do the facts speak for themselves?

Much has been written lately about the Judith Miller and Matthew Cooper cases and the responsibilities of the media to the government, the law and the public trust. I've been thinking a lot lately about what constitutes the public trust, especially with regards to science. Society invests in scientists a huge responsibility to further society through their work. So it's troubling when a recent Nature article reports for US scientists that:

"Of 3,247 early- and mid-career researchers who responded, less than 1.5% admitted to falsification or plagiarism, the most serious types of misconduct listed. But 15.5% said they had changed the design, methodology or results of a study in response to pressure from a funding source; 12.5% admitted overlooking others' use of flawed data; and 7.6% said they had circumvented minor aspects of requirements regarding the use of human subjects.

Overall, about a third admitted to at least one of the ten most serious offences on the list — a range of misbehaviours described by the authors as "striking in its breadth and prevalence"."

An AP article cites this article and notes that the Dept. of Health and Human Services received 50% more complaints of scientific misconduct than two years ago. That's 274 complaints.

What is perhaps most troublesome are the reasons that people cheat. Usually it's not due to lack of adequate training. It's the high profile environment, and the cutthroat race to gain tenure or to maintain one's reputation. The case of Andrew Friedman suggests that high-profile research is a pressure cooker. According to that rationale, cheating helps one stay ahead or at least on-course. Why must this be true? Is it really fair to say that the environment produces cheaters? Somehow that's like a nature vs. nurture argument--with the truth lying on both sides. Yes, a glass ceiling exists in first tier research institutions for those who do not produce as much as their neighbor. However, it's rather unfair to blame the environment rather than one's own moral code.

So what needs to be done? Classes on scientific ethics are a start. Perhaps not just at the graduate level (as is already done), but at the undergraduate level, when many students begin their training. Also, I'm trying to learn more about the process of reviewing an article for publication. What I know so far: it's very, very tedious and actually quite boring. Checking 43 sources is nothing compared to figuring out whether the topic is "novel" and worthy of publication. It would be easy to see why people would not be willing to dig into every article with gusto and severely question data or conclusions. However, a strong sense of responsibility of reviewers to ascertain the validity of data is crucial.

Sometimes it seems that scientific research comes up in the news for two reasons: sexy discoveries or scientific misconduct. The first is natural and likely well-deserved. The second scenario gets the largest outcry in the case of Vioxx/Celebrex and the threat of "Big Pharma." That this latter situation occurs at all is a disservice to the 2/3 of scientists who do not violate the public trust.


Full text of Nature study