Survey: Scientific Misconduct Underreported
ROBERT SIEGEL, Host:
Science is supposed to be about finding facts, not making them up. But occasionally researchers do make things up. It's hard to know how often. A new survey released today suggests that each year in the U.S. a thousand of potential instances don't get reported.
NPR's David Kestenbaum reports.
DAVID KESTENBAUM: Misconduct here means more than just taking home a box of pens from work. It means faking data, altering results, or plagiarism - the very opposite of what science is supposed to be about.
SANDRA TITUS: This serious - the most serious thing probably sort of harming someone.
KESTENBAUM: This is Sandra Titus, who works at the government's Office of Research Integrity.
TITUS: Scientist's motivations are really wonderful. If something goes wrong along the way sometimes that they want to speed the process up, or they want to become famous overnight, or they think they have the right results and they indeed don't have the right result.
KESTENBAUM: Her office deals with allegations of misconduct that involve researchers funded by the National Institutes of Health. And the NIH is huge - annual budget about $28 billion. The universities and other places doing the research are supposed to investigate alleged misconduct and pass their findings on. But in a typical year the Office of Research Integrity only receives 24 such reports. Titus wondered if there were problems they were not hearing about.
TITUS: Nobody really had much of a handle of, is it a really big problem? Is it a small problem? Is this a problem at all?
KESTENBAUM: Titus and two other researchers surveyed scientists who got money from the NIH, over 2,000 respondents. Of them about 7 percent thought they had observed scientific misconduct over the past three years. But what worried her was that when these scientists saw something, they did not always report it. The survey indicated there could be 1,000 unreported incidents a year - potential incidents.
TITUS: So people don't want to go forward, even if they observe something.
KESTENBAUM: And what about the hundreds of times when scientists did report things? Why was Titus's office only hearing about a couple dozen a year? She says one possibility is that institutions and universities are trying to keep things quiet. Everyone seemed to be dropping the ball.
TITUS: The evidence seems to me to be pretty compelling that something is happening and it's happening in everybody's world. It's happening in the universities. It's happening, you know, in the departments. and if you don't look at it and you don't addressed it, you're going to just have something grow bigger and bigger.
KESTENBAUM: Titus and her colleague described their findings in this week's issue of the journal Nature. And some readers will argue it paints an overly bleak picture. Howard Garrison is a spokesperson for the Federation of American Societies for Experimental Biology, which he says represents 85,000 people.
HOWARD GARRISON: We have to understand that the survey is collecting data on suspected misconduct, possible misconduct, and at various stages in the process individuals and organizations look closely at that data and may make determinations that what was initially thought to be misconduct turns out not to be misconduct at all.
KESTENBAUM: Garrison says yes, misconduct happens, more should be done to stop it, but he thinks it's actually pretty rare.
GARRISON: I cannot believe any scientist who's practicing in the United States today and understands their craft would really do this.
KESTENBAUM: Cheaters get found out, he says, because someone will try to replicate the results and find they can't. The authors of the article in Nature proposed a zero tolerance policy, where all suspected misconduct must be reported. Garrison worries that could create a climate of fear and suspicion that in his words would not be helpful.
David Kestenbaum, NPR News.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.