Skip to content


I’m just glad that universities don’t sanction professors for publishing false theorems.

If the guy really is nailed by the feds for fraud, I hope they don’t throw him in prison. In general, prison time seems like a brutal, expensive, and inefficient way to punish people. I’d prefer if the government just took 95% of his salary for several years, made him do community service (cleaning equipment at the local sewage treatment plant, perhaps; a lab scientist should be good at this sort of thing, no?), etc. If restriction of this dude’s personal freedom is judged be part of the sentence, he could be given some sort of electronic tag that would send a message to the police if he were ever more than 3 miles from his home. But no need to bill the taxpayers for the cost of keeping him in prison.


  1. arik says:

    I feel kind of bad for him. The academe punishes misconduct far worse than the private sector does. If you get caught for insider trading or tax evasion, you'll move on at a different firm. But an academic will have trouble ever getting a job again.
    I think of Michael Bellesiles ( who went from award-winning Emory professor to part-time lecturer at Central Connecticut state after the integrity of one of his books on the second amendment was brought to question by the NRA.

  2. Roger says:

    Harvard is not even firing him, so it is not clear that he has committed some sort of criminal fraud. It may just be some sloppy work.

  3. Frank D says:

    Does anyone out there in Gelman blog land have any insight on this? The article is extremely vague.

  4. Hamdan says:

    This is really harsh, Dr. Gelman. From the media coverage, it seems much of this involves coding of how primates respond when presented with various stimuli – something I think is naturally more subjective than we are used to in the quantitative sciences. Sloppy science, however, isn't necessarily the same as an intent to deceive.

    There doesn't seem to be any fabrication of data, which I think would be the clearest definition of misconduct.

    Along the same lines, what do you think of the Anil Potti case at Duke (and the rebuttals by Baggerly et al.)? Where do you draw the line between fraud and faulty statistical analysis?

  5. Jeremy Miles says:

    Here are some more details:

  6. Andrew Gelman says:


    Read carefully. I didn't pass judgment; I wrote, "If the guy really is nailed by the feds for fraud." My speculations about prison, community service, etc., were all conditional on this "if."

    I gotta say, though, it's not such a good sign when a guy writes a book called, "Evilicious: Why We Evolved a Taste for Being Bad."

    Finally, I don't know anything about Anil Potti. I only heard about this Harvard case because I came across the news article.

  7. zbicyclist says:

    @hamdan: "There doesn't seem to be any fabrication of data, which I think would be the clearest definition of misconduct."

    The NYT article
    says “scientific misconduct involved problems of 'data acquisition, data analysis, data retention, and the reporting of research methodologies and results.'”

    To me, that sounds like a euphemism. One might have said that same thing about Cyril Burt.

    The fact that Hauser was able to replicate two of the three experiments says he's been given wide latitude by Harvard — time enough to replicate the experiments, and the implication seems to be that he replicated them, NOT somebody else.

    That's sort of curious. When I've suspected employees of sharp-penciling an analysis, I've never had THEM replicate it.

  8. K? O'Rourke says:

    zbicyclist – nice point – as JC Gardin use to put "investigators have no business replicating their own claims"

    Lots of grey areas here, from some past dealings (in decreasing order)

    using a supportive reference believed to be false or faulty

    writing up the methods section in a way that hopefully the reviewer would miss the critical deficiencies

    not grasping a mathematical point but proceeding anyways

    failing to double check calculations

    But to put people in high pressure low resource occupations where they know there will seldom be any real checks on their work, does argue for the need for real penalties if they ever do get caught.

    In some ways I was lucky – the first major data analysis project I was involved in the primary investigators had a colleague who was dropped from the possible Nobel laureate list we his data analysts left and the work could not be extended or even replicated. So they encouraged much more careful programming and documentation.


  9. John says:

    From the comments above:

    It may just be some sloppy work.

    Sloppy science, however, isn't necessarily the same as an intent to deceive.

    There doesn't seem to be any fabrication of data, which I think would be the clearest definition of misconduct.

    From a Boston Globe article:

    Harvard’s policy describing its procedures for responding to allegations of misconduct in research notes that “research misconduct does not include honest error or differences of opinion.”

Where can you find the best CBD products? CBD gummies made with vegan ingredients and CBD oils that are lab tested and 100% organic? Click here.