Fulltext

Do editors and referees look for signs of scientific misconduct when reviewing manuscripts? A quantitative content analysis of studies that examined review criteria and reasons for accepting and rejecting manuscripts for publication

Bornmann, Lutz ; Nast, Irina ; Daniel, Hans-Dieter

In: Scientometrics, 2008, vol. 77, no. 3, p. 415-432

Ajouter à la liste personnelle
    Titre
    • Do editors and referees look for signs of scientific misconduct when reviewing manuscripts? A quantitative content analysis of studies that examined review criteria and reasons for accepting and rejecting manuscripts for publication
    Auteur
    • Bornmann, Lutz. ETH Zurich, Zurich, Switzerland
    • Nast, Irina. ETH Zurich, Zurich, Switzerland
    • Daniel, Hans-Dieter. ETH Zurich, Zurich, Switzerland
    Type de document
    • Postprint
    Classification
    • Informatique
    Identifiant OAI-PMH
    • oai:doc.rero.ch:314125
    Summary
    The case of Dr. Hwang Woo Suk, the South Korean stem-cell researcher, is arguably the highest profile case in the history of research misconduct. The discovery of Dr. Hwang's fraud led to fierce criticism of the peer review process (at Science). To find answers to the question of why the journal peer review system did not detect scientific misconduct (falsification or fabrication of data) not only in the Hwang case but also in many other cases, an overview is needed of the criteria that editors and referees normally consider when reviewing a manuscript. Do they at all look for signs of scientific misconduct when reviewing a manuscript? We conducted a quantitative content analysis of 46 research studies that examined editors' and referees' criteria for the assessment of manuscripts and their grounds for accepting or rejecting manuscripts. The total of 572 criteria and reasons from the 46 studies could be assigned to nine main areas: (1) ‘relevance of contribution,' (2) ‘writing / presentation,' (3) ‘design / conception,' (4) ‘method / statistics,' (5) ‘discussion of results,' (6) ‘reference to the literature and documentation,' (7) ‘theory,' (8) ‘author's reputation / institutional affiliation,' and (9) ‘ethics.' None of the criteria or reasons that were assigned to the nine main areas refers to or is related to possible falsification or fabrication of data. In a second step, the study examined what main areas take on high and low significance for editors and referees in manuscript assessment. The main areas that are clearly related to the quality of the research underlying a manuscript emerged in the analysis frequently as important: ‘theory,' ‘design / conception' and ‘discussion of results.'