Some scientists bend rules, survey finds

NEW YORK – While rare cases of scientific fraud grab headlines, more mundane misbehaviors are so common among researchers that they pose a threat to the integrity of the scientific enterprise, a new report asserts.

One-third of scientists surveyed said within the previous three years, they engaged in at least one practice that probably would get them into trouble, the report stated. Examples include circumventing minor rules for doing research on people and overlooking a colleague’s use of flawed data or questionable interpretation of data.

Such behaviors are “primarily flying below the radar screen right now,” said Brian C. Martinson, of HealthPartners Research Foundation, who will present the survey with colleagues in today’s issue of the journal Nature.

Scientists “can no longer remain complacent about such misbehavior,” the commentary states.

“I don’t think we’ve been complacent,” said Mark Frankel of the American Association for the Advancement of Science.

Frankel said the survey sampled only a slice of the scientific community.

The survey included 3,247 scientists. They were researchers based in the United States who’d received funding from the National Institutes of Health.

Most were studying biology, medicine or the social sciences, with others in chemistry and a smaller group in math, physics or engineering.

Fewer than 2 percent of the study respondents admitted to falsifying data, plagiarism or ignoring major rules for conducting studies with human subjects. But nearly 8 percent said they’d circumvented minor aspects of such requirements.