A new study finds that fraud in scientific research is growing at a troubling rate, even though it remains rare overall.
A review of retractions in medical and biological peer-reviewed journals finds the percentage of studies that had to be withdrawn because of scientific misconduct has jumped several-fold since the mid-1970s.
The study says fraud or suspected fraud is by far the biggest reason for retractions, outweighing errors and plagiarism.
Fraud is detected only a handful of times for every 100,000 studies published. Study author Arturo Casadevall says a few scientific scofflaws cause big problems that can hurt people. He says one reason may be pressure to hit it big in science.
The study is published Monday in the Proceedings of the National Academy of Sciences.