Plagio en artículos de investigación en revistas biomédicas cubanas. 2016

Introduction: Plagiarism and the fabrication and falsification of results constitute the most serious forms of scientific misconduct. However, there is a lack of solid studies on this malpractice which motivated us to carry out this research work. Objective: To characterize the articles engaged in plagiarism published in Cuban biomedical journals indexed in SciELO in 2016.Material and Methods: A descriptive cross-sectional study was conducted using a probabilistic sample of 50 original articles obtained by simple random sampling of 786 research studies published in 32 Cuban biomedical journals indexed in SciELO. A textual analysis was performed using Google Scholar to identify identical texts. After that, similar articles were compared and the oldest data of publication was taken into account to determine the original article. R Studio Software 1.0.136 was used for the analysis. The titles of the articles, authors, institutions and journals were not shown.Results: More than half of the articles (52 %) met the criteria of plagiarism. Similarities were found in all sections of the article, mainly in the introduction (52 %), discussion (18 %) and material and methods (10 %). The most frequent pattern of plagiarism was in “mosaic” (substitution, addiction, subtraction), followed by “direct plagiarism” (exact copy of sections of text).Conclusions: The frequency of plagiarism in the articles reviewed is high, predominating in the theoretical structures of the article (introduction and discussion), where synonyms, changes in verbs, addition or omission of parts of the text were identified, corresponding to plagiarism in “mosaic”. Introducción: El plagio junto a la fabricación y falsificación de resultados constituyen las formas más graves de mala conducta científica. Sin embargo, la escasez de estudios concretos sobre esta mala práctica, motivó realizar la presente investigación.Objetivo: Caracterizar los artículos que incurren en plagio publicados en revistas biomédicas cubanas indizadas en SciELO, durante 2016.Material y métodos: Estudio descriptivo, transversal, con una muestra probabilística de 50 artículos originales, obtenida por muestreo aleatorio simple de 786 investigaciones publicadas en 32 revistas biomédicas cubanas indizadas en SciELO. Se realizó un análisis textual a través de Google Scholar, para identificar textos idénticos. Luego se cotejaron los artículos con similitudes, y se seleccionó como original el de fecha de publicación más antigua