Resumo

GUTIERREZ, Sneider Alexander et al. Retraction and correction of scientific literature for conserving the integrity of and confidence in science: An analysis of retractions in open access biomedical publications in PubMed, 1959-2015. Rev. acad. colomb. cienc. exact. fis. nat. [online]. 2016, vol.40, n.157, pp.568-579. ISSN 0370-3908.  https://doi.org/10.18257/raccefyn.399.

Informatics tools have enabled the detection of various types of misconduct regarding research studies in scientific literature. The Office of Research Integrity (ORI) (ORI), the Committee on Publication Ethics (COPE), Retraction Watch and countries' academies of science sponsor courses, seminars and research to promote values such as honesty, impartiality, objectivity, reliability, responsibility and skepticism in scientific communication. We analyzed here 1,373 articles recorded in PubMed from 1959 to 2015 with open access to their text and retraction notes. We observed that articles retracted from 2010 to 2015 almost duplicated those accumulated during the previous 44 years; in 32.8% of them retraction was due to admitted error; 23.7% to plagiarism or self-plagiarism, and 19.7% to data falsification or fabrication. Thirty-seven articles were retracted during the first four months of 2015 due to false review or author influence on reviewers, which represents a research misconduct not detected in previous studies. The percentages of open access retracted articles published per year varied from 0.0072% (1/13,861) in 1966 to 0.0472% (213/451,021) in 2013. The percentage of articles retracted compared to articles published from 54 countries throughout the world during the same period (1959-2015) varied from 0.0042% (1/23,761) to 0.2732% (1/366). The amount of articles retracted signed by more than 10 authors was lower than that for 6 to10 or 1 to 5 authors. We found that 794 (57.8%) articles were retracted before the first two years and 579 (42.2%) more than two years after their publication. The retraction of 714 (52%) of the articles was requested by the authors, of 485 (35.3%), by the editors, and of 70 (5.1%) by mutual agreement; 80.8% (1,110/1,373) of the retracted articles had been cited. We discuss here the importance of promotion, education, retraction and correction of scientific literature as a contribution to scientific integrity and society's confidence in the scientific community.

Palavras-chave : Research misconduct; Ethics of scientific communication; Bibliometrics; Retracted article.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )