What analyzing 30 years of US federal research misconduct sanctions revealed

A U.S. federal agency that oversees research misconduct investigations and issues sanctions appears to be doling out punishments fairly, according to researchers who analyzed summaries of the agency’s cases from the last three decades. 

But the authors of the study also found more than 30 papers the ORI said should be retracted have yet to be.

The researchers looked for associations between the severity of penalties the Office of Research Integrity (ORI) imposed on scientists it found responsible for research misconduct and their race and ethnicity, gender, academic rank, and other qualities. The researchers published their findings in late November in Accountability in Research, as the agency is in the process of revising its key regulations

According to the new analysis, ORI’s sanctions correlated with factors indicating the seriousness of the misconduct, such as being required to retract or correct publications, but not with demographics. 

“We did not find evidence of bias,” Ferric Fang, a professor at the University of Washington School of Medicine and one of the study’s authors, said. 

Fang, also member of the board of directors of The Center For Scientific Integrity, Retraction Watch’s parent nonprofit organization, told us: 

The ORI states that the severity of administrative actions should be based on the circumstances and severity of misconduct as well as the presence of aggravating or mitigating factors, such as repeated behavior, the impact of the misconduct on the research record and public health, retaliatory behavior, and whether the individual committing misconduct was directly responsible and accepted responsibility for their actions. Our findings suggest that the ORI has been consistent in applying these criteria.  

“In light of ongoing concerns about disparities in research funding and the STEM workforce, our findings should provide some reassurance that the ORI is applying its administrative actions in an even-handed manner,” Fang said. 

Specifically, the researchers found “factors related to the severity of the misconduct or aggravating factors, such as whether the person interfered with the investigation, had violated a prior agreement with ORI, or was required to retract papers, were positively associated with the severity of the administration action,” said David Resnik, the study’s corresponding author and a bioethicist with the National Institutes of Environmental Health Sciences. He further summarized the group’s findings: 

Whether the person had committed plagiarism only (viewed by many as less serious than data fabrication/falsification) and whether the person admitted wrongdoing (a mitigating factor under most systems of punishment) were negatively associated with the severity of the administrative action.  Factors unrelated to the severity of the misconduct or aggravating or mitigating factors, such as the person’s race/ethnicity, gender, academic position, education, or institutional affiliation were not associated with the severity of administrative actions.   

The researchers found a three year period of research supervision or funding ban was the most common sanction ORI levied, accounting for 65% of cases. ORI has occasionally banned researchers from federal funding for life, and more recently issued a 10-year funding ban and a pair of seven-year penalties.

“I suppose that this is the default length of time,” Fang said of the three-year sanction, “and may represent a penalty that imposes a true hardship on a researcher without necessarily being career-ending.” 

We asked how the findings might be relevant to ORI’s current proposals for revising its regulations. Resnik, a Federal employee, declined to speculate. Fang said: 

I am aware that some of the recently proposed updates to ORI policies have been controversial, but I don’t believe that the administrative actions examined in our study will be substantially changed.

The researchers also found that 32 papers ORI said should be retracted as part of its sanctions have not been pulled from the literature. 

“We should be concerned” about this finding, Resnik said. “It means that the literature has not been properly corrected and that scientists may be unwittingly relying on fraudulent research.  Moreover, these are people who were caught committing misconduct, were sanctioned by their institutions and ORI, and were required to correct or retract papers.” 

Fang said: 

The fact that some papers that the ORI asks to be retracted are not necessarily retracted reflects that fact that only journals can retract publications, and unfortunately, not all appear to take this responsibility seriously. 

Like Retraction Watch? You can make a tax-deductible contribution to support our work, subscribe to our free daily digest or paid weekly update, follow us on Twitter, like us on Facebook, or add us to your RSS reader. If you find a retraction that’s not in The Retraction Watch Database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.

3 thoughts on “What analyzing 30 years of US federal research misconduct sanctions revealed”

    1. Great point, the NCBI should flag these. They should also flag papers with clear evidence of manipulation, whether or not the authors replied or the respective EIC took any action.
      Journals that exhibit a pattern of ignoring fraud should be given a fair warning followed by delisting.
      NCBI is supposed to represent a reliable library / repository of knowledge and the ultimate source of information, as originally intended. It cannot allow itself to be polluted by pseudoscience and noise.

  1. $58 to read one paper, no thanks!

    It’s a problem when folks who are thought to be leaders in the scientific integrity field put their papers behind a paywall, even though open publishing (including all original data) is acknowledged to be one of the key tools to improve scientific integrity.

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.