How is scientific fraud investigated?

With aspersions cast on the work of the Hadley CRU, I’m wondering how an alleged scientific conspiracy of this scale gets investigated. Who would investigate them? And how? Other climate scientists are tainted, either by association or by opposition. My search-fu is weak as it only turns up investigations of single fraudsters.

And do other researchers have to disregard the suspect data until the authors are vindicated? I’m a firm believer in the presumption of innocence, and cock-up over conspiracy, but does the scientific world so believe?

The standard way would be to take a published paper and show that it is fraudulent (or non-repeatable) in another paper. There are no science police. Presumably the fraudsters would be fired from the University or legitimate research institute.

On some occasions, if an academic (in any academic field) is accused by reliable sources of faking data in his papers, the university will appoint a committee to examine the accusations. If the claims are found to be true, the university will often fire the academic. This is apparently part of the contract in many universities. Faking data in any academic field (not just science) is a firing offence.

Yes, I’ve seen that, but in this case it’s alleged to be (and I maintain the presumption of innocence here) a lot more than one scientist. And it’s a very contentious subject.

It should be noted that this is a hypothetical question - in this particular case there is no suggestion whatsoever of academic misconduct. None of the alleged quotes point to withholding data, misrepresenting data, “suppressing” contrary results or misrepresenting the work of others.

The only quote that actually even suggests dodgy behaviour is boycotting “Climate Research” as a journal because they don’t feel that its editorial policy (multiple editors, any of which can allow papers through to publication) guarantees proper peer review. Avoiding dodgy journals is standard academic practice, although I can understand researchers with less well accepted ideas feeling that “peer review” is equivalent to holding a closed shop.

In the case of climate research though, I have a strong belief that skeptic research gets rejected because its badly conducted research, not because it doesn’t support the consensus view.

Accusations of scientific fraud would typically be investigated by the academic institution to which the researchers belonged and/or the journal in which the results were published. If fraud were demonstrated, it would typically result in the firing or resignation of the researchers, and the retraction of any papers containing fraudulent data or conclusions. If it involved the misuse of public funds or grant money, the donor agency or organization could bring a suit to recover the funding or bring other charges.

Nobel laureate David Baltimore was involved in a case involving accusations of fraud against his collaborator Thereza Imanishi-Kari. The case was investigated by NIH, since it had provided funding, and also the US Secret Service as a branch of the Treasury. Imanishi-Kari was originally found guilty of research misconduct, but on appeal the charges were determined to be not proven.

There’s one e-mail. That means one guy sending it, and one guy receiving it. While two scientists is more than one scientist, I fail to see how it’s “a lot more”. I fail to see how anyone else is “tainted” by this.

Alas, I have seen at least one email which suggests precisely that. I’m not going to link to it until I get the all clear in this thread. But for this thread, I’m not really interested in the details of those particular emails but rather the process of what happens now in the general sense. Say rather, a large group of anthropo-archaeologists were similarly accused (Piltdown Man writ large). Or particle physicists…

Again, I have read about what happens when one person is accused. I’m interested in what happens when a group of people are accused.

I’ve not seen anyone with standing actually make allegations yet.
It remains to be seen whether this ‘case’ will ever rise to that level.

Digesting what everyone else said, the answer is that it is up to any group which has the power to do anything about it and which believes there is merit to the claim. How they go about handling it is up to them.

In the case of CRU, this could mean that Congress itself forms a team to examine it, or that some smaller agency like the NIS or CRU itself handles it. Though again, it requires for someone to think there is merit in investigating the topic first. But who it will actually be is going to be ambiguous in any case until someone with the clout to do so, does so.

Nothing different. Each case would be handled individually. Why would you think otherwise?

I was curious that there might be an otherwise, and was interested in what that otherwise might be.

I apologize for my previous post done before I read the link in the OP. I don’t see that there is clearly anything even to be investigated. The article that’s linked to gives us nothing except some random quotes from the E-mails. First, someone is going to have to post the entire set of E-mails. Then someone is going to have to go through all of them and make a plausible case that there is any fraud going on here. They would have to explain who is doing the fraud and what data is being faked. It would then be the university’s job to appoint a committee to investigate the fraud. At most, all it would show is that there is fraud at a particular research institute. The case for global warming does not depend on the research of one particular group. This article is written by someone who is anti-global warming anyway. He claims that the infighting at this institute somehow destroys the entire case for global warming.

Sample sent in PM.

Since this is GQ, the use of the Hadley CRU emails is not relevant to the hypothetical question. There is nothing yet revealed that would constitute academic misconduct at most research institutions.

As to the hypothetical, faked research tends to fall under “deception” rather than fraud. There is no direct monetary gain, so it is limited who can investigate and punish any offender(s). The employer (typically a university or research centre) certainly can, since it would be in breach of an agreed code of conduct. The publication venue typically can’t. If the faked research was used to apply for a grant or other money, then the source of the money has a civil legal claim, and there is a possible criminal charge (although I am unaware of any actual instances of criminal charges in Western countries over misconduct in grant applications - the employer has a lot of motivation to punish and promise the grant body that they’ll be strict with future employees).