SCIENCE : Fraud and Errors in Research Papers Prompt Calls for Federal Scrutiny : Government probes of reporting on experiments draw fire from academics, who say real dishonesty is rare.
Is fraud becoming more common in American science?
That question has surfaced anew since a spate of incidents involving allegations of misconduct in the reporting of scientific research.
In one of the best known of these incidents, an assistant to David Baltimore, president of Rockefeller University in New York, charged that Baltimore used falsified and manufactured data supplied by a colleague in an influential paper, then refused to investigate warnings that the material was flawed.
THE CONTROVERSY: Baltimore eventually apologized after a panel convened by the National Institutes of Health questioned the authenticity of the information. Meanwhile, the case has sparked major controversy in the scientific community as to how scientific reporting should be policed.
“There is no way of knowing how much science fraud there is out there,” said Rep. John D. Dingell (D-Mich.), who headed a controversial inquiry into the Baltimore affair, “but there appears to be enough to be concerned about it.”
The rash of similar incidents has aroused the interest of Congress, which has begun wondering--to the horror of academics--whether the scientific community needs federal policing.
Dingell, who is chairman of a House Energy subcommittee on oversight and investigations, brought in two scientists--Walter Stewart and Ned Feder--and the U.S. Secret Service to probe the Baltimore case.
In poring over Baltimore’s laboratory notebooks, the team found that dates had been altered on some notes and that some portions were written in different ink and on different papers. Their conclusion: that the books had been put together after Baltimore’s studies were questioned.
THE CRITICS: The investigation drew sharp criticism from scientists and academics, many of whom roundly condemned it as a new form of McCarthyism. Some referred to Stewart and Feder as “the science FBI.”
Stewart, for one, is unrepentant about his role in the investigation. “Until very recently,” he said, “there was an unwillingness to call scientists to account for their false statements.”
Phillip A. Sharp, professor of biology at the Massachusetts Institute of Technology, said that despite the rash of bad publicity, fraud in science is still “the exception” and that most such incidents arise from sloppiness and little more.
“There’s a very fine line between actual fraud and genuine error,” Sharp said. “Sometimes, a scientist simply can’t recreate his experiment for any apparent reason, but that’s not fraud, it’s just nature.”
Sharp estimates that out of about 2,500 papers that are published in high-visibility journals each year, perhaps one could really be called into question.
Still, the Dingell investigation appears to have spurred the academic community to tighten the watch on itself.
“It was the worst experience of my life,” said one scientist who vigorously supported Baltimore until recently. “The ethic is that you tell the truth, and if you don’t, then the whole thing breaks down.
“Now what?”
Other Flawed Findings
The Baltimore case is not unique in the recent annals of scientific research: * Tests on children: In 1988, Stephen E. Breuning, a University of Pittsburgh researcher, was convicted of fabricating scientific data about drug trials conducted on retarded children. Breuning served 250 hours of community service and agreed not to take a job as a research scientist for 10 years.
* Heart studies: Cardiologist Robert A. Slutsky, who produced an extraordinary 137 papers at the University of San Diego between 1978 and 1985, was found by university investigators to have “falsified and misused” data. Slutsky ultimately resigned.
* ‘Glaring’ errors: In 1981, colleagues observed John Darsee, a Harvard Medical School cardiologist, forging the dates on computer tapes used to prepare abstracts. Investigators reported an average of 12 flaws in each of 18 papers they combed, some “so glaring as to offend common sense.” The university decided not to recommend Darsee for an assistant professorship, and his fellowships at Harvard and the National Institutes of Health were cut off.
* Kinsey study: A recent book raises questions about potential ethical problems involving the research behind Alfred Kinsey’s landmark reports, in the late 1940s and early 1950s, on human sexual behavior. The book alleges that what Kinsey portrayed as a “carefully planned population survey” actually was based on a small sampling skewed with prisoners, sex offenders and prostitutes. It also charges that Kinsey used children in experiments on sexual behavior.
