Another nail in the Superfreakonomics coffin

By now, I'd expect the authors of Superfreakonomics are having mixed feelings about their new book. On the one hand, they're making good money as the book enjoys healthy sales. On the other, just about every actual expert in the field to which Chapter 5 is devoted -- climate change -- has savaged their take on the subject.

This week comes perhaps the most devastating criticism, from four statisticians whose analysis of global temperature trends demonstrates just how wrong Steven D. Levitt and Stephen J. Dubner got the numbers. And an AP story on the statisticians' analysis raises some serious questions about the sincerity of the duo.

The misguided authors wrote:

"Then there's this little-discussed fact about global warming: While the drumbeat of doom has grown louder over the past several years, the average global temperature during that time has in fact decreased."

As everyone with even a passing familiarity with and respect for climatology knows, this is a standard case of misdirection. But the AP decided to get some real expert opinion on the alleged trend. Seth Borenstein reports:

In a blind test, the AP gave temperature data to four independent statisticians and asked them to look for trends, without telling them what the numbers represented. The experts found no true temperature declines over time.

"If you look at the data and sort of cherry-pick a micro-trend within a bigger trend, that technique is particularly suspect," said John Grego, a professor of statistics at the University of South Carolina.

...

The recent Internet chatter about cooling led NOAA's climate data center to re-examine its temperature data. It found no cooling trend.

"The last 10 years are the warmest 10-year period of the modern record," said NOAA climate monitoring chief Deke Arndt. "Even if you analyze the trend during that 10 years, the trend is actually positive, which means warming."

The AP sent expert statisticians NOAA's year-to-year ground temperature changes over 130 years and the 30 years of satellite-measured temperatures preferred by skeptics and gathered by scientists at the University of Alabama in Huntsville.

Statisticians who analyzed the data found a distinct decades-long upward trend in the numbers, but could not find a significant drop in the past 10 years in either data set. The ups and downs during the last decade repeat random variability in data as far back as 1880.

Saying there's a downward trend since 1998 is not scientifically legitimate, said David Peterson, a retired Duke University statistics professor and one of those analyzing the numbers.

Identifying a downward trend is a case of "people coming at the data with preconceived notions," said Peterson, author of the book "Why Did They Do That? An Introduction to Forensic Decision Analysis."

But is it worse that that? Levitt and Dubner insist they are not climate change deniers. They say as much in their book. But their chapter on the subject contains many of the standard denier talking points. It begins with the planet-has-been-cooling-since-1998 meme. The subtitle of the book is "Global Cooling, Patriotic Prostitutes, and Why Suicide Bombers Should Buy Life Insurance."

From Borenstein we learn that

Levitt, a University of Chicago economist, said he does not believe there is a cooling trend. He said the line was just an attempt to note the irony of a cool couple of years at a time of intense discussion of global warming. Levitt said he did not do any statistical analysis of temperatures, but "eyeballed" the numbers and noticed 2005 was hotter than the last couple of years. Levitt said the "cooling" reference in the book title refers more to ideas about trying to cool the Earth artificially.

It's true that most of Chapter 5 deals with geoengineering schemes to counter global warming. But Could Levitt and Dubner have really just chosen their words carelessly? Did they really think that including the phrase "global cooling" in the subtitle and leading off the chapter by repeating the most common denier argument would lead readers to believe they don't have a problem with the consensus view of anthropogenic climate change?

To talk about global cooling at the end of the hottest decade the planet has experienced in many thousands of years is ridiculous," said Ken Caldeira, a climate scientist at the Carnegie Institution at Stanford.

Ben Santer, a climate scientist at the Department of Energy's Lawrence Livermore National Lab, called it "a concerted strategy to obfuscate and generate confusion in the minds of the public and policymakers" ahead of international climate talks in December in Copenhagen.

I see two possibilities. They could be just dilettantes with no real grasp of the subject or facility with language who think "eyeballing" data is good enough when it comes to challenging the collective wisdom of thousands of genuine experts, in which case they're unwitting allies of the denial movement. Or there could be a more mischievous motive at work, one that puts them squarely in a camp that's anything but well-intentioned. The word "disingenuous" comes to mind.

Given the stakes, an honest answer would be helpful. Which is it, gentlemen?

More like this

There used to be a time when a scientists would carry out experiments in the lab and tell their students: "it doesn't matter what you think, its whether you can come up with something which predicts the data".

How times have changed. These days any old model will do so long as it has a passing resemblance to the data and forget checking whether e.g. that the IPCC predicted 1.4-5.8C warming in 2001 and the actual trend is -1C. Because you don't have to check your models because only "holocaust like denialists" would deny that the models fit the data just because of 8 years of cooling.

No, its a very different subject to the one I took at University. Things are not decided by the evidence, instead they hold a poll and if 90% of people in the sample they selected to poll agree then THE SCIENCE IS SETTLED.

Tell me where I can hand back my degree because it now doesn't seem to be worth the paper is was written on.

and the actual trend is -1C.

Did you even read the post? The actual trend is up.

By David MarjanoviÄ (not verified) on 27 Oct 2009 #permalink

I've heard bad things about the data analysis in other parts of this book. It just keeps getting worse. It's disappointing.

Is there any doubt that they just want to sell books? To muddy the waters to make a few bucks is pretty reprehensible. My verdict: Scumbags.

Borenstein is a known advocate so you lose a little bit of objectivity there. I don't see how this is even a story. Statisticians are supplied number sets that show cooling. Apparently these stat guys are capable of math. That's all we really found out.

Of course there are a few different number sets. If you use the ones that showing cooling since 98 , the stats guys will find that too.

So the real news is that advocate Seth prefers the number sets that show warming.

This is a manufactured news story - nothing more.

Levitt said he did not do any statistical analysis of temperatures, but "eyeballed" the numbers and noticed 2005 was hotter than the last couple of years.

*bangs head against desk*

Do not let this many anywhere near any real data. That is absolutely clueless. Levitt doesn't deserve to be taken seriously if that's how he does his analysis. OK, perhaps it's just a throw-away line, but then why bother to make the comment if you can't be bothered to check it.

Rick - read the post. There is no cooling.

I was reading about the placement of NOAA ground tempature collectors and it's somewhat disturbing how far out of calibration/tolerance these are. Some are placed on the top of blacktop roofs or in parkinglots. others are right next door to a building or next to AC exhausts. I'm neither confirming or denying the climate change hoax, simply pointing out that the junk science that is being passed off as "settled".

Actually, the book was very good. One thing you need to remember is that in a complex system such as climate research there are no "experts" as you can have in, say, dentistry. So every time I hear someone say this or that expert disagrees, well, I understand exactly what that means. Apparently, all these wondrous "climate" experts forgot about Lorenz. Rather than ridicule, one might try reading the book first. Only after having read it, and then reading, again, the negative posts, I realized it was clear that these aren't the people I want doing anything with complex systems because they cannot even understand one simple chapter in a popular book. Also clear was the fact that many of them hadn't read it and were, in fact, derivitive critics using selected information by various people (cough Romm) with an agenda. Loved the book. It was excellent.

There was a time when scientists and statisticians alike would have scoffed at taking a sample of 130 (years of record keeping) working against 4.5 billion (estimated years of earth's age), and ignoring (geological) history of multiple temperature swings from extreme tropical heat to extreme ice-age cold to reach a concrete conclusion about ANYTHING, let alone global temperature trends. Those days of logic seem lost forever... No matter one's belief in the damage we humans do to the planet (and IMO it IS extensive), the science at work here is seriously flawed. In the not so distant past (relative to 4.5 billion years), my home on the shore of the Great Lake Erie would have been crushed beneath a two-mile thick ice sheet. Thank whomever you choose for the warming that occurred to make my neighborhood habitable- I'm glad we weren't there to argue flawed science to stop the thaw!

And, once again, this place gets inundated with people who are labouring under the illusion that they know what they're talking about.

"The temperature stations are all bogus!" "There's no such thing as a climatology expert, and none of them understand Lorenz!" "Paleoclimatologists don't exist!" Arrgh. Crap onna stick people, will you get it through your heads that "I don't know of any ..." is NOT the same as "there aren't any ...", and GO LEARN.

Your contempt for people in the field is evidently based on not knowing what climate researchers actually do, what they learn, and what the physics of the real world is like -- in other words, it's YOUR failing, not theirs. Other people look on YOU with contempt for your displays of the arrogance of ignorance.

By Luna_the_cat (not verified) on 27 Oct 2009 #permalink

Of course, Levitt and Dubner's response would be, "but all this really isn't even the point of the chapter!" I do kind of wonder if they were just sloppy with the denial talking points and justified it to themselves as not really the point of the chapter anyway. Then again, if your point is to promote geoengineering, casting doubt on global warming doesn't really strengthen your case, whereas if you're just throwing out every reason you've heard why we shouldn't rein in CO2 emissions, then the chapter makes sense.

J,
The warming is confirmed by satellites -- if you read the AP article, you'll see they sent the UAH satellite data set to the statisticians as well. It is also confirmed by natural indicators. Also, the fact that warming has been strongest in the Arctic would be problematic for your idea.

charles,
Lorenz's models showed unpredictable weather, not unpredictable climate. We don't know what the weather will be next week, but we know that winter will be colder and summer will be warmer, we don't know if a hurricane will strike Miami on August 11, 2015, but we know it won't on Christmas. The climate does not show long term warming or cooling trends caused by the butterfly effect.

Mike,
Talk about your impossible goalposts! So once we get to 1 billion years of climate data with rising CO2 levels, maybe at that point we'll know something about our contribution?

What if we knew the dramatic but slow swings from ice ages to interglacials were pretty well predicted by slow, predictable changes in the Earth's orbit, and we knew that those factors in isolation should be gradually cooling the Earth, can we then know anything about today's climate? What if we now have better measurements than ever of everything that is believed to have changed the climate in the past? And tell me, was there ever a time when scietists and statisticians would have scoffed at taking a sample of 4.5 billion years to determine the effect of a process (human CO2 emissions) that has really only changed the atmospheric concentration significantly over the last century or so?

This is absurd. They said this:

Then there's this little-discussed fact about global warming: While the drumbeat of doom has grown louder over the past several years, the average global temperature during that time has in fact decreased.<\blockquote>

All the AP statisticians are saying,
"The experts found no true temperature declines over time." - Not what Levitt and Dubner are saying
"Saying there's a downward trend since 1998 is not scientifically legitimate" - Not what Levitt and Dubner are saying

They are saying that average global temperature has decreased in the past few years. Not that this is indicative of any trend. Not to say that the earth is cooling because of these few observations. I know people are critical of Levitt and Dubner because they feel they haven't gotten the science right and they haven't read the experts research. Well let me level the same accusations at their critics: Have you read any of Levitt's hundreds of peer-reviewed articles in prestigious economics journals? He knows statistics. Read some of his papers. It won't take you long to determine that he couldn't possibly confuse a few years worth of data on temperatures with a statistically significant trend.

Wait, he chose to say "the temperatures have decreased" ironically?

I'm reminded of the time Jennifer Morohasy put up two guest posts that had (mutually exclusive...) denialist theories in them that violated the laws of thermodynamics. She responded by posting, without comment, the definition of Socratic irony.

Isn't this just a slightly more sophisticated version of the '90s teen prank of shouting "not!"?
.
.
.

Mike: Regardless of whether or not he knows economics, Levitt admitted to eyeballing the temperature data in regard to the "cooling" line. This is How Not To Analyze Data in a nutshell, and amateur mistakes like this cannot be defended by saying "but he's an expert!" - especially since he's fessed up.

He is saying that it is interesting to note that, while concerns over global warming have intensified over the past few years, the global temperatures are lower in these years than in years in our recent history. Yes, he looked at the last few years on the charts and said "Oh, over the past few years, temperatures have gone down." There's nothing wrong with that provided you don't go on to say, "This must mean that there's no problem." Which he did not.

Looking at a few years in isolation is the precise wrong thing to do if you are trying to extrapolate beyond those few years. But I don't think he was trying to do that. Why? Because, "Levitt, a University of Chicago economist, said he does not believe there is a cooling trend." If his thesis was "There is a trend of declining global temperatures that will continue into the future, which is a reversal of the global warming trend of the past few decades," then I agree that he would have to present significant evidence to back up that claim, including some sort of statistical analysis of the time trend and a very good reason to think that there is something cyclical about global temperatures and that we are at the end of a cycle. Instead, he agrees that there isn't a cooling trend and says, but isn't it weird that temperatures have gone down even when we're talking about global warming, much as any scientist would pay attention to outliers.

In short, if you are trying to make the point that uneducated readers may see that and say, "Oh, that must mean he thinks the Earth is getting colder! Global warming is bullshit!" then okay, point conceded. If you are trying to argue that Steven Levitt himself believes that the Earth is getting colder because of a few years worth of data at the end of a long data series, then you are just plain wrong.

Here's the real question: how can two intelligent, truth-seeking individuals look at that quote and have such starkly different reactions to it? I see it and say, "Completely innocent quote. It's clear he's making an ironic point and doesn't truly believe in a cooling trend." James (and others) reads it and says, "Outrageous! How can he extrapolate a cooling trend from just the last few years worth of data?"

Mike,

I think we can all agree that your degree isn't worth the paper it is printed on.

Cheers,