as saying "you don't get to choose on such grown-up matters, so suck it up". Doubtless, though, there'll still be somebody brazen enough to claim a Yes vote today as a "vote against nationalism", just as Roy Hattersley did after the Scottish devolution referendum of 1997. Tell me, Roy - how exactly could I have gone about voting for independence in that referendum? A No vote? An abstention? How?
Still, in a way it's a pity that the AV referendum question won't be fronting up to its limitations in quite such a direct manner...
"At present, our voting system is not proportional. However you vote today, that will not change. Tough luck, baby."
"To be clear, this isn’t a referendum to extend devolution or bestow more powers on the National Assembly for Wales. They already have the opportunity to get the powers which will be delivered in the event of a Yes vote in today’s referendum – they were bestowed on the NAW by the Government of Wales Act (2008).
This is more about speed of delivery – rather than having to apply to Westminster for individual powers in each of 20 fields specified in the Act using a lengthy process known as Legislative Competence Orders (LCOs)..."
To my mind, the vote plainly is about the extension of devolved powers. The whole point of having to apply for a power, surely, is that the application can be refused - or to put it another way, the power isn't actually 'bestowed' on the Assembly until an application is accepted.
Malc also reveals the entire lengthy preamble to today's referendum question, which includes this rather startling sentence -
"The Assembly cannot make laws on subject areas such as defence, tax or welfare benefits, whatever the result of this vote."
Which is as good as saying "you don't get to choose on such grown-up matters, so suck it up". Doubtless, though, there'll still be somebody brazen enough to claim a Yes vote today as a "vote against nationalism", just as Roy Hattersley did after the Scottish devolution referendum of 1997. Tell me, Roy - how exactly could I have gone about voting for independence in that referendum? A No vote? An abstention? How?
Still, in a way it's a pity that the AV referendum question won't be fronting up to its limitations in quite such a direct manner...
"At present, our voting system is not proportional. However you vote today, that will not change. Tough luck, baby."
Stranger things are starting to begin
Perhaps this is the time of year that I always think this, but I'm beginning to seriously despair of the 2011 Eurovision crop. Of the songs I've heard from the national selections so far, a mere three have really reached out and grabbed me (although there may be some good ones I haven't caught up with yet), and of those, two failed to win the ticket to Düsseldorf. Admittedly there were special circumstances - a sympathy vote following the death of another song's original performer - that partly explain Jóhanna's failure in Iceland, but the overlooking of Nicki Ponte's I Don't Wanna Dance in this evening's Greek final is utterly beyond me.
The one gem that has made it through comes from that unlikeliest of sources - the hosts Germany. Much as I never found Lena Meyer-Landrut half as irritating last year as some people seemed to, I was still slightly dubious about the wisdom of allowing her to 'defend her title'. However, with a brave choice of song that once again suits her quirkiness down to the ground, it looks like it might just pay off - at least to some extent. Taken By a Stranger is probably a bit too low-key to win outright, but the way things are going it may well deserve to.
One thing that occurred to me when I was listening to it the first time round was that, just like Satellite, the lyrics seemed a bit too offbeat and intricate to have been written by a non-native speaker of English. Sure enough, they appear to have been penned by an American. That's surely one of the biggest downsides of the scrapping of the language restrictions twelve years ago - now that Germany have established it can be an advantage, we'll probably see more and more songwriters from English-speaking countries being drafted in, narrowing the contest's 'gene pool' considerably.
The one gem that has made it through comes from that unlikeliest of sources - the hosts Germany. Much as I never found Lena Meyer-Landrut half as irritating last year as some people seemed to, I was still slightly dubious about the wisdom of allowing her to 'defend her title'. However, with a brave choice of song that once again suits her quirkiness down to the ground, it looks like it might just pay off - at least to some extent. Taken By a Stranger is probably a bit too low-key to win outright, but the way things are going it may well deserve to.
One thing that occurred to me when I was listening to it the first time round was that, just like Satellite, the lyrics seemed a bit too offbeat and intricate to have been written by a non-native speaker of English. Sure enough, they appear to have been penned by an American. That's surely one of the biggest downsides of the scrapping of the language restrictions twelve years ago - now that Germany have established it can be an advantage, we'll probably see more and more songwriters from English-speaking countries being drafted in, narrowing the contest's 'gene pool' considerably.
Wednesday, March 2, 2011
Are we really supposed to "worry" about the prospect of less discrimination?
Ruth Sunderland on the European ruling banning discrimination by insurance companies on the basis of gender -
"Lord Davies is quite rightly trying to tackle prejudice based on sex. Presumably, the ECJ thinks it is doing the same thing - but it has confused unfair discrimination with the perfectly reasonable act of making well-founded distinctions between groups of people."
I can't help wondering if Ms Sutherland is quite up to speed with the actual definition of the word "prejudice". If judging a person on the basis not of their individual traits but instead of the traits of a "group" they are simplistically assigned to doesn't constitute prejudice, I don't know what does. Are we supposed to regard the greater likelihood of black youths being randomly stopped by the police as "perfectly reasonable" on the grounds that "well-founded distinctions between groups of people" are being made?
"The big worry is that this may only be the start. Lawyers say Europe may turn its sights on other areas such as age, where there is already an EU directive against discriminatory pricing that currently gives an exemption for insurers."
Now, I "worry" about a great many things, but I must say the prospect of further blows against discrimination on the basis of age, race, gender or disability doesn't rank high on my list.
"Lord Davies is quite rightly trying to tackle prejudice based on sex. Presumably, the ECJ thinks it is doing the same thing - but it has confused unfair discrimination with the perfectly reasonable act of making well-founded distinctions between groups of people."
I can't help wondering if Ms Sutherland is quite up to speed with the actual definition of the word "prejudice". If judging a person on the basis not of their individual traits but instead of the traits of a "group" they are simplistically assigned to doesn't constitute prejudice, I don't know what does. Are we supposed to regard the greater likelihood of black youths being randomly stopped by the police as "perfectly reasonable" on the grounds that "well-founded distinctions between groups of people" are being made?
"The big worry is that this may only be the start. Lawyers say Europe may turn its sights on other areas such as age, where there is already an EU directive against discriminatory pricing that currently gives an exemption for insurers."
Now, I "worry" about a great many things, but I must say the prospect of further blows against discrimination on the basis of age, race, gender or disability doesn't rank high on my list.
A few more thoughts on that YouGov poll
I've just been having a belated look at the details of the controversial YouGov poll from the weekend, and there's no great mystery about the two factors that led to a double-digit SNP lead on the raw data being transformed into a handsome Labour lead in the headline figures. The original sample already included more Labour identifiers than SNP identifiers - but the difference was just two (291 Labour to 289 SNP). That was reweighted to become a whopping 478 to 201 advantage for Labour. And on newspaper readership, another crucial aspect of YouGov's methodology, the weighting of the 94 Record/Mirror readers in the actual sample was almost tripled so that they became 251 'virtual' people - who were disproportionately likely to vote Labour, naturally.
In one sense, these figures reinforce the obvious need for weighting - it seems highly likely that Labour do have a greater advantage over the SNP in terms of party identification than the trivial one suggested by the original sample, and it's clear enough that tabloid readers were underrepresented relative to readers of quality papers. But the real issue is where the target figures for the weightings come from, and whether they have any objective basis. YouGov have form on this - after being subjected to persistent criticism in the run-up to last year's Westminster election, they suddenly conceded that their previous methodology had been failing to take account of changes in party ID in Scotland following the SNP's victory in 2007, and that the Nationalists would be "weighted up" by about 2% in subsequent polls. Encouraging though it was to see their willingness to accept an error, it was hard to escape the impression that both the old and the new weightings had to some extent been plucked out of the air on the basis of what 'looks right'. I'd suggest the next lesson they need to learn is that a one-dimensional party ID model simply isn't appropriate for a country with such an entrenched pattern of diverging party preferences for devolved and Westminster elections. If weighting by party allegiance is considered essential due to the difficulties thrown up by YouGov's panel method, a better bet might be to look at how people actually voted in the previous election for the relevant institution, with adjustments to take account of the 'spiral of silence' problem and false recall.
Regardless of what you think of the massive inbuilt 'head-start' awarded to Labour in YouGov polls for Holyrood, one other observation ought to be relatively uncontroversial - the greater the need to reweight certain portions of the sample, the more unreliable the poll results become, even if the weighting system is well-founded. For example, because barely a third of the target number of Record/Mirror readers were interviewed, there's a greater statistical chance that the voting preferences they reported will be unrepresentative of that section of the population, and of course any such distortion will be magnified when the figures are scaled up. The need for such extreme weighting is an inescapable weakness of YouGov's panel method - and yes, James MacKenzie, that remains a valid point no matter how many times the SNP have commissioned YouGov polls in the past.
My own gut feeling - and I said this at Political Betting even after the Ipsos-Mori poll showing the SNP ahead - is that Labour probably do have some kind of lead at the moment, but that the mammoth advantage suggested by the TNS-BMRB poll at the start of the year seemed highly implausible. The same applies to Labour's lead on the list in this poll. Either way, what we all need to urgently remind ourselves of is that this is a PR election, and that the First Minister will be chosen by a vote of all members of the Scottish Parliament, not by a plurality of one party over another in terms of the popular vote or seats. In many ways the most important figure is the combined support for the SNP and Conservatives - not because those two parties are remotely likely to agree a coalition deal, but because the Conservatives are overwhelmingly unlikely to enter into coalition (or indeed any sort of formal deal) with Labour. Even if we assume this poll is accurate, Labour and the Lib Dems in combination have a lead of just 2% on the constituency vote over the SNP and Tories in combination, and 6% on the list. That may put the state of play in somewhat better perspective.
In one sense, these figures reinforce the obvious need for weighting - it seems highly likely that Labour do have a greater advantage over the SNP in terms of party identification than the trivial one suggested by the original sample, and it's clear enough that tabloid readers were underrepresented relative to readers of quality papers. But the real issue is where the target figures for the weightings come from, and whether they have any objective basis. YouGov have form on this - after being subjected to persistent criticism in the run-up to last year's Westminster election, they suddenly conceded that their previous methodology had been failing to take account of changes in party ID in Scotland following the SNP's victory in 2007, and that the Nationalists would be "weighted up" by about 2% in subsequent polls. Encouraging though it was to see their willingness to accept an error, it was hard to escape the impression that both the old and the new weightings had to some extent been plucked out of the air on the basis of what 'looks right'. I'd suggest the next lesson they need to learn is that a one-dimensional party ID model simply isn't appropriate for a country with such an entrenched pattern of diverging party preferences for devolved and Westminster elections. If weighting by party allegiance is considered essential due to the difficulties thrown up by YouGov's panel method, a better bet might be to look at how people actually voted in the previous election for the relevant institution, with adjustments to take account of the 'spiral of silence' problem and false recall.
Regardless of what you think of the massive inbuilt 'head-start' awarded to Labour in YouGov polls for Holyrood, one other observation ought to be relatively uncontroversial - the greater the need to reweight certain portions of the sample, the more unreliable the poll results become, even if the weighting system is well-founded. For example, because barely a third of the target number of Record/Mirror readers were interviewed, there's a greater statistical chance that the voting preferences they reported will be unrepresentative of that section of the population, and of course any such distortion will be magnified when the figures are scaled up. The need for such extreme weighting is an inescapable weakness of YouGov's panel method - and yes, James MacKenzie, that remains a valid point no matter how many times the SNP have commissioned YouGov polls in the past.
My own gut feeling - and I said this at Political Betting even after the Ipsos-Mori poll showing the SNP ahead - is that Labour probably do have some kind of lead at the moment, but that the mammoth advantage suggested by the TNS-BMRB poll at the start of the year seemed highly implausible. The same applies to Labour's lead on the list in this poll. Either way, what we all need to urgently remind ourselves of is that this is a PR election, and that the First Minister will be chosen by a vote of all members of the Scottish Parliament, not by a plurality of one party over another in terms of the popular vote or seats. In many ways the most important figure is the combined support for the SNP and Conservatives - not because those two parties are remotely likely to agree a coalition deal, but because the Conservatives are overwhelmingly unlikely to enter into coalition (or indeed any sort of formal deal) with Labour. Even if we assume this poll is accurate, Labour and the Lib Dems in combination have a lead of just 2% on the constituency vote over the SNP and Tories in combination, and 6% on the list. That may put the state of play in somewhat better perspective.
Tuesday, March 1, 2011
F - P - T - P - Can - We - Fix - It? F - P - T - P - Yes - We - Can!
I'm just back from visiting ConHome, in search of my nightly fix of daft defences of our glorious first-past-the-post voting system. Yet again, I wasn't disappointed. At first I thought Ed Hall was seriously trying to compare those who point out the unfairness of FPTP with a tearful three-year-old stamping his feet and demanding that he be allowed to watch Bob the Builder again, but no, it's even better than that - the comparison is actually with Hall himself, and how he feels about the 'unfairness' of electoral reform. A few unkind souls might suggest that's rather revealing.
Pointing to the example of the forthcoming London mayoral election, which will use a system similar to AV, Hall claims -
"If the leading candidate doesn't get 50% of the first preference votes, then London's Mayor will be chosen by the second choice votes of the electors that voted for the political wings of fundamentalist Christian, Karmic flying and neo-Nazi parties. That is not just unfair, it's plainly irrational. It gives undue weight to the voters who chose to use the election as a platform to promote their fringe views."
No, it doesn't, Ed. It gives precisely the same weight to every single voter, which is how - in my naivety - I always thought a democratic system was supposed to work. But I can certainly understand Hall's confusion, given that his only previous experience is with a system that saw Russell Johnston elected MP for Inverness in 1992 on 26% of the vote. To use Hall's terminology, first-past-the-post gave 100% "weight" to the one-quarter of people who happened to vote for Johnston, and literally zero "weight" to the three-quarters of people who voted against him. I dare say some of the latter group were sorely tempted to go off and indulge in a bit of foot-stamping afterwards - after all, in those days they didn't even have episodes of Bob the Builder to distract them from the injustice.
Whether Hall realises it or not, what he's actually arguing against is not the idea that fascist-inclined voters and "Karmic fliers" should have a greater say than the rest of us, but rather the idea that they should even have an equal say. That may be superficially attractive to some, but it's an argument against universal suffrage, not AV. Hall doesn't seem to have spotted that under FPTP, fascist voters are already able to decide the outcome of elections, either because they've switched tactically to a mainstream candidate, or because there is no BNP candidate and they consequently have no choice but to vote for a mainstream party. If he finds that so distasteful and "unfair", I presume Hall will be proposing fundamental changes to the current system to - somehow - ensure it can never happen again in future? Mysteriously, it seems not.
But if you think Hall's argument can't get any weaker, you're underestimating the man. He goes on to compare the self-evident 'fairness' of the current system with the way TV talent shows work -
"In modern politics, a winner should be a winner. Try it round your dinner table or next time you watch the X Factor or Strictly Come Dancing. Everyone votes for their favourite book, film or act: surely the candidate with the most votes wins? How would the BBC or ITV possibly explain or justify a programme format with public voting in which the candidate that got the most votes did not win?
'Thank you for calling Britain’s Got Talent. Your vote for the Trapeze Sisters has been counted. Now you can choose a second choice contestant instead. If The Trapeze Sisters don't win, your vote will be transferred to your second choice contestant. Press 2 to register a second preference vote.'"
Oh dear. I'm afraid, Ed, that the voting system for Britain's Got Talent and Strictly Come Dancing is much closer to AV than first-past-the-post. If it worked like the latter, Strictly wouldn't have run for ten or fifteen weeks or whatever it was - there would have been just one programme and one set of dances, at the end of which Ann Widdecombe would have been declared the series winner, on the grounds that her 14% of the popular vote was fractionally higher than any other individual. Your average dinner-party would deem that much fairer than what actually happened, apparently.
Comically, Hall seems to spot this gaping hole in his reasoning almost before he finishes uttering it -
"Of course you do get to vote again in the TV formats as the candidates are knocked out, and the next week's programme starts, but do we really want General Elections every week until we get a winner? That would be the only way to give equal weight to everybody’s second choice votes."
No, it wouldn't, Ed. There's another ingenious way of achieving just that, and you'll be thrilled to hear we're about to hold a referendum on it. It's called AV, or to use its US name, Instant Run-off Voting. The Americans call it that because it produces much the same effect as successively eliminating the least popular candidates and holding subsequent run-off votes until someone wins a majority, but without any of the additional time, hassle and expense. If what I've just described is Hall's ideal - and unless he's guilty of intellectual dishonesty I can only assume it must be - then why on earth is he supporting the status quo in this referendum?
Pointing to the example of the forthcoming London mayoral election, which will use a system similar to AV, Hall claims -
"If the leading candidate doesn't get 50% of the first preference votes, then London's Mayor will be chosen by the second choice votes of the electors that voted for the political wings of fundamentalist Christian, Karmic flying and neo-Nazi parties. That is not just unfair, it's plainly irrational. It gives undue weight to the voters who chose to use the election as a platform to promote their fringe views."
No, it doesn't, Ed. It gives precisely the same weight to every single voter, which is how - in my naivety - I always thought a democratic system was supposed to work. But I can certainly understand Hall's confusion, given that his only previous experience is with a system that saw Russell Johnston elected MP for Inverness in 1992 on 26% of the vote. To use Hall's terminology, first-past-the-post gave 100% "weight" to the one-quarter of people who happened to vote for Johnston, and literally zero "weight" to the three-quarters of people who voted against him. I dare say some of the latter group were sorely tempted to go off and indulge in a bit of foot-stamping afterwards - after all, in those days they didn't even have episodes of Bob the Builder to distract them from the injustice.
Whether Hall realises it or not, what he's actually arguing against is not the idea that fascist-inclined voters and "Karmic fliers" should have a greater say than the rest of us, but rather the idea that they should even have an equal say. That may be superficially attractive to some, but it's an argument against universal suffrage, not AV. Hall doesn't seem to have spotted that under FPTP, fascist voters are already able to decide the outcome of elections, either because they've switched tactically to a mainstream candidate, or because there is no BNP candidate and they consequently have no choice but to vote for a mainstream party. If he finds that so distasteful and "unfair", I presume Hall will be proposing fundamental changes to the current system to - somehow - ensure it can never happen again in future? Mysteriously, it seems not.
But if you think Hall's argument can't get any weaker, you're underestimating the man. He goes on to compare the self-evident 'fairness' of the current system with the way TV talent shows work -
"In modern politics, a winner should be a winner. Try it round your dinner table or next time you watch the X Factor or Strictly Come Dancing. Everyone votes for their favourite book, film or act: surely the candidate with the most votes wins? How would the BBC or ITV possibly explain or justify a programme format with public voting in which the candidate that got the most votes did not win?
'Thank you for calling Britain’s Got Talent. Your vote for the Trapeze Sisters has been counted. Now you can choose a second choice contestant instead. If The Trapeze Sisters don't win, your vote will be transferred to your second choice contestant. Press 2 to register a second preference vote.'"
Oh dear. I'm afraid, Ed, that the voting system for Britain's Got Talent and Strictly Come Dancing is much closer to AV than first-past-the-post. If it worked like the latter, Strictly wouldn't have run for ten or fifteen weeks or whatever it was - there would have been just one programme and one set of dances, at the end of which Ann Widdecombe would have been declared the series winner, on the grounds that her 14% of the popular vote was fractionally higher than any other individual. Your average dinner-party would deem that much fairer than what actually happened, apparently.
Comically, Hall seems to spot this gaping hole in his reasoning almost before he finishes uttering it -
"Of course you do get to vote again in the TV formats as the candidates are knocked out, and the next week's programme starts, but do we really want General Elections every week until we get a winner? That would be the only way to give equal weight to everybody’s second choice votes."
No, it wouldn't, Ed. There's another ingenious way of achieving just that, and you'll be thrilled to hear we're about to hold a referendum on it. It's called AV, or to use its US name, Instant Run-off Voting. The Americans call it that because it produces much the same effect as successively eliminating the least popular candidates and holding subsequent run-off votes until someone wins a majority, but without any of the additional time, hassle and expense. If what I've just described is Hall's ideal - and unless he's guilty of intellectual dishonesty I can only assume it must be - then why on earth is he supporting the status quo in this referendum?
Monday, February 28, 2011
Green on the constitution?
I had a slightly irritable exchange with James MacKenzie at Better Nation yesterday, both on the queries that have been raised about the new YouGov poll commissioned by his party, and on the possibility that they (the Greens) might enter a coalition with Labour should the latter emerge as the largest party in May. It has to be said that James simply didn't have a defence of the poll's methodology on its own merits, although to be fair other commenters did have a go later on. On the coalition front, the impression I got was that James was trying to present his party as roughly equidistant between Labour and the SNP - while he concedes they are closer to the Nationalists on justice policy, nuclear power and the constitution, he claims they are somewhat closer to Labour on public transport and carbon emission targets. He pointed out (and I accept) that they entered into coalition negotiations with the SNP in good faith four years ago, but that they had subsequently gone on to oppose the government vigorously on the basis that it had turned out to be "a business-as-usual administration on poverty, the economy and the environment".
I'd naturally refute the latter point, but it does raise an interesting issue in relation to any future deal with Labour. James has previously proved very touchy whenever it's been suggested to him that the Greens' support for an independence referendum and/or enhanced powers for Holyrood has been quietly dropped, or isn't much of a priority for them. He even memorably worked himself into a state of apoplexy over the synthetic controversy of the 'lapsed tax powers', asking where the equivalent anger from Nationalist bloggers was - the implication seemed to be that the Greens are now the "true believers" in Scottish self-government. Well, let's see the evidence for that. If coalition with Labour becomes a possibility, will the Greens prioritise constitutional progress in the negotiations? I'd suggest the fairly obvious answer is - no, they will not.
The dogs on the street know what an Iain "the Snarl" Gray-led government will look like on the constitution - it will be, to coin a phrase, a "business-as-usual administration". So whatever (probably minor) concessions the Greens might secure on environmental policy, that's the sort of government they would be supporting. They are, of course, perfectly entitled to put constitutional progress for Scotland on the back-burner for five long years if they so wish, and in some ways it might be considered perfectly natural that a Green party would regard such matters as a relatively low priority. But if that is the case, let's have a bit of honesty about it, and at a minimum let's hear no more nonsense about how they are the true 'guardians of the flame'.
I'd naturally refute the latter point, but it does raise an interesting issue in relation to any future deal with Labour. James has previously proved very touchy whenever it's been suggested to him that the Greens' support for an independence referendum and/or enhanced powers for Holyrood has been quietly dropped, or isn't much of a priority for them. He even memorably worked himself into a state of apoplexy over the synthetic controversy of the 'lapsed tax powers', asking where the equivalent anger from Nationalist bloggers was - the implication seemed to be that the Greens are now the "true believers" in Scottish self-government. Well, let's see the evidence for that. If coalition with Labour becomes a possibility, will the Greens prioritise constitutional progress in the negotiations? I'd suggest the fairly obvious answer is - no, they will not.
The dogs on the street know what an Iain "the Snarl" Gray-led government will look like on the constitution - it will be, to coin a phrase, a "business-as-usual administration". So whatever (probably minor) concessions the Greens might secure on environmental policy, that's the sort of government they would be supporting. They are, of course, perfectly entitled to put constitutional progress for Scotland on the back-burner for five long years if they so wish, and in some ways it might be considered perfectly natural that a Green party would regard such matters as a relatively low priority. But if that is the case, let's have a bit of honesty about it, and at a minimum let's hear no more nonsense about how they are the true 'guardians of the flame'.
Sunday, February 27, 2011
A silver lining from YouGov
After the excitement of the recent Ipsos-Mori poll, YouGov have provided a reality-check for the SNP, but certainly not an extinguishing of all hope that the calamity of an Iain "the Snarl" Gray premiership might yet be averted. Here are the full figures -
Constituency vote :
Labour 41% (+1)
SNP 32% (-2)
Conservatives 15% (+1)
Liberal Democrats 8% (-)
Others 4% (-1)
List vote :
Labour 40% (+4)
SNP 26% (-5)
Conservatives 15% (-)
Liberal Democrats 7% (-1)
Greens 6% (-)
Others 5% (-)
The SNP have already raised an important technical objection to the poll, namely that the raw figures have been reweighted in accordance with party identification for Westminster elections, which certainly seems counter-intuitive given the traditional disparity between Westminster and Holyrood polls. I've no idea how much credence to give to that complaint, but if it is true that the unadjusted figures showed a double-digit lead for the Nationalists, it's at least worth keeping an open mind on the point. But there are a number of other crumbs of comfort for the SNP as well -
1) For a poll showing a significant Labour lead, the SNP are proving remarkably resilient on the constituency vote - just 1% down on their winning position from 2007.
2) The greater Labour lead on the list vote doesn't pass the 'smell test' somehow. In all of the three Holyrood elections to date, the SNP have managed to retain more of their constituency votes on the list than Labour have, and it's hard to think of any particular reason why that pattern should be completely reversed now.
3) The dire showing for the Lib Dems continues to jeapordise Labour's chances of cobbling together a stable parliamentary majority, whether in coalition or a looser arrangement with Tavish Scott's party. On the seat projections from this poll, the two parties would just about have a majority between them, but there's very little room for slippage from Labour's heady current standing.
4) Although Patrick Harvie seems to be touting the Greens as potential alternative partners for Labour, his party's level of support remains very much on the borderline between a breakthrough in terms of seats, or a 2007-style result. In any case, if they couldn't find sufficient common ground with the SNP four years ago, it seems somewhat doubtful that they'd be able to suffer the 'born to rule' arrogance of Gray's mob for a full term.
It's also worth pointing out that the Ipsos-Mori poll arrived in the wake of Scottish issues (the budget and the Megrahi report) featuring very prominently in the news. With the recent blanket coverage of the revolutions in the Middle East, the reverse is true this time round. We'll have to wait until the formal campaign to get a proper sense of how the stark choice between Salmond and Gray plays with the public when placed before them on a nightly basis.
UPDATE : Stuart Dickson has forwarded me this email he sent to YouGov's Peter Kellner and Anthony Wells, raising a further issue I hadn't previously spotted -
BPC disclosure rules - new YouGov/Scottish Green Party poll published in today's Sunday Herald
Hello Peter and Anthony,
I note that the detailed tables for this YouGov poll...have not appeared yet at the YouGov website. You have been very good lately, so I look forward to perusing the detailed tables as soon as possible (later today?).
It is unsatisfactory that the newspaper article ONLY reports the "certain to vote" VI figures: this is non-standard for YouGov polls, and I can only presume that it is designed to mislead. Please slap the Scottish Green Party / Sunday Herald on the hand.
COPY OF MY POST ON UK POLLING REPORT:
A new YouGov poll on behalf of the Scottish Green Party is published in today’s Sunday Herald.
Dire for the Lib Dems, but unclear whether it is good for Labour or the SNP. Prior to weighting being applied the SNP were 13 POINTS ahead!! After weighting was applied, the SNP were suddenly 9 POINTS behind !! Go figure.
But I am confused by one thing: why on earth are the SH only publishing the “certain to vote” figures ?!? YouGov polls are not normally reported in this fashion. Anthony, can you provide the normal headline figures, as they will appear when YouGov get round to publishing the full tables? Thanks in advance.
On one point YouGov can probably be absolved of blame - it's not unusual for them to wait until Monday to put the details of a Sunday poll on their website. However, if Stuart is right about their normal practice in relation to certainty to vote (they've changed their methodology so many times over the years I've lost track) the partial reporting of this poll in the Herald does seem a bit suspicious. Given that the Scottish Greens commissioned the poll, it's hard not to wonder whether the figures restricted to those certain to vote just happened to be more favourable for Patrick Harvie's party.
Constituency vote :
Labour 41% (+1)
SNP 32% (-2)
Conservatives 15% (+1)
Liberal Democrats 8% (-)
Others 4% (-1)
List vote :
Labour 40% (+4)
SNP 26% (-5)
Conservatives 15% (-)
Liberal Democrats 7% (-1)
Greens 6% (-)
Others 5% (-)
The SNP have already raised an important technical objection to the poll, namely that the raw figures have been reweighted in accordance with party identification for Westminster elections, which certainly seems counter-intuitive given the traditional disparity between Westminster and Holyrood polls. I've no idea how much credence to give to that complaint, but if it is true that the unadjusted figures showed a double-digit lead for the Nationalists, it's at least worth keeping an open mind on the point. But there are a number of other crumbs of comfort for the SNP as well -
1) For a poll showing a significant Labour lead, the SNP are proving remarkably resilient on the constituency vote - just 1% down on their winning position from 2007.
2) The greater Labour lead on the list vote doesn't pass the 'smell test' somehow. In all of the three Holyrood elections to date, the SNP have managed to retain more of their constituency votes on the list than Labour have, and it's hard to think of any particular reason why that pattern should be completely reversed now.
3) The dire showing for the Lib Dems continues to jeapordise Labour's chances of cobbling together a stable parliamentary majority, whether in coalition or a looser arrangement with Tavish Scott's party. On the seat projections from this poll, the two parties would just about have a majority between them, but there's very little room for slippage from Labour's heady current standing.
4) Although Patrick Harvie seems to be touting the Greens as potential alternative partners for Labour, his party's level of support remains very much on the borderline between a breakthrough in terms of seats, or a 2007-style result. In any case, if they couldn't find sufficient common ground with the SNP four years ago, it seems somewhat doubtful that they'd be able to suffer the 'born to rule' arrogance of Gray's mob for a full term.
It's also worth pointing out that the Ipsos-Mori poll arrived in the wake of Scottish issues (the budget and the Megrahi report) featuring very prominently in the news. With the recent blanket coverage of the revolutions in the Middle East, the reverse is true this time round. We'll have to wait until the formal campaign to get a proper sense of how the stark choice between Salmond and Gray plays with the public when placed before them on a nightly basis.
UPDATE : Stuart Dickson has forwarded me this email he sent to YouGov's Peter Kellner and Anthony Wells, raising a further issue I hadn't previously spotted -
BPC disclosure rules - new YouGov/Scottish Green Party poll published in today's Sunday Herald
Hello Peter and Anthony,
I note that the detailed tables for this YouGov poll...have not appeared yet at the YouGov website. You have been very good lately, so I look forward to perusing the detailed tables as soon as possible (later today?).
It is unsatisfactory that the newspaper article ONLY reports the "certain to vote" VI figures: this is non-standard for YouGov polls, and I can only presume that it is designed to mislead. Please slap the Scottish Green Party / Sunday Herald on the hand.
COPY OF MY POST ON UK POLLING REPORT:
A new YouGov poll on behalf of the Scottish Green Party is published in today’s Sunday Herald.
Dire for the Lib Dems, but unclear whether it is good for Labour or the SNP. Prior to weighting being applied the SNP were 13 POINTS ahead!! After weighting was applied, the SNP were suddenly 9 POINTS behind !! Go figure.
But I am confused by one thing: why on earth are the SH only publishing the “certain to vote” figures ?!? YouGov polls are not normally reported in this fashion. Anthony, can you provide the normal headline figures, as they will appear when YouGov get round to publishing the full tables? Thanks in advance.
On one point YouGov can probably be absolved of blame - it's not unusual for them to wait until Monday to put the details of a Sunday poll on their website. However, if Stuart is right about their normal practice in relation to certainty to vote (they've changed their methodology so many times over the years I've lost track) the partial reporting of this poll in the Herald does seem a bit suspicious. Given that the Scottish Greens commissioned the poll, it's hard not to wonder whether the figures restricted to those certain to vote just happened to be more favourable for Patrick Harvie's party.
Subscribe to:
Posts (Atom)