carolo7/iStockphoto
Criminology researchers are retracting five studies that have sparked a bitter battle over potential scientific misconduct and issues of race. The episode has riveted the criminology community—and severed a once close relationship after one of the researchers accused his former mentor of falsifying data.
On 10 November, Justin Pickett, a criminologist at the State University of New York in Albany, announced on Twitter that he and his co-authors have agreed to retract a 2011 study published in Criminology that examined public support for taking a suspect’s ethnicity into account at sentencing. Four additional disputed papers, published between 2015 and this year in the journals Criminology, Social Problems, and Law & Society Review, have been or are in the process of being be retracted with the agreement of all the authors, ScienceInsider has learned. Eric Stewart, Pickett’s former mentor and a criminologist at Florida State University (FSU) in Tallahassee, is a co-author of all five studies.
The studies being retracted cover a range of topics. Two found that the number of black people lynched in a U.S. county 100 years ago influences whether white people in the same area today perceive black people as a threat and favor harsh punishments for them. Another examined the role of social context in antiblack and anti-Latino sentiment in the U.S. criminal justice system.
The upcoming retraction notice for the 2011 Criminology study—which Pickett shared with ScienceInsider—states that Stewart, the study’s second author, “identified a mistake in the way the original data were merged” while responding to concerns raised about the paper. The problems are “coding and transcription errors,” the notice says, which “collectively exceeded what the authors believed to be acceptable for a published paper.” It also notes that Pickett “disputes that the identified discrepancies are attributable to researcher error.”
Behind that language is a tangled tale that The Chronicle of Higher Education first brought to light in September. In May 2018, someone using the pseudonym “John Smith” raised concerns about alleged statistical irregularities in the five papers. The concerns prompted Pickett, who worked on one of the papers as a doctorate student under the supervision of Stewart, to take a second look at the 2011 paper and eventually publish a 27-page critique of it. Pickett’s main concern is that the study mentions a survey of 1184 people about whether judges should take ethnicity into account when deciding sentences, but he claims to have evidence that the survey had just 500 respondents. Pickett also says he asked Stewart several times for access to the survey’s raw data, but never received them.
Pickett believes the data in the 2011 study were falsified. “That means that although there was a real survey and real data, the findings reported in the article are not based on the actual data,” he says.
Stewart did not reply to ScienceInsider’s requests for comment. (He also reportedly did not reply to the Chronicle, although the publication quoted an email that Stewart reportedly sent to FSU administrators; it stated that a co-author “essentially lynched me and my academic character”—an especially loaded phrase because Stewart is black.)
The five papers were also scrutinized by Nick Brown and James Heathers, two researchers who have gained notoriety as “data thugs” for exposing poor science and potential misconduct, after “John Smith” emailed them as well. The pair identified a number of troubling issues within the papers, including an unexpectedly high phone survey response rate and no mention of who funded the work.
Gary Ostrander, vice president of research at FSU, says the university launched a preliminary inquiry into the case. “At the conclusion of the inquiry,” he says, “the committee felt that there was no need to move to the full investigation as the professor had already been working with the journal’s editors to address any questions they had about the work.”
Pickett says he doesn’t regret being outspoken about the studies. “I am afraid that I have burnt many bridges, and it worries me a great deal,” he says. “I very much wish the world of science was more receptive and more kind to people who speak out about problems in published research, whether those problems result from honest error or misconduct.”