Peer-reviewed publications are the standard of excellence in the scientific world. They are the reliable source of progress for each of the disciplines they represent. When peer-reviewed publications publish and then retract a report, it is because something in that report was inaccurate. In a recent issue of the prestigious journal Proceedings of the National Academy of Sciences, the authors reported that there has been a 10-fold increase in retracted scientific papers since 1975.
The authors, who reviewed more than 2,000 reports, concluded that 67.4 percent of the retractions were due to misconduct, including fraud, suspected fraud, duplicate publications and plagiarism. Only 21.3 percent were attributable to error.
Some people believe that this precipitous rise in retractions is caused by sloppy work in a competitive publish-or-perish environment, or that the rate of errors is the same as it always was, but that detection is more frequent because of increased sophistication of computer systems that detect errors. But computer searches that can detect errors after publication should also be able to detect errors before publication.
Not long ago, you could trust that information in a scientific journal or presentation was truthful. Scientists may have made mistakes in procedure or evaluation, but you could believe they were telling the truth as they understood it. Increasingly, this is not so today. The Department of Health and Human Services' Office of Research Integrity Division of Investigative Oversight reports that it handles approximately 200 allegations of research misconduct each year.
Not that long ago, you could walk into a drug store and be pretty certain that the non-prescription drug you took from the shelf or the prescription drug you received from the pharmacist was pure. Increasingly, this is not so today.
We are now facing questions of integrity and failure of self-regulation, which are the underpinnings of scientific credibility. In previous eras, the scientific community was relatively small and concentrated in Western Europe and the United States. Scientists in a particular field knew one another and read each other's reports. Peer pressure kept most honest and accurate. Data was shared among researchers. Scientists would repeat experiments to verify the accuracy of each other's results. Peer-reviewed publications would make information available to all in a field and government regulation could catch mistakes before they affected the general public.
Science has always been competitive and it is becoming more so. Today, academic researchers are increasingly competing for scarcer government grants. Industrial scientists and corporate laboratories are being put under tremendous pressure because of continuing cost-cutting initiatives and increasing demands for productivity. The scientific community has become huge, spanning countries that were never before centers of scientific advancement, and the pace of scientific progress has made it impossible to track all advances in a discipline, even in some that are highly specialized.
In a competitive world, where pursuit of patent protection is king, scientists are loath to share data. Peer-reviewed publications have failed to detect fraud and mistakes, while rogue companies and sometimes established name-brand enterprises escape government oversight through loopholes in regulation or the inability of the government regulators to be everywhere.