Daniel BuenoWhen a scientific article is retracted – i.e., disqualified by the journal that published it – due to the discovery of errors or fraud, the impact can spread to the entire community of researchers that study topics related to the subject of the excluded paper. This conclusion was based on an analysis of 1,104 papers that were published then retracted from the PubMed database. According to the news blog in the journal Nature, a study published by the National Bureau of Economic Research examined a selection of articles about similar topics to the retracted papers, but written by different authors. It was observed that after a retraction, citations of the correlated papers dropped 5.7% compared to a selection of non-related papers used as a control group. Even more impressive was the effect on funding for research related to the suspended articles. Given that PubMed monitors investments made by the National Institutes of Health (NIH) in the research that gave rise to the papers, it was possible to identify that funding for correlated studies dropped between 50% and 73% after the retractions. “The impact on funding was quite significant,” says the economist Pierre Azoulay, from the Sloan School of Management at the Massachusetts Institute of Technology (MIT), the study’s main author. Azoulay and his colleagues believe that there are two explanations for the ripple effect. Either the scientific community realized that the field of knowledge in question is more limited than it looked, or it decided to keep its distance out of fear of being associated with the controversy. The second hypothesis seems more likely, as the effect is more visible when the paper is retracted due to misconduct than when it is retracted in cases of unintentional error. One example mentioned in the study is the South Korean researcher Woo Suk-Hwang, who announced in a paper published in the journal Science in 2005 that he had successfully cloned human embryos for stem cell harvesting (see Pesquisa FAPESP No. 120). Naturally, many researchers attempted to reproduce the experiment, which turned out to be fraudulent. But after the paper was retracted, the South Korean government cut its investments in embryonic stem cell research in half, and many researchers abandoned the field. One solution for preventing losses, says Azoulay, is to develop a coding system for retractions, to make it clear why a paper has been disqualified. He acknowledges that editors will probably dislike the suggestion, as they would be forced to clearly state what was wrong with the eliminated article.
Print issue
Current | Past
