We received four letters of accusation on the same published article from 4 different email addresses (namely A, B, C, D) consecutively with the interval of each being 1 month.
The first letter of accusation identified several issues with the use and interpretation of statistics, and noted that no reference number was provided for the study’s ethics approval. In response the editorial office started an investigation in accordance with the COPE guidelines, reviewing the whole peer review process of the article and contacting the corresponding author. The author gave a point-to-point reply to the issues and supplied related materials including the original data on which the table in question was based, and also the ethics approval document. The author confessed honest mistakes in the article and said that they would like to publish an Erratum.
Before we could reply to complainant A we received a similar letter from complainant B. As there was no evidence of academic conduct, we conveyed the author’s explanation to complainant A and B but both complainants raised more concerns regarding potential manipulation of data.
We then asked the author to provide all original data and further explanation of the complainants’ question. After receiving the author’s reply, we asked the opinion of two statisticians who completed verification of the original data and found that the data of all tables and related conclusions are repeatable. Our statisticians provided their comments and some questions for the author to answer. The author again replied point by point, and the statisticians came to an agreement that the data and the findings of the article are supported by the original data. However, the statistical methods used are unclear, and more detailed info of correction should be provided in the Erratum. The author was asked to write an Erratum.
The third letter of complaint was similar again, and arrived before we had replied to the two previous correspondents. We asked complainant C to provide their full name and affiliation and they did so.
We gave an official reply to complainants A/B/C regarding our investigation results and plan to publish an Erratum but they soon replied with further complaints. Meanwhile, the fourth letter of accusation came with more complaints about the article. The author began to lose patience, concerned that the complaints would be endless. We have the same feeling and also have a sense that the complaints may be personal in nature, not just academic.
Questions for the Forum
- What can we do next based on the current investigation results? Is it appropriate to ask the complainants to publish a Letter to the Editor and the authors to write a response letter for publication? The author has agreed to make the complaints public and give a reply in public.
- Can we conclude the case based on our current investigation results or shall we contact the ethical committee for further investigation?
- In retrospect, do we need to ask the author to give a point-to-point reply to the complaints from each complainant? The author was willing to cooperate at first but lost patience in the back-and-forth communication. If not, how should we reply to the complainants?
- Could you please confirm if our process of the case is in accordance with the COPE flowchart?
- Are there more efficient ways to handle the multiple complaints like the above case?
- What could we do if we receive new letters of accusation on the same article in the future?
This type of scenario was familiar to many participants at the Forum and several possible solutions were offered. First, it was noted that if the issues raised by the complainants all relate to data interpretation then they may be based on a difference of opinion rather than a fundamental flaw in the article. In this case the journal should focus on determining if the data are valid; if they are convinced that they are, then they may not need to investigate every complaint and instead give the same response to each complainant. Second, if the criticisms were raised and addressed at the peer review stage then the journal may also choose not to take it further, but to state to the complainants that they are satisfied with the authors’ response. It is also possible to present an article with any errata and updates as a single package. Any further complaints must then address the whole package, avoiding the danger of this becoming a recursive process of addressing essentially the same complaints over and over again. This solution also permits the journal to compartmentalise the different issues: they can go ahead and issue a correction if necessary while continuing to look into other issues.
Third, the journal could ask their IT team to look at the IP addresses of the emails of complaint in order to probe their origins. This would reveal if they are all in fact from the same individual, which could suggest a more personal programme of harassment and would merit getting their institution involved. The journal editor is within their rights to ask correspondents for their names and affiliations, but that they are not obliged to provide them. COPE advice regarding complaints is that anonymous complainants be treated with the same level of seriousness as those who have identified themselves.
In essence, it appears that the journal has followed the correct course of action, and if they are satisfied with their processes and their plan for a correction, then they may have done all that is in their remit. If they feel that there are sufficient grounds to think that the complainants may be correct in their allegations then they could contact the author’s institution to suggest that they explore the matter further. The appropriate person to contact would be the research integrity office, or a named research support or governance officer.
If the complainants have not provided sufficient grounds for concern about the article then there is no need to correspond with them further.
Related resources
-
Responding to whistleblowers when concerns are raised directly, COPE flowchart
-
Responding to whistleblowers when concerns are raised by social media, COPE flowchart