American Political Science Review Issues "Expression of Concern" Following Karlstack Investigation
"the dog ate my dataset!"
Note to my several hundred new subscribers — while I often write about politics/economics, this Substack actually began as an outlet to investigate fraud and corruption in academia. If you don’t care about academia, please note that Karlstack is split up into 5 sections: Economics, Academia, Politics, Crypto and Personal. To opt-out of this “Academia” section you can click the “Settings” button (at the top right corner of your screen) and then click “Manage Subscriptions.”
This is a short update to my article from 6 months ago, “The American Political Science Review Goes Woke:”
The complete expression of concern is linked here. It says:
"The Editors of the American Political Science Review would like to alert readers that the article “The Tragedy of the Nomenklatura: Career Incentives and Political Radicalism during China’s Great Leap Famine” may contain errors, including inaccurate reporting of model specifications for some estimates in Tables 4 and 5 and incorrect compilation or calculation of indicators, that make the core findings unreliable. At the time of publication, APSR editorial policies did not require data sharing in a public repository, and according to the authors, the original data are unavailable."
This investigation was kicked off when I hounded the editors, and they eventually replied to me:
The issue with this paper is straightforward. It claimed to have included fixed effects in the regressions, but did not actually include fixed effects. They blatantly lied about including fixed effects. Running the regression as reported completely kills the findings.
What I don’t understand is why exactly the paper was not retracted?? It is fraudulent, and the data is now conveniently unavailable — another case of “the dog ate my data!” So frustrating. A clearly fake paper with clearly fake results, and the best they can do is issue a feckless “expression of concern.”
No accountability. Nobody cares that one of the authors (James Kung, professor at University of Melbourne econ department with a PhD from Cambridge) is the most prolific cheater in the profession (see this 38+ page EJMR thread, “James Kung: A Serial Manipulator”), and the other author (Shuo Chen, professor of economics at the University of Fudan) has also been caught several times with his hand in the cookie jar (see this Karlstack article, “Shuo Chen & Yiming Cao: The Men Who Made History“).
What it boils down to is that the the academic discipline of Political Science just doesn’t matter. At all. If you fake a medical paper, people die. If you fake a political science paper (or dozens of political science papers), you get tenure.
It should be noted that the APSR is the #1 ranked Political Science journal in the world. This would be like if a medical doctor published a totally fraudulent paper in The Lancet, and the best The Lancet could do is issue an expression of concern while leaving the paper unchanged and still published.
More to the point of the article, it is becoming increasingly apparent that peer review does not work for finding errors in academic research. Refereeing a manuscript is primarily about rendering an aesthetic judgement.
https://experimentalhistory.substack.com/p/the-rise-and-fall-of-peer-review
Glad your work is reaching a larger audience.
My spouse reviews many medical research papers and grant proposals. His opinion is that people are becoming much more sloppy with interpretation and analysis of data. He asks the hard questions and this is not always popular.
Go after those fraudsters. Get them. The integrity of science and economics depend on people who are willing to do that , who do not need to fear their own grants will be unfunded in retaliation.