.Back to listing

Let’s redefine “research misconduct” more realistically


Guest post by Professor Ross McKinney.

The saga of Magdalena Koziol at Yale, reported in Science (“Sabotaged scientist sues Yale and her lab chief”, 343 (07 March 2014) p.1065-66) is certainly an unhappy one. One of Koziol’s coworkers, Polloneal Ocbina, repetitively poisoned her zebrafish for reasons that remain unknown outside the halls of Yale, although there seems to be general agreement on the basic facts of the poisoning.   The story, as it has played out, looks like one of those situations where it will take some time before both sides are allowed to array all the facts as they know them for public review, if ever.  Cases like this are often sealed with contractual agreements to maintain confidentiality.  In the meantime, the case raises a worthwhile question: is our current definition of “research misconduct” fundamentally flawed?  The Office Research Integrity of the U.S. Department of Health and Human Services evaluates cases of misconduct, and its formal definition is constrained to fabrication, falsification, and plagiarism.  In lay terms, this translates into making up data, manipulating data improperly, or stealing other people’s ideas without adequate credit.  Other bad behaviors in science may be truly egregious (e.g. intimidation by a senior investigator, harassment, systematic sloppiness, devious selection of statistical tests) but they won’t be considered research misconduct because “research misconduct”  is a term of art, a piece of jargon, and constrained in ways that common sense doesn’t reconcile.  This artificial definition seems particularly limited when considering the question of sabotage, as illustrated by the Koziol case. Was this “research misconduct” on the part of Ocbina? In one prior case, the Office of Research Integrity managed to gerrymander the definition of misconduct by declaring that research record was falsified as a result of sabotage.  This would be quite a stretch in this case: research misconduct is usually done by the researcher, not to the researcher.  In fact, the regulation says fabrication occurs when “the research is not accurately represented in the research record.”  I suspect Dr. Koziol’s lab book very accurately described a tank of fish swimming belly up.  Some institutions include vague language that considers behaviors outside the norms of science to be misconduct, but clearly those behaviors are being judged in a subjective way that will likely be inconsistent in its application.  Perhaps it’s time to create a longer list of bad scientific behaviors that cross the threshold and should be considered under a broader and improved definition of “research misconduct”.