Category: Boneheaded stupidity

Statistical significance

John Quiggin27 Comments

I know I should just ignore the Oz, but faced with its continuous campaign to promote innumeracy, cheered on by the likes of Alan Jones and Andrew Bolt, I can’t help but try to set things straight. We’ve seen on many occasions that nearly all “sceptics” either misrepresent of misunderstand the concept of statistical significance, assuming it to correspond to the ordinary meaning of “significant”. The classic example is the Lindzen talking point, made in 2008 that “there has been no statistically significant warming since 1995”. As everyone who understands statistical signifance (notably including Phil Jones, who gave an accurate response and saw a distorted version of his words become a delusionist meme), that’s because statistical significance depends on sample size. Roughly speaking, to see a significant upward trend in a noisy time series, the trend, multiplied by the number of years of data, needs to be about twice the standard deviation of the random variation about trend. So, if you have an upward trend of 0.015 degrees per year, and a standard deviation of 0.1 (these are estimates, but feel free to check)) you typically need 14 or 15 years of data to see a statistically significant trend. Over shorter periods, it’s easy to eyeball a pause or decline, as this graph from Skeptical Science shows.

Lindzen obviously knew this, and it was easy to check that he could go back 13 years from 2008 (but no further) without finding a statistically significant trend. He also knew that, given a few more years of data, the trend for the period since 1995 would be statistically significant, but correctly assumed that no-one on the delusionist side would know or care. Now, the Oz has this, from Michael Asten, professor of geophysics at Monash University. It’s worded carefully enough for me to think he knows he’s pulling the same swifty as Lindzen, but it’s hard to tell for sure[1]

Global temperatures have not increased in a statistically significant sense in the past 15 years. A pause of 10 years in the upward trend of the past 40 years would be unsurprising from existing models. A pause of 20 years would definitely surprise. Changes across the next five years will be watched closely.

As you would expect, Asten has to move Lindzen’s goalposts forward by a couple of years, to an implied starting date of 1997. Note also that he slides from “no statistically significant trend” to “a pause”. What can we say about this? In one sense he is right. As I’ve said, we need about 15 years of data to get a statistically significant trend, so we wouldn’t expect to find one with 10 years, and we would usually expect to find one with 20 years. But, of course, that number itself is variable. Asten is repeating basic facts about time series, in a way that would lead unwary or gullible readers (the vast majority, given the outlet) to suppose that recent evidence casts doubt on the observed warming trend. The only thing that’s hard to figure here is whether he is fooling himself as well as his readers.

fn1. (Lindzen himself often slipped from “no statistically significant warming” to “no warming” either out of sloppiness or because he thought no one was looking.

Impressive logic

John Quiggin32 Comments

Watching the 7:30 Report the other night, I saw Bronwyn Bishop (once touted as a possible PM) oppose legislation requiring automatic enrolment of 18-year olds to vote (already in place for state elections in NSW and Victoria). Her argument “it is a binding requirement, under the Electoral Act, for people to enrol themselves”. Umm, yes, and tomorrow, when the new legislation is passed, no such binding requirement will exist.

Fortunate in my enemies

John Quiggin218 Comments

That’s how Robert Vienneau described me after some of my stoushes last year.

It seems as if my luck is holding in that respect at any rate. While I’ve had plenty of supportive responses after being booted from the Fin, I’m sure not everyone is sorry to see me go. Most of those in the latter class, however, haven’t seen any need to gloat.

I would have been disappointed, however, if Andrew Bolt had not lived down to his usual form on this occasion. Sure enough, as his fans have advised me both by email and in comments here, he’s written a gloating column, expressing the hope that Laura Tingle (a far better journalist than Bolt could ever be, even if he was trying) will be next to go.

Bolt can’t even manage an original line of attack, dragging out the tired misrepresentation of a 2007 blog post that the Telegraph ran last week.

The great thing about having Bolt as an enemy is that you get his fans thrown in as part of the package. There’s something comforting in knowing that, if someone dislikes you, there’s a high probability that they are the kind of person who comments on Bolt’s blog.

Of course, it isn’t much of a distinction to be one of Bolt’s enemies. With the exception of the late Paddy McGuinness (who at least had some style to combine with the vitriol) I can’t think of anyone who is less discriminating in his hatreds.

Megalogenis messes up the generation game

John Quiggin23 Comments

George Megalogenis is a smart and insightful writer, so I opened his new book The Australian Moment with anticipation. Alas, the very first sentence is the most tiresome of generation game cliches

Every rich nation reveals its character in its most selfish generation – the baby boomers

Showing that invoking the word “generation” instantly reduces your IQ by about 50 points, Megalogenis goes on to support his claim by a discussion of the results of a poll of voters taken in 1972, the year Gough Whitlam was elected, and the year before his government lowered the voting age from 21 to 18.

The baby boom began in 1946, when returning members of the armed services started having big families. The end is usually placed somewhere between 1960 and 1963.

Whichever end date you choose, even the many boomers who were still in primary school ought to have been capable of doing the subtraction that would inform George that anyone born after 1951 (that is, the vast majority of boomers) was not a voter in 1972.

Austerity and academia

John Quiggin179 Comments

Austerity is hitting lots of people, across pretty much all social classes, except for the top 1 per cent, who are rapidly recouping their losses in the GFC and will soon be pulling even further away from the 99. Just at the moment, academics seem to be in the crosshairs, from Washington to Sydney and beyond. Here’s a post on the subject from my friend and former colleague Rohan Pitchford. To forestall a possible line of criticism, let me observe now that, while academics have it better than plenty of others under attack from austerity policies, anyone who plays on this kind of division is a tool of the 1 per cent, and will be treated as such by me.

Sackings Hit Economics School Hard (Guest post from Rohan Pitchford)

I was surprised and dismayed to hear that that several of my former colleagues at the School of Economics at the University of Sydney have been told to leave their jobs by July. A remarkable aspect is that the sackings happened by edict, several layers of administration above the School level. How is it possible for such a removed group to know the details of people’s work life, their roles and the reasons behind their roles without any form of consultation? (ANU also faces budget cuts, but is taking the enlightened bottom-up approach.)

I know all of those concerned personally, and the group includes both talented researchers, teachers and administrators. They were apparently selected using retrospective publication criteria. We all know that most people experience a ‘bare patch’ in their publications during the life course, whether it be because of a new child, a death in the family, illness (admin duties!) etc. Knowing the people involved, the criteria seem to me unfair and random.

Perhaps the most astounding fact in all of this is that Sydney Economics has been perennially understaffed– as a Professor there, I estimated that they were some 10 academics short of what is required to deliver requisite courses: The average class size is 110, I conjecture larger than any other department. There are not enough staff to cover all the classes taught, let alone to reduce class sizes to educationally appropriate levels. Sydney typically has had to hire part-timers to fill the gap. The department generates some 20 million dollars per year in revenue from its teaching program. I cannot imagine that this will do anything but hurt this important revenue base.

A big question is this: Are the sackings due to productivity, or are they in response to an administration that has grossly over-spent on buildings? I have heard rumours of expenditure of 100m on a new medical centre, and 360m on a new obesity centre preceded these sackings.

Solutions? I discuss a possible way forward for Australia here:

Alesina, Ardagna, Austerity, Australia

John Quiggin13 Comments

A while ago I published a blog post, and later a Fin article, pointing out that the influential Alesina and Ardagna article, Tales of Fiscal Adjustment, that coined the term “expansionary austerity” was riddled with factual and analytical errors in its discussion of Australia. That piece has elicited a string of lengthy replies from the Catallaxy/CIS team, notable for the absence of any substantive content. Sinclair Davidson produced a mammoth post with multiple updates, entirely devoted to refuting a parenthetical snark on my part that the paper wasn’t peer reviewed. Now there’s another one from Steve Kates, who wants to quibble about the chronological relationship between Jean-Baptiste Say and the Mills, father and son.

So, an open challenge. In my original post, I give seven quotes from the Alesina and Ardagna article, all of which I say are wrong or at least misleading. Does anyone want to defend any of these? Special bonus points for anyone who can defend the opening sentence of their Australian section, which reads “In 1985, a single-party left-wing government took office and launched a stabilization plan to correct the internal and external imbalances (the current account deficit was 4.13% of GDP and the total deficit/GDP ratio was above 3% in 1984). ” (emphasis added).

UpdateKates has added a lengthy update to his post, without, AFAICT, defending the erroneous claims in A & A

The zombie economics of austerity in Australia (updated again)

John Quiggin128 Comments

Yesterday’s Fin ran a piece from Stephen Kirchner and Robert Carling of the Centre for Independent Studies, under the headline “Give austerity a chance” which was a pretty accurate summary of the contents. It’s paywalled, but you may be able to read it by clicking here. The piece relies almost exclusively on the work of Alberto Alesina and his colleagues, promoting the zombie idea of expansionary austerity. As I pointed out here, the most influential of these pieces by Alesina and Ardagna, is riddled with errors, at least as it applies to Australia.

Although Kirchner is a blogger himself, he and his co-author could be forgiven for missing my post. But Alesina’s work is probably the most-refuted piece of economic analysis put out (though never published in a peer-reviewed journal) in recent decades. It’s been demolished not only by the usual suspects like Krugman and DeLong (and me), but by the Economist, the IMF and even by one of Alesina’s own co-authors, Roberto Perotti.

Charitably assuming that Kirchner and Carling had managed to miss just about every publication on the question of austerity in the last year, could they not have spent 30 seconds with Google before hitting “Send”? A search on Alesina+austerity reveals a torrent of criticism, none of which they mention.

It is hard to know which is worse – the possibility that Kirchner and Carling, presented by the CIS as expert economists, were ignorant of all this, or the alternative hypothesis that they knew it and decided not to mention it. Either way, it’s an appalling breach of elementary standards of research.

I’m pretty sure the facts have been brought to the attention of Kirchner and Carling. The honest thing to do would be to write to the Fin pointing out that the work on which they relied was, at best, highly controversial. If Kirchner, Carling and the CIS are unwilling to do this, we can draw the conclusion that they cannot be trusted in anything they write.

Update Sinclair Davidson at Catallaxy has a lengthy reply, but the sole substantive criticism is that contrary to my parenthetical remark, Alesina and Ardagna did finally publish a peer-reviewed paper in 2010. But the work that was actually influential was done back in the 1990s. I’ll republish my blog post pointing out what a shoddy job that paper in describing developments in Australia. Davidson’s piece is notable for the lack of any substantive defence of Alesina’s work, and also for this , offered in response to my observation that the research in question had been comprehensively demolished by the IMF among many others

Fancy that – cutting edge research into a highly politicised aspect of public policy is “controversial”. Does Quiggin think AFR readers are so dumb they wouldn’t realise that?

So, next time you read an opinion piece from the CIS you can safely assume the caveat lector “This research is probably discredited, the authors almost certainly know it, but, if so, they’re not going to tell you”.

No one expects opinion pieces to be “fair and balanced”, but if you are going to rely on work that has been subject to serious and credible criticism, you should at least point out the main criticisms and (if possible) say briefly why you think they don’t stand up. As an example, Wilkinson and Pickett’s The Spirit Level produces some striking evidence of relationships between inequality and bad social outcomes.. This work has been subject to a lot of criticism, not fatal in my view, but enough that it needs to be mentioned. I did this when I cited the work in Zombie Economics and then at greater length here

Further update While still not disputing any of the substantive points I’ve raised, Davidson digs deeper on the question of whether the original Alesina and Ardagna work was published in a peer-reviewed journal. The work was published in Economic Papers, which does not take unsolicited submissions. Rather the editors commission pieces, or you can propose a piece to them. That is, this is, as the webpage says, a policy forum, not an academic journal. Standard practice for publications of this kind is for the editors to approve (or return for revision, or, very rarely, reject) the pieces they’ve commissioned. This isn’t peer-review in the normal sense. I’ve always assumed that Economic Papers follows the standard practice in this respect, but Davidson is welcome to check it out, if he cares enough.

As a PS, I couldn’t resist checking a 700-comment thread on the US elections. I shouldn’t link, but I will. While there is plenty of not-so-innocent amusement to be had, what struck me was that most of the commenters appear to be creationists – the handful holding up the flag for evolution are getting hammered.

Not Lake Wobegon

John Quiggin26 Comments

I haven’t been paying much attention to the Oz since it went behind the paywall, but I happened to pick up a copy of today’s paper edition, and came across a fascinating piece by Christian Kerr (who, IIRC, used to write for Crikey as “Hillary Bray”). Trying to talk up public opposition to equal marriage as a reason for Labor hesitancy to push hard on the issue, he cites a survey showing that around 25 per cent of Australians agree with the proposition homosexuality is immoral. Conscious that 25 per cent is, well, a minority, he decides to look at individual electorates.

What’s really striking is Kerr’s discovery that ‘In 80 of the 150 federal electorates, an above-average number of people support the proposition’. I did some quick math of my own, and it turns out that 80 is almost exactly half of 150. So, next time you see a sample estimate that doesn’t suit your case, be sure to check subsamples. You, too, may find that half of them are above the average.

Expansionary austerity: some shoddy scholarship (repost)

John Quiggin37 Comments

I’ve just read ‘Tales of Fiscal Adjustment’ by Alesina and Ardagna, which appears to be the founding text for the idea of expansionary austerity. The level of scholarship, at least as it applies to Australia (which is their first illustration) is exceptionally poor, to the extent that it requires a rescuscitation of the ancient Internet tradition of Fisking. I’m going to quote excerpts from their text (about 50 per cent of the total), and intersperse them with my comments.

Read More »