Failure in the self-correcting system? Maybe not.

The NYT magazine has a fascinating article by Janeen Interlandi on the case of Eric Poehlman, a medical researcher at UVM who was discovered faking his data by a student at his lab. Eventually he was sentenced to 1 year and 1 days jail time for scientific fraud, but he'd been getting away with it for years - indeed he was hired to UVM based on work he'd faked and seemed never to have stopped faking data until he was investigated.

The scientific process is meant to be self-correcting. Peer review of scientific journals and the ability of scientists to replicate one another’s results are supposed to weed out erroneous conclusions and preserve the integrity of the scientific record over time. But the Poehlman case shows how a committed cheater can elude detection for years by playing on the trust — and the self-interest — of his or her junior colleagues.

... Not only does any research touched by tainted data have to be re-examined, but high-profile cases of misconduct can also shake public confidence. “We already have a large subculture in society of people who don’t trust science to begin with,” says John Dahlberg, one of the Office of Research Integrity investigators who oversaw Poehlman’s case. “This doesn’t help at all.”

Peer review and repeatability in science are self correcting, but deliberate fraud is harder to guard against. If you get a result that disagrees with a colleagues you don't immediately think "fraud" - you check your analysis, try to see how they might have analyzed or interpreted the data differently. It might even be a typo in their manuscript, or put in the journal. And thats if you're using the same data, often you're not. Deliberate fraud of the "making up data" variety is probably rare, but this case shows that it is caught... eventually. The self-correcting part of science is necessarily speedy.

Another interesting part of the article is when it asks how did he get away with it for so long?

The length of time that Poehlman perpetrated his fraud — 10 years — and its scope make his case unique, even among the most egregious examples of scientific misconduct. Some scientists believe that his ability to beat the system for so long had as much to do with the research topics he chose as with his aggressive tactics. His work was prominent, but none of his studies broke new scientific ground. (This may also be why no other scientists working in the field have retracted papers as a result of Poehlman’s fraud.) By testing undisputed assumptions on popular topics, Poehlman attracted enough attention to maintain his status but not enough to invite suspicion. Moreover, replicating his longitudinal data would be expensive and difficult to do [emphasis mine].

“Eric excelled at telling us what we wanted to hear,” Matthews, Poehlman’s former colleague, told me.“ He published results that confirmed our predisposed hypotheses.” Steven Heymsfield, an obesity researcher at Merck Pharmaceuticals in New Jersey, echoed Matthews’s sentiments and added that Poehlman’s success owed more to his business sense and charisma than to his aptitude as a scientist.