Watch the introduction to "Claiming institutional affiliations" with Ana Marušić
Standards and guidance on "institutional authorship"
Authorship is the currency of research and academia. But what about author affiliations? Research assessment and university ranking systems, as well as national accreditation systems, put demands on institutions and researchers to boost publication productivity. This has led to an increasing practice of inflating the number of affiliations claimed by an author on their publications, which is not solely due to increased research collaboration or mobility.
Journals and research institutions rarely provide guidance or standards on the definition of “deserving” affiliations or the number of affiliations per author on a publication. Multiple undeserved affiliations may be considered as research malpractice and may affect different stakeholders and the integrity of the published record.
During this Forum we will engage in a discussion on “institutional authorship” – the definition of affiliation on a publication, standards and guidance on their number and challenges to trustworthy institutional representation on publications. We welcome your thoughts and feedback.
Questions for the Forum discussion
1) Is there a definition of institutional affiliation i.e. institutional authorship, on a publication?
2) When does institutional support merit inclusion as an author affiliation on a publication?
3) Are there accepted standards on the number of reported institutional affiliations per author on a publication?
4) Who should create such standards?
Comments from the COPE Forum, December 2023
NOTE, comments do not imply formal COPE advice or consensus
- In many institutions the first and last place in the authorship list have particular prestige as they denote the lead author and a senior author respectively. This can lead to authors being encouraged to include their affiliation by several different institutions.
- For authors, the institution can be the current employer, the place where the most substantial part of the research was done, or the body who provided the funding. It may not be as clear-cut as ‘where the research was done’.
- For journals, however, it is important to be clear which is the institution that has given ethics approval, will pay any publication fees, and will be the point of contact for any issues with the publication.
- Definitions of eligible affiliations are still rare. The Publication Manual of the American Psychological Association says that only organisations that have made some substantial contribution to the work should be included.
- Journals could be clearer in how they define the institutional affiliation.
- Some journals do not permit more than one affiliation to be given per author.
- Some of the strain in the system arises from increasing use of affiliations by corresponding authors to claim eligibility for publishers’ Open Access funding schemes. Should publishers be using something other than institutional affiliation to determine eligibility for this funding? It might be helpful for publishers to specify corresponding or billing affiliation instead.
- Are affiliations as important for non-research articles? Who is accountable for opinion pieces or editorials, or for articles signed only on behalf of the journal?
- Affiliations matter to institutions who rely on rankings to boost their recruitment and income. Demanding that authors include them as an affiliation is a way to boost their indicators of productivity. However, it can become a case of gift authorship if it does not accurately represent where the work was done or where responsibility for it lies.
- Researchers who are at an early career stage may legitimately have several affiliations because of institutional mobility, or they may have no permanent affiliation. ‘Current affiliation’ may be a more useful way to declare their status.
- Researchers who are retired or who are not attached to any institution (independent scholars) are not well served by the emphasis on affiliation and may feel discriminated against.
- Should authors be expected to declare how much of the work was done at each institution listed? Or which institution is the employer and which funded the research if these are different?
- The issue goes beyond authors and publishers: author affiliations and funding organisations are treated separately by some indexing and registration databases such as CrossRef. Publishers need to be clear in what they are collecting so that the data can be registered accurately.
- Databases of persistent identifiers such as the Ringgold Identity Database may be of use to publishers in tracking and identifying institutional affiliations. Ringgold Database identifiers can be linked to the International Standard Name Identifier database and via that, to The Research Organization Registry.
- More guidance on affiliations would be welcome since it has become so important as an indicator of research output, quality, credit, and prestige.
Further reading
-
What’s in a name? How false author affiliations are damaging academic research Vivienne C. Bachelet, LSE Blog
-
Misrepresentation of institutional affliations: The results from an exploratory case study of Chilean authors Research article
- Octopus affiliations Khaled Moustafa, Arabixiv Papers
A summary of the discussion at the virtual Forum event will be added to this page shortly.
Page updated: 12 December 2023
We welcome comments on this discussion from COPE Members and non-members here:
Your comments
You must be logged in to leave a comment on the COPE website. If you are not a member of COPE you can register to create a guest account.
Comments will not appear immediately. We review comments on our website before publishing them, to ensure they are respectful and relevant.
- Login to your account or register
to post comments
Comments
There is another possible problem that you didn't mention, namely a concealed affiliation. Affiliations are often thought of as an employment or student relationship, but they can also come in the form of financial support or non-employment membership in an organization. Some research has been criticized in the past for having been motivated by concealed funding. Examples include funding by tobacco industry to dispute health concerns, or funding by the petroleum industry to combat competing climate change research.
It feels like affiliations could use a taxonomy, much like the CRediT taxonomy for contributor roles. That way we could cover things like "research started while employed at..." or "research concluded while employed by..." or "research conducted while visiting ..." or even "Research independent of the author relationship to...". The exact taxonomy would require considerable discussion.
to post comments
I feel somewhat symphatic with this proposal yet it is quite bureaucratic. The same could be done by adding short explanations to the affiliation list, like "current address" for the contact address of an author while the affiliation during the actual research would be plain. If the author has been a visiting scholar during much of the research, again plain affiliation to the organisation where the work was done with the contact address with note "permanent address". I think that accepting two affiliation is totally valid e.g., in cases like above. More would need a justification. Funding does not make affiliation. The scientists working in a research organisation on a foundation grant (quite common in my home country) should list the organisation where they work as the affiliation even in the case they do not have an employer status.
to post comments
I believe that only the major primary affiliation should be included in research papers. All other ancillary connections and vaguely related positions should not be listed in the publication. If a free-for-all facility is provided by journals, there will not be a limit to which authors possibly could go to.
There should be agreed concensus on the part of journals in this regard.
to post comments
At our journal we discussed the problem of submissions from paper mills and faked affiliations. During this discussion the suggestion came up that submissions should be admitted only if coming from well known academic or industry affiliations. This, however, makes a big problem for independent researchers, for example retired academics or people like me, when I was unemployed in maternal leave, working from my kitchen table in Chemical Graph Theory together with my husband and using his academic affiliation, which was not a problem at that time (the nineties). I will have the same problem in near future when I'll loose my academic affiliation because I am retired. My opinion is that freedom of research allows every researcher publishing their research in academic journals, independently of affiliation.
to post comments
I'm not sure what the problem is here. We permit author to simply insert their department/reserch unit and university/college (inc town/city and country). On the rare occasions that the contributor works outside higher education we note as appropriate.
to post comments
authors must be affiliated with the institution/organization where they primarily work. This is because every moment they spend contributing to a publication is ultimately acknowledged, rewarded and supported (completely or partially) by this affiliation body. If the author receives financial support from another institution/organization/company, he/she should acknowledge this support. But this does not grant the author the right to affiliation. if the author is hosted as a contributing researcher in one of the institution /organization laboratories, then he/she has the right to claim affiliation (maybe a second affiliation). Claiming more than one affiliation must be justified by the author and acknowledged by the affiliation body. Journals may generate a system for the verification of the affiliation of the contributing authors in which the head of the department/the manager / the director has to verify this piece of information (the author affiliation).
to post comments
In our portal (BAHIANA Journals, Brazil), are considered proper afilliations only those with which the author maintains formal contractual ties of employment either for teaching, research activities and/or clinical practice. Independent researchers may submit papers and are permitted to declare themselves as such. For those authors who do not have an employment relationship with educational and/or research institutions, the student relationship may be considered, usually the highest academic rank to be obtained by the author at the time of submission (i.e., doctoral student, master's student, undergraduate student). Laboratories, research groups and institutions in which the author is a postdoctoral fellow, visiting professor or consultant are not considered as affiliations for the purposes of publication in order to avoid having authors taking advantage of the status of institutions where they might be working on in a short-term basis and/or without contractual ties. The authors are requested to comply with that at the time of submission. As we are unable to verify all claims, we have made ORCID authentication mandatory for all authors of any given submission to any of the journals of our portal, allowing for the independent verification of the authorship claims. Mandatory ORCID authentication for all authors may also be effective in curbing ghost authorship.
to post comments
Starting point: whether we should indicate the authors' affiliation (or independence from their involvement) at all, and if so, on what basis. Why is it important/mandatory? For whom is it important (stakeholder perspectives)? And only after that move on to the definition (if required).
to post comments