A few weeks back I blogged about a paper by Arturo Casadevall, Ferric Fang and others from the University of Washington and the Albert Einstein College of Medicine that investigated retractions in scientific publications and concluded that the majority of retractions could be traced to misconduct, with the majority of misconduct in turn arising from fraud.

Now in a recent study, the same authors hone in on some of the details of misconduct and unearth another interesting gem; they find that men are more likely to engage in misconduct compared to women. And yes, this is true even when you correct for the overrepresentation of men over women in academic research.

The authors look at 215 cases of fraud in life sciences research uncovered by the Office of Scientific Integrity (OSI). They were able to determine the gender of the perpetrator in all but one of the cases, and 65% of the wrongdoers were male.

What was also interesting was that the misconduct depended on the rank of the researcher; it seems that 88% of faculty members committing fraud were men, compared to 69% of postdocs and 58% of students.


This is perhaps not too surprising. Students and postdocs are much more worried about their future career prospects compared to tenured professors and less likely to jeopardize them. However the conclusion is not as obvious as it seems since pressure to publish or perish and other factors can also drive students and postdocs to engage in fraud; two prominent cases in the last couple of years make this clear. Based on their findings that fraud among faculty members is much more likely to be centered on men, the authors don't discount the possibility that female scientists might be committing fraud on an equal basis but may be escaping detection.

The real question of course is why men are much more likely to commit fraud. It's one of those classic nature vs nurture dilemmas with nurture seeming to play the dominant role. The authors cite studies reflecting higher crime rates among men compared to women, but cultural factors are likely to play a much greater role in the academic world. As the authors state, data on cheating among students is rather inconclusive with respect to gender, so it's likely that there's something about the peer-based research environment that contributes to this behavior. It's generally accepted that the cut-throat arena of scientific research often discourages women from pursuing research careers. Gender bias in hiring is increasingly being validated as a contributing factor. So is the inability of the academic system to properly take the burdens of pregnancy and childcare into account. But all this simply explains the paucity of women scientists at the highest levels of academia, not why they would be less likely to commit fraud.

Or does it? The authors hint at something which I think may be a clear reason for the reluctance of female scientists to commit fraud; the censure of the scientific community. The same pushback from a male-dominated scientific hierarchy that thwarts women at every stage of their scientific career likely makes them much more sensitive to criticism. With the result that they are far less likely to invite an unfavorable response from a predominantly male community by engaging in misconduct. Thus it might be female scientists' acute awareness of possible condemnation - ironically engendered by an unhealthy male-dominated environment - that might be keeping them more honest than the men. This of course does not mean that we let the unfavorable environment persist, but it does mean that we need to have other disincentives for fraudulent research.

This is a relatively small study that raises many more questions than it answers; what would the findings be for other sciences? Are there certain areas of research that seem to be more inundated with male fraud? Does the percentage of male fraud depend on other factors like nationality, tenure, and the funding situation? However, all these factors may be less important than possible solutions. The authors point out that current ethics courses are mainly targeted toward students and postdocs while the preponderance of fraudulent faculty members in the study points to a much greater need for instructing professors in the ethics of research and publishing. The data seem suggest a certain laxity in behavior that might accompany tenure and a stable academic job. At the same time the findings may again illuminate the intense pressure and battles for funding that often tempt academic scientists to stray from the righteous path. Ultimately, studies like this may put the spotlight more on the dysfunctional aspects of our current academic research system rather than simply on gender bias.


Males Are Overrepresented among Life Science Researchers Committing Scientific Misconduct

Ferric C. Fang and Joan W. Bennett and Arturo Casadevall, mBio, 4, 1, e00640-12