A federal agency that funds scientific research nixed punishments recommended by its own ethics watchdog for some academics who plagiarized and manipulated data in grant proposals and taxpayer-funded research, public records show.
The inspector general for the National Science Foundation identified at least 23 instances of plagiarism in proposals, NSF-funded research, and agency publications in 2015 and 2016. It found at least eight instances of data manipulation and fabrication in those years. NSF officials disregarded recommended sanctions against some of the scientists and academics implicated in those findings. Though many were temporarily barred from receiving additional federal funding, nearly all will be eligible for taxpayer support and official roles in NSF-funded research in the future.
In one investigation that concluded in Nov. 2015, the IG found that an NSF-supported researcher had "knowingly plagiarized text into five NSF proposals."
"These actions were a significant departure from the standards of the research community, and therefore constituted research misconduct," according to a report on the investigation's findings.
The IG recommended to NSF that the agency officially classify the plagiarism as research misconduct, require the researcher to undergo "a course in proper research methods," certify that all research over the subsequent three years was not plagiarized, and bar the researcher from serving as an NSF consultant, advisor, or peer-reviewer.
The NSF accepted most of the recommendations, but it chose not to bar the researcher from working for NSF in an official capacity, as the IG had proposed. The researcher would be free to continue advising, consulting, and peer-reviewing taxpayer-funded research.
In another investigation, which concluded in Aug. 2016, the IG found that a university professor supported by an NSF grant "falsified the status of a total of seven manuscripts in four NSF annual grant reports and four NSF proposals" and "engaged in a total of twelve acts of research misconduct in a continuous pattern spanning several years."
"The professor's fabrication of data and falsification of manuscripts' status were intentional acts, fit a pattern of research misconduct, and were a significant departure from accepted practices," the IG concluded.
The NSF agreed, pursuant to the IG's recommendations, to debar the professor for one year and require a course on proper research methods.
However, "contrary to our recommendations," the IG wrote, the agency did not require the professor to submit certifications of data integrity after that period of debarment and did not ban him from serving as an NSF advisor, peer-reviewer, or consultant going forward.
While it was more common for NSF officials to accept and implement IG recommendations, those examples were two of a handful of investigations in which the agency opted to impose more lenient punishments for serious academic misconduct than its watchdog recommended.
In nearly every case, the punishments imposed on scientists and academics found to have manipulated data or plagiarized agency-funded research were temporary and allowed the culprits to continue conducting taxpayer-funded work after a probationary period.
One researcher working on projects funded by the NSF and the National Institutes of Health admitted "that he falsified data in three publications," according to the IG's findings. The IG recommended that he be debarred for five years, but the NSF opted to reduce that period to three years.
The NSF did not respond to questions about the IG's findings and concerns over academic misconduct in federally funded scientific research.
Publicly available versions of the IG reports are heavily redacted to remove information that might identify the researchers and projects at issue, preventing efforts to request comment from the subjects of the IG's investigations.
While many of the researchers in question admitted misconduct when confronted, some offered dubious explanations for violations of basic academic standards.
One researcher found to have plagiarized from five sources in an NSF grant proposal told investigators that she "was never instructed regarding use of quotation marks while a graduate student in the U.S."