Ethical publishing – should plagiarized pieces be retracted ? – well, perhaps not all

What should be done about an article when it has been discovered that it contains sections which have been copied from previously published works: a) rejectit during the editing process, b) retract it even if it has already been published, or c) take some other action? This dilemma is raised in a recent issue of Nature¹, by Praveen Chaddah who has himself been the victim of two incidences of plagiarism.

Chaddah is in agreement that plagiarism constitutes a wrongdoing, but he makes a distinction between “text plagiarism”and the plagiarism of another’s ideas. When all is said and done, scientists are not writers of books and novels and the most important thing is to uphold the principle that they do not take credit for the ideas and results of other researchers. This is because plagiarism strikes right at the heart of research as a creative and innovative enterprise. An idea can be a hypothesis to explain observations, or an experiment designed to test a hypothesis, and this type of plagiarism is difficult to detect with automated plagiarism detection software which simply scans texts of documents. In addition, as Chaddah points out, the copying of research results without due reference being made constitutes a clear case of fraud more than the act of plagiarism itself.

The issue raised by Chaddah is how to deal with these three types of plagiarism.

1. The copying of text in the introduction to a paper. This type of plagiarism is easily detected by plagiarism detection software, and for some journal editors constitutes sufficient reason for retraction. But if the experiments are performed as described, the results in the errant paper could be significant and of benefit to this field of research.

2, 3. The copying of text in the “method”and / or “results”sections can constitute a more severe case of plagiarism if ideas and data are being copied. Plagiarism detection software cannot detect this type of plagiarism –this requires the eyes of independent experts in the relevant field.

While it is true that authors could have crossed a serious ethical line, the question is whether these works should be withdrawn, simply retracted, or corrected with additional information. The response is: it depends …. At any rate, the discovery of plagiarism, although it may not merit a retraction of the paper in question, requires investigation at the very least. However, the policies followed by journal editors during the revision process are not clear as far as the technology for the detection of plagiarism is concerned, nor is it clear how editors should face this situation so that cases of plagiarism of this type can be resolved.

This is not a simple issue, since for researchers the “methods”section of a paper presents a unique set of challenges when trying to be original.

The reason is that, even when doing original research, authors will inevitably find themselves repeating steps that they or others have taken before. Whether they are using an established technique to study something new or simply replicating an earlier study, it can be frustrating to try and find new ways to accurately describe something already covered countless times before. (Bailey 2013)

This is the opinion of Jonathan Bailey, a regular contributor to the iThenticate plagiarism blog.

The difficulty researchers experience in writing an original text about something which has been described many times before has meant that many authors, when they repeat a previously used methodology, simply copy the original text along with the descriptions. This happens most frequently with authors whose mother tongue is not English. The mistake they make is to copy the “methods”section without acknowledging the original source.

According to a studyrecently published in Scientometrics² about this fairly common practice, 20% of editors in the biosciences do not consider it important if the copied piece represents less than 40% of the “methods” section.

The study in Scientometrics showed that authors use different strategies when writing the “methods”section in an article. The most frequent practice is to copy methods already published, then to a lesser extent by referencing previous works, and lesser yet a complete re-write of the section in their own words. The decision paths of the Committee on Publication Ethics indicate that the decision as far as plagiarism is concerned is the responsibility of the editor and will depend on the type and extent of the copied part.

The increasing use of CrossCheck in recent years has increased the odds of detecting text plagiarism, so editorial best practices should be defined so as to include some rules to aid decision making, especially for this very difficult section of an article, and principally in bioscience journals.

The article in Scientometrics suggests some best practices rules for writing the “methods” section. They were also proposed last year in the journal Journal of Zhejiang University³, another journal that covers the same topic.

The suggestions here are:

1. Whenever possible, authors should use their own words to describe the methods.
2. If the description of the methods has been published previously, the copying of the description, in whole or in part, should be clearly identified (in bold, italics or in quotes) by the same or different authors, and should make full reference to the original text.
3. Different journals prefer different strategies. Some require the use of a phrase such as ““by standard method”or “as previously described” (preferred by The Lancet), followed by the appropriate reference.
4. Other journals such as Science require authors to include the methods section as an annex, or as part of the supplementary materials.
5. A detailed description should be given only when the method is completely new.
6. If the methods section refers to equipment from vendors (kits), it would sufficient to refer to the vendor manuals.

My thoughts

In summary, even though plagiarism is always unethical, it does not mean that the article contains bad research or invalid results. It is simply a warning that should be investigated.

The progressive incorporation of plagiarism detection technologies to improve the editorial quality of SciELO journals, as we have recommended in many previous articles, and the use of iThenticate and CrossMark should not lead us to lower our guard and put all out hope in computer algorithms. Fortunately, intelligence does not lie in the computers but rather in the reviewers, at least up until now.

In any case, the mistake is in copying the methods section without giving proper attribution to the original source. Finally, it is always better to submit too many references than too few.

Notes

¹ CHADDAH, P. Not all plagiarism requires a retraction. Nature. 2013, vol. 511, nº. 7508. Available from: http://www.nature.com/news/not-all-plagiarism-requires-a-retraction-1.15517.

² JIA, X., TAN, X., and ZHANG, Y. Replication of the methods section in biosciences papers: is it plagiarism? Scientometrics. 2014, vol. 98, pp. 337-345. Available from: http://www.zju.edu.cn/]jzus/download/editorpapers/SCIMsurvey.pdf.

³ ZHANG, YH., et al. Be careful! Avoiding duplication: a case study. Journal of Zhejiang University Science B. 2013, vol. 14, nº. 4, pp. 355-358. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3625532/.

References

BAILEY, J. Should Research Plagiarism Cause an Automatic Retraction? Ithenticate. 2014. Available from: http://www.ithenticate.com/plagiarism-detection-blog/should-research-plagiarism-cause-an-automatic-retraction#.U9Ueymwg_3g.

BAILEY, J. The Challenge of Repeating Methods While Avoiding Plagiarism. Ithenticate. 2013. Available from: http://www.ithenticate.com/plagiarism-detection-blog/bid/94140/The-Challenge-of-Repeating-Methods-While-Avoiding-Plagiarism#.U9Ud9Wwg_3g.

CHADDAH, P. Not all plagiarism requires a retraction. Nature. 2013, vol. 511, nº. 7508. Available from: http://www.nature.com/news/not-all-plagiarism-requires-a-retraction-1.15517.

External link

Committee on Publication Ethics (COPE) – http://www.publicationethics.org

About Ernesto Spinak

Collaborator on the SciELO program, a Systems Engineer with a Bachelor’s degree in Library Science, and a Diploma of Advanced Studies from the Universitat Oberta de Catalunya (Barcelona, Spain) and a Master’s in “Sociedad de la Información” (Information Society) from the same university. Currently has a consulting company that provides services in information projects to 14 government institutions and universities in Uruguay.

Translated from the original in Spanish by Nicholas Cop Consulting.