As in past years, Nicholas Gruen is organising a group subscription to Crikey. Check it out here
Lindzen and “No statistically significant warming since 1995”
I discussed the ‘no statistically significant warming since 1995’ talking point in this post. For those who came in late, this talking point has been around the delusionist blogosphere for some time, though with a lower profile than ‘global warming stopped in 1998’, and was put as question to Phil Jones of UEA in a BBC interview. Jones answered honestly, if a bit clumsily, that the data period since 1995 is marginally too short to derive a statistically significant trend, a response which was headlined by the Daily Mail as “Climategate U-turn as scientist at centre of row admits: There has been no global warming since 1995?” and became the talking point of the day. As has been widely noted, confusing not statistically significant’ with ‘not significant; in the ordinary sense indicates either deliberate dishonesty or ignorance of a point covered in excruciating detail in every introductory stats course.
But where did this silliness come from? I’d seen Janet Albrechtsen quote Lord Monckton on the point, and it seemed about right for him, an innumerate debating point that would take a fair while to refute, during which time he could move on to the next one.
Imagine my surprise, then, when I discovered the point being made (and apparently originated) by Richard Lindzen of MIT who is (or ought to be) by far the most credible figure on the delusionist side. In a piece published on “Watts Up With That” Lindzen says ‘There has been no warming since 1997 and no statistically significant warming since 1995’ and in this piece for Quadrant he gives a variation, saying “has been no statistically significant net global warming for the last fourteen years” and “the fact that warming has ceased for the past fourteen years is acknowledged” . Note the slide from “has been no statistically significant net global warming for the last fourteen years ” to “warming has ceased”, committing the basic newbie error against which all budding stats students are warned.
Lindzen has published a couple of hundred papers in climatology, so I think we can assume he knows that the statement “there has been no statistically significant warming since 1995” means nothing more than “given the variability in the data, we need at least 15 observations to reject the null hypothesis at 95 per cent confidence”, a fact so trite as not to be worth mentioning.
It is sad to see a respected scientist reduced to this kind of thing. And as far as I can tell, all this is simply to avoid admitting that he backed the wrong horse back in 1990, when he bet that he was smarter than the majority of climate scientists who thought humans were (probably) causing global warming. The data since then has supported the majority view, but instead of revising his position, Lindzen has resorted to dishonest statistical trickery.
To quote The Economist, with respect to the Daily Mail
Since I’ve advocated a more explicit use of the word “lie”, I’ll go ahead and follow my own advice: that Daily Mail headline is a lie.
But at least the Daily Mail headline writer could plead ignorance. Lindzen has no such excuse.
Update: More on this from Deep Climate
Sockpuppet attack
I’ve just edited several comments from sockpuppets ‘Julia’ and ‘Colin’ posting Liberal Party talking points. Both posted using IP 114.78.24.168, then ‘Julia’ switched to another one, also in the range 114.78.2x.xxx. Spurious email addresses jules77@hotmail.com and col.TRE67@hotmail.com. Anothe commenter advises ‘Julia’ has been seen at LP.
If anyone can out these pests, I’ll happily publicise the results.
Four lies and an empty set
The switch from ‘scepticism’ to an overtly anti-science, and anti-scientist, position has paid some dividends for the advocates of delusional views on climate change. Although the only wrongdoing in the CRU harassment/hacking campaign was that of the hackers and their allies (more on this soon), the false charges of scientific fraud got much more media attention than their refutation, and the early responses to alleged FOI breaches (including that of some pro-science writers such as George Monbiot) failed to recognise of the deliberately vexatious misuse of FOI by the anti-science group. In this context, the discovery of a genuine error in the 1700 page IPCC Fourth Assessment Report regarding projected changes in glaciers could easily be spun into something much bigger.
It seemed for a little while as if the delusionists had scored another win, when Phil Jones, the scientist who has been most viciously target by the hackers/harassers gave an honest answer to a deliberately loaded question prepared by them and put to him in a BBC interview
BBC: Do you agree that from 1995 to the present there has been no statistically-significant global warming?
Phil Jones: Yes, but only just. I also calculated the trend for the period 1995 to 2009. This trend (0.12C per decade) is positive, but not significant at the 95% significance level. The positive trend is quite close to the significance level. Achieving statistical significance in scientific terms is much more likely for longer periods, and much less likely for shorter periods…
This was headlined by the Daily Mail as “Climategate U-turn as scientist at centre of row admits: There has been no global warming since 1995?.
But, now it looks as if this is starting to backfire. Anyone who has successfully completed a basic course in statistics knows that “statistically significant” has a special meaning which bears almost no relationship to the usual sense of the term “significant”, and this fact is beginning to sink in.
Monday Message Board
It’s time, once again for the Monday Message Board. Post comments on any topic. As usual, civilised discussion and no coarse language.
Invulnerable zombies – the EMH
My book is due for publication in the US Fall, so work on my added sections on reanimation is starting to feel like one of those games where you have to kill all the zombies to get to the antidote before you fall victim yourself. Here’s another one – more on the way
Weekend reflections
It’s time again for weekend reflections, which makes space for longer than usual comments on any topic. Civilised discussion and no coarse language please.
Zombie economics gets a mulligan: or, how Obamacare caused the Global Financial Crisis
I’m adding a little section to each of the chapters in my Zombie Economics book called “Reanimation”, about the attempts that are already under way to revive economic ideas killed (at least according to the standard rules of hypothesis refutation) by the global crisis. I wasn’t surprised to find plenty of examples for the efficient markets hypothesis (easy to render immune from any kind of refutation by an appropriate formulation) or for policy ideas that yield big benefits to the rich and powerful, such as privatisation and trickle-down economics. But I was surprised a little while ago to see the crisis described as a transitory blip in the continuing Great Moderation. Still that pales into insignificance compared to this piece by Casey Mulligan of Chicago (h/t commenter Daniel ), in which (I swear this is true!) the crisis is the result of financial markets correctly anticipating the adverse labour market impacts of possible legislation under Obama, such as a health plan that might include means tests.
Two kinds of ignorance
Also, in yesterday’s Fin, Geoffrey Barker accused Abbott of going for the bogan vote (paywalled), where bogan is taken to mean ignorant. Leaving aside the class/cultural analysis implicit in the term “bogan”, which I think is wrong, the argument is the same as I made in my post on agnotology, as his characterization of Rudd as a technocrat, not really at ease with the kind of politics that includes demands for authenticity and so on. Coming back to “bogan”, the big issue in agnotology is not ignorance in the ordinary sense of the term (people who don’t know much about political issues, and don’t care to learn – that is certainly part of the stereotypical bogan image, and may perhaps be descriptive of the actual demographic groups commonly associated with the term, though I don’t know of any evidence of this).
The ignorance associated with climate change delusionism and other rightwing factoids is metacognitive and has much more to do with the Dunning-Kruger effect of overestimating one’s own competence. The classic example is the kind of person who eagerly circulates reports that there has been no statistically significant warming since 1995. The only information content in such a report is that the person doing the reporting doesn’t understand the concept of statistical significance[1], and therefore is incapable of assessing any issue involving statistical analysis, of which climate change is a prime example.
The stereotypical candidate, in relation to climate change, is that of a 50+ male[2] with a business background in engineering or some similar field where practical judgement is accorded more value than theoretical expertise, and where a willingness to push on regardless is an important element of success. Journalists and opinion columnists[3], accustomed to “mastering a brief” at short notice are also highly susceptible – lawyers who may actually have to master briefs involving technical issues seem mostly to recognise that this is the kind of problem where expert judgement is required, as does the more sensible kind of economist[4]
fn1. Note for pedants. A Bayesian statistician would say that confusion over the concept of significance reflects the logical problems of the concept and the underlying classical theory of statistics. But that only makes sloppy misuse of the concept even worse. I’ll have more to say on this soon, I hope.
fn2. A demographic group to which I belong
fn3. This one, too.
fn4. This one, too, I hope.
Greece: Haven’t we seen this movie before
My piece in today’s Fin compares Greece with some Australian states who also play games to conceal debt. The emerging news (also in today’s NY Times) is that the same banks who facilitate the dodgy debt deals established a CDS market for Greek sovereign debt and some have large short positions (translated from marketspeak: they are betting that the deals they set up will go bad and Greece will default). This can of course be defended as insurance, but it obviously changes your relationship with your bank or financial advisor if they can steer you into a deal and then bet on its failure. The potential for moral hazard in the CDS market has yet to be fully explored, but I think we will get to find out the hard way before too long.
John Quiggin 