Small group discussions
Case 1: Adding references after final proof
When their paper is published, the authors of a study are
surprised to see that changes have been made that were not in the final proof
sent to them before publication. A comment has been added to the introduction,
emphasising a view contrary to the article's premise, and two references have
also been added, both of which are to review articles written by the editor of
the journal. The authors complained to the editor.
In his response, the editor says that he did this because the
article was insufficiently balanced. The references he added were to
comprehensive reviews of the subject that were published recently in the journal
of which he thought readers should be made aware. The journal in question
belongs to COPE.
- Does this represent editorial misconduct?
- What should the editor have done?
- If you were the subcommittee invited to deal with this case
in the light of the COPE code, what would you do (step by step) and what
recommendations would you make? - What problems, if any, do you foresee in implementing the
code? - What do you think the outcome will be for the author,
editor, and for COPE?
Comments
- Editors may want to make substantial editing changes as part
of acceptance, or suggest ways in which the manuscript can be improved,
including additional references. But the way in which this has been done is
wrong. - The editor should have cleared the sentence and the references
with the authors before acceptance, although how much an editor can influence
authors before acceptance and how much authors could feel pressurised to accept
the suggestions is a rather murky area. - If the article is insufficiently balanced, a better process
would be for the editor to have invited a commentary posing the opposite view.
Why wasn't it brought up during the peer review process or at least
independently of the editor? - Could this be construed as massaging the impact factor as the
review article has been written by the editor of the journal and published in
the journal? - The authors should go back to the journal, then the journal
ombudsman, then to COPE. If the editor is found to be in breach of the code,
COPE could ask for an explanation. If none was forthcoming – no apology
published for example, the authors should go straight to owners/publishers of
the journal. - The code says that editors should be responsible for
everything, but it neglects the responsibility of authors. If published in an
author's name, the
author is ultimately for what it says; the editor only monitors the process, and
this case illustrates why the editor can't interfere with judgments
made by the author. - The responsibilities of authorship are outlined in the Guidelines on Good
Publication Practice. - Where would the author stand if the journal inserted a
couple of sentences in the author's name that provoked litigation? - It is important to publish a correction as an well as an apology that would link
back to the original paper. The authors could also decide to retract the paper.
Case 2: Changing decision after acceptance
An author whose paper had been accepted for publication by a journal receives a
letter form the editor saying that the journal has too many papers to publish
and that he is no longer able to publish it. The author complains, but the
editor stands firm. The author refers the case to COPE.
- If the journal is a COPE member, what should COPE do?
- Conversely, if the journal is not a member, what should
COPE do?
Comments
- This represented a fundamental problem of misconduct. But
it's possible
that the editor may have discovered something else and used this as an excuse.
But it is not acceptable behaviour. - The guidelines refer to investigation of members of COPE. But COPE could still
apply pressure to have the decision reversed. - There could be good publishing reasons for not going ahead, for example,
allegations that consent had not been obtained from all the authors, and the
authors responding that the paper has already been accepted so they are not duty
bound to do anything further. - Could publication threaten the journal's financial future?
This is unlikely: surely it should just be a matter of delay and increasing
the journal's backlog? - Journals should always have a conditional clause, whereby they stipulate that
acceptance is subject to no other issues arising. In a recent COPE case a paper
clearly duplicated something that had already been published. It was therefore
rejected at a late stage. It had been accepted subject to compliance with peer
review changes. - Journals' commitment to authors should always be phrased in such a way so
as not to represent a binding contract. - The implication in this is case, however, is that there is no other reason for
turning down a perfectly good paper, other than space or money. - Papers can be withdrawn after acceptance, if, for example, the same paper has
been published elsewhere, or an editorial board member has uncovered fraud,
because the author has then broken the agreement between him/herself and the
editor.
Case 3: Publication bias
An editor receives complaints from readers that the journal is biased in the
papers it publishes relating to a specific issue. It has published only those
reflecting one point of view. One author has written 14 review articles over the
past five years, the most recent of which makes the same points as the previous
one, and cites mainly work from that author and the editor of the journal. A
reader refers the case to COPE. The journal is a member of COPE.
- What should COPE do?
Comments
- This is not a case of editor misconduct. The readers may well have a legitimate
complaint, but they should write a letter to the editor or the editorial board,
but the editor has not done anything wrong. - If the editor does not have a letters column, the readers should write to
another journal, or still write to the board. - The code deliberately tries to steer clear of decisions
about content, because that's the responsibility of editors. - Let the reader decide. If it's such a biased journal, people will stop
buying/subscribing to it.
Case 4: Massaging the impact factor
An editor of a new journal wants to boost the impact factor, and discovers that
one technique being used elsewhere is to ask authors to include references to
papers already published in that journal. The editor in chief questions this
approach, but the editor is adamant that his competitors are doing it, so he
should be allowed to do it too, if the journal is not to be unfairly
disadvantaged. The editor in chief refers the case to COPE.
- What do you think of the editor's approach?
- What issues does this case raise?
- What should COPE do?
Comments
- Would we feel same way if the editor had asked the author to strip out
references to another journal? The process was dishonest, and the paper
would be skewed by adding references so scientific content. The reader suffers. - COPE should write to ISI and request that self referential citations should be
discounted, and the editor in chief should recommend that this practice be
stopped. - Is it inappropriate to bring to authors' attention,
relevant references that have been published in that journal? Because they
happen to be in that journal does not undermine their validity, and it could
be laziness on the author's part. - There is a difference between that and demanding that
publication is dependent on including at least 25% of references from that same journal. But many
editors would not have the time to trawl through MedLine, looking for other
valid references. - Would the addition of these references necessarily skew the
article? Conversely, editors can't monitor everything the author decides to
include. - It is unethical to selectively point authors to references in one journal if
there are equally good references in others, but editors do tend to be more
familiar with their own journals. - The integrity of the journal and the quality of the material should be the
decisive factors, not where a reference is published. - A peer reviewer who is expert in the field should be able to advise the author
on the relevant research, and not just what has been published in the journal. - Peer reviewers should be required to look at the references
and suggest ones that might have been missed; it is their job, not that of the editor. - What happens if the peer reviewer recommends acceptance,
providing the authors cite two papers that have been written by the peer
reviewer? And if the authors refused, would the article be rejected? - This would constitute a deliberate attempt to massage the impact factor, and is
a clear case of unethical behaviour. - Citing a paper is not an editorial question; it's a
scientific question. If there are no scientific reasons, it should not be
cited, irrespective of where it has been published. Rogue citations are a
problem. Steering citations partly depends on the country in which the
reviewer or editor live, so it's important to prevent that bias and give examples of
publications that authors have missed. - Should editors take responsibility for ensuring balanced citation?
Citation is biased anyway, but editors should do all they can to prevent it.
Requesting a reference should only be done after checking PubMed for something
better first. On smaller journals, editors tend to act as surrogate peer
reviewers, so blurring the roles. - Courting publicity through electronic tables of contents
and press releases increases the visibility of the journal, which could be
construed as massaging the impact factor. An editor might feel the need to do
this because the journal is not issuing press releases and if it can't compete, will struggle in
terms of revenue/supplements, etc. That could have major consequences for a
small journal.