Criminology researcher to lose sixth paper

via Tony Webster/Flickr

A criminologist whose work has been under scrutiny for a year is set to have a sixth paper retracted, Retraction Watch has learned.

Last July, Justin Pickett, of the University of Albany at the State University of New York, posted a 27-page explanation of why he was asking for one of his papers to be retracted. The paper in question had been co-authored by Eric A. Stewart, a professor at Florida State University, whose work had been questioned by an anonymous correspondent.

Following pickup of the story by The Chronicle of Higher Education, that paper was eventually retracted, along with four others Stewart co-authored. But that was not the end of the tale. 

Stewart is now having another paper retracted, this one from 2003 in Justice Quarterly. The paper, titled “School social bonds, school climate, and school misbehavior: A multilevel analysis,” has been cited 186 times, according to Clarivate Analytics’ Web of Science.

Last fall, scientific sleuths Nick Brown and James Heathers analyzed the paper using their RIVETS — for “Rounded Input Variables, Exact Test Statistics”: A Technique For Detecting Hand-Calculated Results in Published Research — tool and found a positive result. Such a result, they write,

cannot, by itself, tell us anything definitive about what may have led to it. However, when such positive tests are encountered, it may be worthwhile to contact the author of the article in question and ask for details of the method used, and perhaps also a copy of the data set so that the method can be used to reproduce the published results.

Pickett wrote to the editors of Justice Quarterly in January of this year, saying that the publicly available data Stewart claimed to have used did not match what was reported. As he told Retraction Watch:

The editors immediately contacted the publisher, and set out to do a thorough and professional investigation, which they did. They used three outside reviewers who reanalyzed the data. They kept everyone updated. It was all quite impressive. I wish all editors and journals were this serious about the validity of what they publish.

In a June 8 letter to Pickett, Zoe Sternberg, head of Taylor & Francis’s criminology portfolio, writes that

The referees found that the article contains numerous errors in the reporting or analyses of the data. None of them could replicate Dr. Stewart’s analyses, and they further expressed concern that Dr. Stewart was also unable to replicate his analyses. Consequently, the referees questioned the validity of the results reported in the article. It is not possible, in our view, to reach a conclusion as to whether misconduct occurred.

Sternberg wrote to Pickett that

We will send the retraction statement to our production team today, and they will prepare it for publication shortly.

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].

3 thoughts on “Criminology researcher to lose sixth paper”

  1. I hope that I am allowed to post a link to our RIVETS preprint here: https://psyarxiv.com/ctu9z/

    I thought of the principle behind RIVETS some while back, but as far as I know it’s the first time that anyone has used the principle of reported statistics *not* suffering from rounding error to identify problems in a published article. It has some faint echoes of John Carlisle’s test for implausibly small baseline difference in RCTs.

    1. I’m no statistician but I find that a fascinating idea. I’m curious if any more papers have been flagged by your RIVETS.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.