Discussion > Two Hundred Million Dollar Scientific Grant Fraud Case
Robert Christopher. I have reviewed manuscripts for publication in scientific journals and, for a short period, was an associate editor who sent papers out for peer review. The main criteria used to judge papers in my particular field were 1) does the study present new data and/or interpretations? 2) is the study relevant to the Journal's usual content? 3) is the paper too parochial and potentially be of interest to only a few of the Journal's readership? 4) are the conclusions reasonable and fully supported by the evidence and discussion? 5) are other interpretations possible? followed by more technical matters (are all illustrations necessary? Can the text be shortened? Are references complete and in the correct format?). You will notice that there is no mention of data quality. It must be assumed that this is appropriate. A reviewer should not be expected to judge that. After publication is when those matters are usually discussed by those who might have their own evidence.
ACK
Interesting that in the USA there is a "bounty" for those exposing the fraudulent activity of colleagues, is this only for taxpayers money?
Mann cites various UK investigations that followed Climategate as finding him innocent, and those "investigators" got paid.
golf charlie
ACK ON Sep 5, 2016 at 1:52 PM
"You will notice that there is no mention of data quality."
Do reviewers even bother to review the data, that it exists, and will be available in a timely fashion? :)
For some papers, is it ever available? :) :)
"A reviewer should not be expected to judge that [data quality]. After publication is when those matters are usually discussed by those who might have their own evidence."
As I said, my thoughts were that a peer reviewer isn't meant to seek out fraud, but what if it is obvious that something is 'not quite right'? Is there any agreed procedure, any agreed options available to escalate the issue, that does not threaten the reviewer's existence? Does the reviewer have to remain silent, as a priest would having received confession, or decide whether to question it or not?
Many do think a 'peer reviewed' paper is some sort of guarantee. What of, I am not sure :) , but that illusion persists, particularly with those members of the public concerned and wanting to Save the Planet.
The UEA Climategate related investigations do not encourage anyone to come forward with 'issues', given the outcome.
Robert Christopher
Robert, I can only speak for myself, my particular area of science and my own experience. I have never engaged in anything to do with climate science but I suspect that before 1980 or so it was like almost any other area of science, except medical science. We do science because we find it interesting. We publish to get recognition for our endeavours and to help get funding to do more research. Some have higher aspirations, others lower ones (advancement and all that entails in terms of control and power). Reviewing other people's work is an onerous duty that takes up research time or time needed to write proposals.
The convention is that you may challange a person's conclusions, methodology or assumptions, but you should not question their data. It is assumed that the author is not faking data. You are correct, a reviewer's task is not to detect fakery. Of course if a reviewer has their own data that contradicts, that is another matter.
The position is I think different with modern climate science and medical science, simply because people's reputations and prestige are more involved (and in medical science the competition to succeed is so much more intense). Climate science has become so bound up with politics at all scales, that the usual checks and balances regarding publication are no longer sufficient. The dubious nature of much of the data and its manipulation, within what is essentially still a small field, changes everything.
Perhaps the scientific community can learn from how medical science regulates publications.
ACK
ACK, how does medical science regulate publications?
The Lancet is owned by a commercial publisher.
golf charlie
golfCharlie. Not too sure. I believe peer review is much more onerous - more competition. Also some publications only "publish" first online, giving time for objections and challenges to be given, before final approval and acceptance. I'm commonly hearing that this paper or that paper has been withdrawn or has not been approved, even after some time has elapsed since the paper's initial appearance. But I really don't know.
ACK
Robert. If I suspected plagiarism or outright fraud I would communicate this privately to the Journal editor. It should be their responsibility, together with the Journal's publisher. Fortunately this never happened to me but I often had to complain against shinglelism (self plagiarism). I have always insisted that, as a reviewer, my identity, should be communicated to the author(s). If I cannot stand up for my opinions and judgements, they are worthless. I have a reputation for being harsh, but I hope a fair reviewer.
I once recieved a letter from.an author whose papers I frequently was asked to review. It started by saying that on many occasions I had "put my axe" into his manuscripts, but then he thanked me for doing so. I treasured that letter for many years.
ACK
(Harry, at 10:37 AM in Unthreaded, I totally agree with you!)
Here is a report that ought to be brought to the public's attention, but will any PSB touch it? We will see :)
WattsUpWith That: Two Hundred Million Dollar Scientific Grant Fraud Case
Federal Prosecutors have launched a gigantic fraud case against Duke University, North Carolina, accusing Duke University of embezzling $200 million in federal research grants, by presenting doctored data with their grant applications.
The (first few, at least) comments are worth reading as well.
It goes right to the heart of the problem: how do those outside a specialised field check on the work within it. It is what peer review is wishfully thought to do, but doesn't, and I don't think is meant to do, even if it may stumble on 'errors', accidentally.
It isn't just the public and their representatives (our politicians :) ), even those knowledgeable in related subjects can find it difficult to determine whether something is amiss, let alone be in a strong enough position to blow the whistle.
Maybe we need an auditing function, that isn't such a large button to push as Whistle Blowing, to clean up Science, and State Sponsored Science in particular.