A recent post on Scholarly Kitchen  raised some interesting points about the ethics surrounding citation, and specifically self-citation. Previously, COPE has discussed related issues surrounding self-citation by journals and editors  and citation of preprints . During this forum, we broadened the discussion to include some of the questions related to self-citation by authors in scholarly publication.
Written by COPE Council Version 1 July 2019 How to cite this
COPE Council. COPE Council. COPE Discussion Document: Citation Manipulation. July 2019.
Our COPE materials are available to use under the Creative Commons Attribution-NonCommercial-NoDerivs license https://creativecommons.org/licenses/by-nc-nd/4.0/
Attribution — You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they
endorse you or your use of the work).
Non-commercial — You may not use this work for commercial purposes. No Derivative Works — You may not alter, transform, or build upon this work. We ask that you give full accreditation to COPE with a link to our website: publicationethics.org
The topic for discussion at this Forum was ‘Citation manipulation’. The issue of self citation has been discussed in a number of places before. The focus here is on a form of citation manipulation that qualifies as coercion, where an editor or others affiliated with a journal pressure an author to add citations from that journal for the implied purpose of increasing citation rates and, by extension, journal impact factor.
An editor in chief of a major medical journal in a specialty field is also an author. The editor submits a manuscript to a competing journal in the same field. The manuscript receives moderately favourable reviews and the authors are invited to respond to the reviewer input and submit a revised manuscript. In the communication from that journal's editor in chief, the authors are asked to cite additional references, both of which are from the same journal. The references are only peripherally related to the topic of the manuscript and are within the time frame of publication that will influence the impact factor.
Question(s) for the COPE Forum
• Is using the manuscript review process to increase citations of your journal ethical?
• In a related question, is writing editorials which cite large percentages of manuscripts from your journal ethical?
In summary, is it ethical to use these tools to influence the impact factor metric?
The Forum agreed with the advice given previously by COPE council on this case. The summary is given below. Citation concerns have been raised by Clarivate Analytics as evidenced by recent dropping of some journals for citation manipulation. It may be unavoidable to cite the journal if the subject is a new or niche topic as there may be fewer options for citing peer reviewed work in the field.
However, acceptance of an article should not be contingent on introduction of inappropriate or tangential citations. For reviewers, when authors self-cite for more of the authors’ prior work, an extra check by editorial staff is needed.
The new CEO at Clarivate in charge of impact factor issues is Annette Thomas. She holds an AMA (Ask Me Anything) Reddit on a regular basis according to one of the attendees at the Forum. Questions such as these can be raised to obtain more specific information, for example, the question about citing papers in an issue in the editorial. There is generally a limit of 15% for self-citations but it might be possible for editors to calculate this so that they come up to but not over the limit.
Summary of previous advice: In general, the answer to all of the questions is no. Using the manuscript review process to increase citations of your journal is unethical, especially if the references are not germane to the paper. Sadly, this practice is common. A figure of 15% for journal self-citations is deemed acceptable by Clarivate Analytics.
However, it is very difficult to know whether in this specific case the editor-in-chief's behaviour was unethical. Reviewers and editors routinely suggest papers that the authors may have neglected to include in their review of the prior literature. It is also usual that other relevant papers would have come from the same journal. An editor-in-chief may be choosing references to bolster the level of scholarship of a borderline paper, regardless of where they come from. Were those the only references the editor-in-chief asked for? If the suggested papers are truly wide of the mark, authors can simply ignore them.
It is also unclear if the additional references came from one of the reviewers and are being reiterated by the editor-in-chief or if the editor-in-chief stated them independently. In either case, the tone and context would indicate coercion versus trying to improve a paper ("you are strongly recommended" versus "you could consider"). If a reason is given but the relevance of those specific references is low, then the authors could find other references that address the point better.
Writing an editorial and using citations to manipulate the impact factor is wrong, but in a more generic case, editorials have been written which have cited nothing but the journal articles. For example, editorial introductions to a special issue or themed article set, in which those articles that appear are cited, is not a case of citation manipulation.