Andrew Gelman is not the plagiarism police because there is no such thing as the plagiarism police.

The title of this post is a line that Thomas Basbøll wrote a couple years ago.

Before I go on, let me say that the fact that I have not investigated this case in detail is not meant to imply that it’s not important or that it’s not worth investigating. It’s just not something that I had the energy to look into. Remember, people can be defined by what ticks them off.

And now here’s the story. I got the following email from someone called Summer Madison:

I think this might interest you:

http://www.econjobrumors.com/topic/new-family-ruptures-aer-nber-is-rip-off-of-obscure-paper

I replied:

Hi, thanks for sending. Doesn’t quite seem like econ to me, except to the extent that econ is taken to include all causal inference for the social sciences.

Summer wrote back:

I was hoping you might consider posting on your blog re: the plagiarism issues raised by the article.

You might want to look at the link again — the paper is causing quite an uproar on EJMR.

I should add that if you read through the entire thread, you will see that a number of posters at EJMR have expressed the hope that you would comment on the plagiarism issues raised by the article.

From EJMR:

“First, to the best of our knowledge, our study is the first to document a causal link between fetal stress exposure and mental health in later life.”

Not cited:

http://archpsyc.jamanetwork.com/article.aspx?articleid=491893

http://www.bmj.com/content/348/bmj.f7679.full

http://archpsyc.jamanetwork.com/article.aspx?articleid=482586

http://www.sciencedirect.com/science/article/pii/S0022395610003468

Also not cited but uses the exact same data set:

http://www.ncbi.nlm.nih.gov/pubmed/21321257

So the paper is promoted as a breakthrough in terms of new data and new question, but it is neither.

It looks pretty bad to me (but I have to admit that I did not look at the original articles, nor did I even read the whole Economics Job Rumors thread). So I’m making no statement on whether I think this really is plagiarism, how bad it was, what should be done about it, etc.

Instead I want to explore the question, Why to some cases of plagiarism or, more generally, unscholarly behavior, get me angry, while others amuse me, and others (such as the case above) don’t pique my interest at all?

These get me angry:

– That political scientist Frank Fischer who attacked the person who called him on his plagiarism.

– Matthew Whitaker, who ripped off the city of Chicago by selling them materials written by somebody else.

Mark Hauser.

Weggy.

Weick.

These get me amused/annoyed:

– Bruno “Arrow’s Other Theorem” Frey.

– That spy novelist who put together a book ripping off over spy novels.

Chrissy.

This one I can’t quite bring myself to care about:

– The econ example above.

Why does this example not interest me? Maybe because it’s not in either of my main academic research fields (statistics or political science), it doesn’t have the amusing twist of the story of Frey (who, you might remember, had written a lot about motivations of researchers to cheat), and no monkeys were harmed during the course of the research.

I’m not saying my attitudes on these different cases are appropriate, and I’m certainly not trying to defend (or, in this case, adjudicate on questions of) unethical research practices. I’m just trying to give some sense of why I’m writing about some of them and not others.

Part of transparency is being open about what we decide not to write about.

P.S. I posted this awhile ago but it’s still not scheduled to appear for a few months. In the meantime I got an email from someone else which I’ll just append here:

Have you been made aware of the accusations of plagiarism of a forthcoming article by Petra Persson and Maya Rossin-Slater (PP-MRS) in the American Economic Review (AER)?

Retraction Watch has a story about the accusations

Economists go wild over overlooked citations in preprint on prenatal stress

The article in question originally claimed to be the first to estimate the causal effect of maternal stress during pregnancy on the child’s mental health later in life. The paper is here

http://www.nber.org/papers/w22229

Some economists noted that this article was not the first to find such a link. How so? Scholars in medicine way back in 1978 already estimated this causal effect. Moreover, economists noted that a psychology study from 2011 also estimated this causal effect. This latter study used the same data set as the PP-MRS study.

PP-MRS responded by adding footnotes to the forthcoming AER paper, maintaining that their study is novel because it is the first to estimate causality, while the 1978 and 2011 studies merely estimate correlations.

Other economists have responded to this rebuttal with two critiques.

The first critique is by Brett Matsumoto, a research economist at the Bureau of Labor Statistics:

https://www.dropbox.com/s/8t7k906ze2mfkk7/comment.pdf?dl=0

This critique is formal and focuses squarely on the econometrics of the PP-MRS paper, casting doubting on PP-MRS’s claim to be the first to estimate a causal effect.

The second critique is a crowdsourced paper written by various anonymous economists writing under the pseudonym Nicolas Bearbaki (a play on ‘Nicolas Bourbaki’). The reason for the anonymity is for fear of professional retribution.

https://mpra.ub.uni-muenchen.de/71699/

This critique is formal (less so than the Matsumoto paper), but is more blunt in its accusations against PP-MRS. This critique also mentions the editorial misconduct at the AER, as the handling editor (Hilary Hoynes, as mentioned in the Retraction Watch article linked above) is a recent coauthor of MRS. Such coauthorship is a violation of AER’s conflict of interest policy.

I, along with scores of economists in the shadows, would greatly appreciate if you could weigh in on this controversy. Many economists are scared to speak out against the AER, while many journalists generally do not understand the intricacies in estimating causal effects.

I responded that I don’t really have the energy to “weigh in on this” in the sense of looking at the details, but I’m happy to post something on it just to get the discussion out there. If the criticisms are indeed correct, I can really understand the frustrations out there.

P.P.S. This story continues to develop during the months between writing and posting. Today (1 Jul) I got another email from Summer Madison pointing me to this post from economist George Borjas who expresses concern about insider bias in the peer review process for the article in question.

7 thoughts on “Andrew Gelman is not the plagiarism police because there is no such thing as the plagiarism police.

  1. The biggest problem with the PP-MRS paper is not plagiarism, but corruption in the peer-review process. The editor who managed this paper (Hoynes) had recently coauthored with one of the authors, which is a violation of AER policy. This editor also didn’t send the paper out for further reviews once the extensive overlap with existing work was revealed. Presumably reviewers might not favor publication of a minor twist on existing work in the flagship journal. However, the editor simply allowed the authors to insert some footnotes referencing the older work, and proceeded to publication.

  2. Andrew,

    Earlier this week, you said in another post (about Fiske) to “remember that, for each of these people, there may well be three other young researchers who were doing careful, serious work but then didn’t get picked for a plum job or promotion because it was too hard to compete with other candidates who did sloppy but flashy work that got published in Psych Science or PPNAS.”

    The same is true in the current case. I think this partially explains the outrage on EJMR, particularly among young scholars. I’m not an economist, but I was outraged. I appreciated the widespread negative reaction on EJMR. It was great to see that strong norms opposing such behavior existed among so many scholars. I was incredibly impressed by the crowdsourced response! Truly, that has to be a first in the history of academia. Then, many of us looked to you to reaffirm and legitimize this reactionary sentiment, because you have been one the foremost interdisciplinary leaders in the movement to improve research norms. From articulating how to appropriately respond to criticisms of your own work, to calling out plagiarists (Hurwitz and Peffley), you have always led the charge. At least on my end, I hate that it didn’t happen that way this time.

    • Justin:

      I can well believe this is a big deal in the field of economics. I do not at all intend my brief post above to be any sort of de-affirmation or de-legitimization of the concerns you discuss. I just don’t know enough about the details here, and I couldn’t motivate myself to study this particular case in detail. There’s only one of me! And I have research to do, classes to teach, books to write, and blog comments to respond to! (That last might seem unnecessary but I learn a lot from these exchanges.) I encourage others to look into the matter. For example, Retraction Watch has a whole team, and I think they’re a trusted source.

  3. As a small fish in a very large scientific ocean, I’ve experienced the exact opposite: having reviewers criticize my claims of originality by refusing to recognize the specific moderators in my models (e.g., “so-and-so [possibly the reviewer?] has already studied the relationship between X and Y,” when my hypotheses dealt with the relationship between X and Y moderated by Z). Not sure if they just didn’t bother to read the paper in detail or if there is a more insidious COI going on.

    Of course, I tend to think that originality is overrated, but it should be noted that there are two sides to this particular coin.

  4. This is just an extreme case of what is my biggest pet peeve with the economics field right now. As the causal inference wave leads economists further afield into new substantive topics that are outside the traditional topics examined by economists into terrain typically held by political scientists and sociologists and historians, etc, there is an incredibly arrogant belief among some economists that its not necessary to acknowledge that other people have already studied these topics and already came up with similar answers to the same research questions.

    This can be very frustrating. You’ll see NBER working papers claiming to be the first to do XYZ, and if you’ve ever bothered to look into the other field’s literature you’ll see that there is already a decades long debate in political science or sociology, etc, on XYZ, which has already extensively discussed the economists’ supposedly novel theoretical argument. And some of that other literature is already quite methodologically sophisticated, so its not just that these political scientists can be dismissed out of hand for their “inferiority”. But because the economists are submitting to a different set of journals, with a different set of reviewers, who also don’t know that poli sci or sociology literature well either, they often can get away with this. Then the resulting AER publication starts getting cited more than the original articles, because AER is just much higher profile across the social sciences.

    Being higher status helps, I guess.

Leave a Reply

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