Liz Wager, the chair of the Committee on Publication Ethics, knows something about retractions. In April, she and University College London’s Peter Williams published a paper in the Journal of Medical Ethics showing that journal editors’ approaches to retractions aren’t uniform.
The pair is back with another paper, using the same dataset of retractions and published in Science and Engineering Ethics, in which they ask journal editors why they retract — or don’t. The findings — more on them below — informed COPE’s 2009 guidelines on retractions, as did those in the April paper.
From the introduction to the new paper (link added):
Retraction is one of the most serious sanctions journals can take against authors in cases of misconduct, and can cause permanent damage to an academic career. However, journal policies and practices are not consistent. For example it appears that some journals will always retract in cases of redundant publication, while others will issue a notice of duplicate publication or do nothing. Even in well-publicised fraud cases, some journals have not retracted the affected articles (Sox and Rennie 2006).
Discussion of cases at COPE indicates that editors or publishers are sometimes reluctant to retract articles. Reasons for such reluctance may include beliefs that retractions may be made only by the authors; author disputes in which some authors request retraction while others oppose it; and concerns about, or actual threats of, litigation from authors.
Wager and Williams wanted to see which of these reasons were most salient, so they approached 11 editors, of whom 5 “agreed to be interviewed during the study timeframe.” Here are some of the highlights of those interviews.
First, an editor after our own hearts, who seems very interested in transparency:
It is worth noting here that one of the respondents felt that, in a retraction, where there is a dispute between authors, it is important in every case to ‘spell out’ who doesn’t agree and about what. Similarly, credit should be given to the person who noticed any error and brought the case to the attention of the editors.
…one editor felt so strongly about a particular case of plagiarism that he wanted to ban the infringing author from publishing in the journal for 3 years. He took the case to COPE, who said that further punishment (i.e. beyond the retraction) was unnecessary, and reluctantly abided by this recommendation.
And there was a reminder that just because science moves beyond a particular finding, it doesn’t need to be retracted — although error-laded papers should still be withdrawn:
…one editor said that her journal often receives complaints from readers about articles that are years old, to the effect that the results or methods described are no longer valid. However, the editors consider that there is a distinction between a paper containing errors and one that is superseded in later years by further scientiﬁc or technological advances. As long as the paper took into account everything that was known at the time, the paper is considered legitimate.
A related response:
Another problem raised is that the act of ‘retraction’ has negative connotations, and so it was very important to be clear about the reason for taking this step. In some cases, it was pointed out, ‘correction’ notices could be used instead, which may have less negative effects on authors. Another interviewee made the point that although papers may be retracted, many of them still contained useful research and, of course, the paper still exists and therefore may be read and cited. For these reasons it is very important to choose one’s words very carefully in a retraction, and to emphasise what is correct in the paper.
Retraction Watch is looking at you, Journal of Biological Chemistry.