Fraud: My story in tomorrow's New York Times

A new look at retracted papers since 1975 paints a picture that’s none too pretty. Retraction rates are zooming up, and most of those retractions, a new study finds, are due to misconduct such as fraud and plagiarism. I write about the study in tomorrow’s New York Times. Check it out.

6 thoughts on “Fraud: My story in tomorrow's New York Times

  1. Can’t help but notice that the papers looked at were all from the “biomedical and life sciences” where one might expect a significant amount of fraud; would be interesting to know how the ‘hard’ sciences, physics, engineering, computer science, etc. as well as the ‘softer’ sciences, psych., sociology, anthro., etc. would compare… And of course the fraud that is detected and retracted is likely only a tip of the iceberg of what exists.

  2. @ Shecky R: That would be really interesting. This is especially true because the biomedical and life sciences are more complex and therefore more difficult to study (e.g., cause and effect of a pool ball collision versus cause and effect of cigarettes causing cancer). Also, we probably have more conviction to subjects that have bigger impacts on the human condition like biomedical sciences.

    Not reading any of these papers YET, in my opinion, I would attribute this to moral turpitude to: commodifying research through competitive processes. It is not the field of these sciences per se, but rather, it is being on the field scientists are expected to play on determined by somebody else’s rules (e.g., “Produce product X,” “Publish in Y,” “Publish in N journals”). The rules are both emergent (outcomes of living in a population) and directly determined (leaders of government and other financiers, research institutions, and publishing outlets). We just need to slow down and work on recapturing some integrity. Just some thoughts.

  3. One of the things I noted in the 20 years I worked in biomedical research was the increasing emphasis on getting published, and its use as a measure of “productivity.” When I started, a paper every other year or so was considered “good,” with maybe a technique paper thrown in there for good measure. By the end, it was a constant run to put together another paper or abstract. While we didn’t falsify data, I wasn’t particularly happy with the caliber of papers we were turning out. I can see how that pressure would start tempting – and some succumbing – people to “fill in the blanks” or “massage” the data to get a paper out of it.

  4. Carl, did they consider the possibility that there is more recognition of misconduct (rather than or in addition to actual misconduct), as for example much of the rise in autism diagnosis has been attributed to changing diagnostic criteria or additional attention?

  5. So far I just skimmed the original PNAS article, but I noticed a mistake in your NY Times piece. You lead with a 10-fold increase in retractions in the past decade, but I think the article is showing a 10-fold increase since 1975.

    Just skimming the article, it’s difficult to tell how much of the increase in retractions is simply due to more research groups and more journals publishing more papers. I don’t think this accounts for all if the increase, but it should account for a non-trivial amount and it doesn’t seem like the article accounts for this in any way.

    One interesting aspect, which is stated with supplemental figure 3 is that “Thirty-eight laboratories with greater than or equal to five retractions accounted for 43.9% (n = 390) of retractions for fraud or suspected fraud (n = 889).” Smaller-scale fraudulent labs clearly exists, but I wonder if the real rise in fraud is actually driven by these big cases.

Leave a Reply

Your email address will not be published. Required fields are marked *