Survey: Misconduct in research likely underreported

Instances of falsified results, fabricated data and plagiarism in scientific research may be vastly underreported, according to findings published in a commentary in the current (June 18) issue of Nature.

Seeking to know if the low number of research misconduct cases it sees each year — on average just 24 — were the tip of a much larger iceberg, the U.S. Department of Health and Human Services‘ Office of Research Integrity surveyed researchers with National Institutes of Health (NIH) funding nationwide about potential misconduct they’d witnessed.

Of the 2,212 scientists at 605 institutions who responded, nearly 9 percent said they had observed a possible occurrence of research misconduct during the past three years. Extrapolating those results to the larger universe of NIH-supported investigators suggests that up to 2,300 observations of potential misconduct take place each year, with about 1,000 going unreported.

Considering the massive scientific output of NIH-funded scientists, including grant proposals, dissertations and journal articles, 2,300 isn’t a terribly large number, says co-author Jim Wells, now director of the Office of Research Policy at the University of Wisconsin–Madison.

"On the other hand, it’s a lot higher than anyone expected," he says. "And it’s discomforting to think that so much potential misconduct never gets reported and thus we can’t bring to bear our policies for investigating these cases, protecting whistle blowers and so on, to make sure things are handled properly."

He hopes to see more attention paid to research integrity in the future, including better education for graduate students and other young scientists on the issue, and increased protections for those who blow the whistle on possible data falsification and fabrication.