Imprimir Republish

Good Practices

Localized effects of retraction

BoasPraticas_242aaDaniel BuenoA study by six researchers from Thomason Reuters, a communications company that produces data on science, examined the impact that cases of scientific misconduct have on their surroundings and concluded that the damage is concentrated on the retracted articles and their principal authors, and does not affect the institution where the researcher accused of fraud or plagiarism worked.  Those who inadvertently cited the canceled article in their papers or published other papers in co-authorship with the accused researcher also retain their academic reputation, seemingly exempt from the deleterious effects of the retraction.  These findings were published in February 2015 on the electronic repository arXiv (pronounced “archive” in English).

Researchers analyzed 2,659 articles in various fields of knowledge that had been retracted between 1980 and 2014.  All had been indexed in the Thomson Reuters database known as the Web of Science.  The study identified the reason for the request for retraction for 1,666 of the papers.  More than 25% of the cases involved plagiarism.  Approximately 24% of retractions resulted from unintentional errors committed by authors and about 23% of the articles were canceled because of data falsification or fabrication.  One theory espoused by the authors of the research is that plagiarism has become more frequent owing to the increasing availability of scientific literature on the Internet.

The study also found that authors of retracted papers are cited less frequently after they were discovered.  Authors who falsified or fabricated data suffer greater damage to their reputations than those who committed unintentional errors, and press coverage of the scandal enhances that effect.  The institutions or fields of knowledge with which the retracted article was associated suffer practically no impact.  The study cited as example Woo-Suk Hwang, author of two fraudulent articles about stem cells published in the journal Science in 2004 and 2005.  He was severely punished by a drop in citations of his other works but the effect did not extend to the Seoul National University College of Veterinary Medicine, which maintained an ascending curve of citations.  The study also shows that retractions occur more often in the medical and biological sciences.

“The most original aspect of this study is that it assesses the effects of a retraction on institutions and specific fields of knowledge,” states Ferric Fang, an author of research studies about retraction and professor at the University of Washington.  “The evidence that retractions result in a decline in the rate of citations, particularly when there is misconduct, are a sign that the system is working the way it is supposed to,” he told Retraction Watch blog.