Earlier this year, a whistle-blower email pointed out data discrepancies in five articles co-authored by Dr. Eric A. Stewart, a Florida State University criminologist. The email triggered suspicions of data fabrication by Stewart, the only author included on all five papers. FSU began an investigation, but an inquiry committee containing two of Stewart’s past co-authors decided not to move forward with it. So far, four of the five articles have been retracted.
Stewart and his co-authors received the email from “John Smith” in May 2019. It outlined multiple data irregularities and implied that the extreme errors may have been more than oversights or mistakes.
Soon after, Dr. Justin T. Pickett, Stewart’s co-author and former mentee, published a paper encouraging Criminology to retract the article that he co-authored with Stewart and others, “Ethnic threat and social control: Examining public support for judicial use of ethnicity in punishment.”
In his paper, “Why I Asked the Editors of Criminology to Retract Johnson, Stewart, Pickett, and Gertz (2011),” Pickett listed eight glaring problems with the data of the article he co-authored. These included dramatically incorrect figures; for example, Pickett wrote, the article listed 1,184 respondents versus the actual 500 respondents. Based on his analysis, Pickett argued that the data were altered in some way.
The situation deepened in September when The Chronicle of Higher Education published “The Criminologist Accused of Cooking the Books,” a longform article by journalist Tom Bartlett which amplified Pickett’s argument.
According to the article, before the co-authors had received the whistleblower email, “John Smith” had sent data to James Heathers and Nick Brown. The two have become known for exposing faulty data, and they have prompted retractions of prominent papers.
They reviewed the data that “John Smith” sent them, focusing their analysis on “A Legacy of Lynchings: Perceived Black Criminal Threat Among Whites,” written by Stewart, another FSU professor Daniel Mears, and other co-authors. Heathers and Brown concluded that the data contained multiple irregularities.
“Among the most unusual features was the sample size: 2,736 American adults,” Bartlett wrote. “That’s a sizable survey, the cost of which would almost certainly run into six figures, and no funding source was listed. Who footed the bill? Maybe even more remarkable was the response rate—61 percent—which is extraordinarily high for a phone survey.”
This information further implied that the numbers were either fabricated or unintentionally altered.
Stewart has remained publicly quiet regarding the situation, but he has defended himself privately. In a text to Pickett, Stewart said he had been targeted by “data thugs,” and, to FSU administrators, Stewart expressed that his character had been “lynched,” Bartlett wrote. In the frame of the Chronicle article, Stewart seemed to boil the situation down to a personal attack.
In another text to Pickett, Stewart said he expected FSU to launch an investigation. At the time that the Chronicle article was published, the investigation was underway. However, since then, FSU has abandoned the investigation, according to a statement released by an FSU representative.
“Dr. Stewart had been working with his colleagues and the journal long before the FSU administration was alerted to the issue,” the statement reads. “However, when informed, we embarked on the first of a two-part process that is divided into an initial inquiry and then an investigation. At the conclusion of the inquiry, 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.”
The inquiry committee included two of Stewart’s past co-authors: William D. Bales and Sonja E. Siennick. However, FSU’s conflict of interest policy lists collaborations and co-authorships as potential conflicts, bringing into question FSU’s compliance with its investigative policies. It is unclear whether or not the inquiry committee’s ties to Stewart influenced the decision to not investigate.
As of now, Stewart is working with Criminology to resolve the data problems without the involvement of FSU.
“The journal’s process is completely independent from the university,” the statement concludes, “and we respect Professor Stewart’s decision to ultimately request a retraction.”