New at Reason

Julian Sanchez | December 30, 2005

Ron Bailey takes a look at some recent scientific frauds—and how science's liberal institutions catch them.

Comments

|12.30.05 @ 1:35PM|#

Of course, peer review has come out about as strongly in favor of anthropogenic global warming as is possible, and yet, Ronald and his ilk...

Ken Hagler|12.30.05 @ 1:46PM|#

Like it says in the article, "Peer review is designed to catch errors, not fraud. Consequently, scientific fraud is often caught only after peer review publication."

|12.30.05 @ 1:52PM|#

There's actually a preprint archive that many physicists use. Articles get posted there and discussed, and not ll are submitted to peer reviewed journals. I've stayed away from it because most of my immediate colleagues don't use it either, and it's not as popular in the subfields where I've worked.

To me, peer reviewed journals are fine. They serve an important purpose in academic science, especially when you consider that there's a hierarchy of journals. You do something, and now you want recognition for it. So you take it to a peer reviewed journal. You ask them to certify that your work is as meritorious as the other works published in their journal. The experts examine it, and if they concur then it's published.

Basically, it's a form of GPA for practicing scientists.

And yes, I recognize that peer review can't catch most fraud. But peer review isn't really designed for that. It's mostly designed to evaluate merit. It may not do an ideal job of that, but let's at least evaluate this mechanism based on its purpose, not some other purpose that we might wish it served.

|12.30.05 @ 2:11PM|#

"So how does peer review work? Typically, what happens is that a researcher sends in a manuscript to a journal describing her recent work."

Oh, how grammatically PC!! :-)

Great article, though peer review can't overcome mediocrity and mythology that has already reached the level of conventional wisdom.

Dave W.|12.30.05 @ 3:16PM|#

I am not really clear on the motivations of the scientists who caught the fraud. Were the one's that "blew the whistle" motivated by:

(a) science, scientific method, quest for truth, etc, etc;

(b) investment money, eg, diligent investor-scientists mindingtheir investments in Hwang's companies and/or their competitors;

(c) legal liability fears, eg, fear of shareholder suits, egg donor suits, fear of consume suits.

(d) fear of gov't regulation

Mr. Bailey's article seems to assume it is (a). My guess is that it was mostly (b), which is a different thing than (a). Hopefully someday I can read an account of who exposed the fraud and exactly why.

Dave W.|12.30.05 @ 3:51PM|#

You are probably correct, T., but this is one of those situations where I would rather know than guess. I mean if it was some scientist who got put on the project because his boss was worried about SOX requirements, then that would be a good thing to know, because it would force us to re-evaluate. If it is truly a victory for the liberal institutions of science, then that would be great, too.

I just want credit where credit is due and only where credit is due.

|12.30.05 @ 3:52PM|#

To explain why it's difficult to turn in a colleague, let's leave aside the fact that nobody likes bad blood. There's also the fact that the people in charge of ethics and safety policies in most research institutes are not practicing scientists. You go to these obligatory seminars, and you hear them say things that may not be wrong per se, but definitely betray a total ignorance of how science works. And you know that if a colleague does something wrong, they'll come down on everybody around him as well. Everybody's work will be scrutinized by people who are clueless.

Nobody wants that sort of drama, but nobody likes cheaters either. So you want to be sure before you accuse somebody. Which is fine: Extraordinary claims deserve extraordinary proof. But the fact that there are risks involved with making an accusation suggests that scientists who expose fraudulent colleagues must have at least some sort of altruistic motive. Otherwise it wouldn't be worth the hassle.

Ron Hardin|12.30.05 @ 6:03PM|#

There's formal peer review, and there's working peer review.

The latter type actually works, and it's when stuff falls into the hands of some guy who's actually curious about the work (huh, that's odd), and so likely to catch stuff you want caught.

The former type, as with global warming, is sending out work that is of no interest, everything a reference to some other work of the same type, getting the party line straight.

You see no global warming papers starting at the beginning and going to the end, in a way that might interest an actual physicist (but then it's not physics). Yet it's peer review.

|12.30.05 @ 10:53PM|#

You see no global warming papers starting at the beginning and going to the end, in a way that might interest an actual physicist (but then it's not physics). Yet it's peer review.

I haven't followed the global warming literature. Is it really that bad, in your professional opinion?

|12.31.05 @ 12:00AM|#

A straightforward radiative transfer calculation applied to constant atmospheric properties clearly demonstrates that an increase in the concentration of carbon dioxide will cause the temperature of the surface of the Earth to rise. The difficulty is that atmospheric properties will not remain constant under such warming, and the total warming (direct CO2 + atmospheric response) is highly uncertain. But that doesn't make it fraud.

|12.31.05 @ 12:15AM|#

You see no global warming papers starting at the beginning and going to the end, in a way that might interest an actual physicist (but then it's not physics).

Ah, spoken like a true upper-case-P Physicist! I, on the other hand, am a lower-case-p physicist. The difference between the two is that the upper-case Physicist thinks that physics research happens only in Physics departments whereas the lower-case physicist uses physics to solve all sorts of interesting research problems in all sorts of other fields.

The real reason why Physicists are not interested in global warming that the Earth's climate is an extremely complex system and Physicists do not know how to work on problems that are not easily amenable to reductionism. But we welcome all lower-case physicists!

|12.31.05 @ 1:12AM|#

Late to the thread as usual, but this thread looks like an excellent excuse to link to this:

Electron Band Structure In Germanium, My Ass

|12.31.05 @ 7:36AM|#

Perhaps One-Trick Pony can simply sue the whistle-blowers for intentional infliction of emotional distress -- not knowing something sure seems to be causing him some -- then find out everything during discovery.

Well, why not? It's how he proposes learning pretty much everything else.

|12.31.05 @ 9:21AM|#

joel-

Always good to see a fellow physicist here! What do you work on? I work in medical research.


dead elvis-

That lab report is absolutely fucking hilarious!

|12.31.05 @ 1:00PM|#

Pleased to meet you, thoreau.

Actually, I work on climate. It's a tough research problem at the intersection of turbulence, radiative transfer, and phase changes of water and spans spatial scales from 10e-7 m to 10e7 m. Moreover, the answers are needed at the precision of 1%, so the traditional physics order-of-magnitude approximations are not sufficient. These days I am primarily doing statistical analysis of various historical climate observations so I suppose you might say I am more akin to an astronomer than to a physicist.