Defining who are the authors of your scientific papers, while apparently trivial, is sometimes a surprisingly difficult decision (especially, the first times we do it). As novice researchers, we may operate under conjectures or assumptions about how scientific authorship works, which may not necessarily be true. In this post, I go over several factors that often weigh in into that decision, and I provide a couple of tips and resources about how I would go about taking that decision, ideally.
These days, more of us seem to be writing our scientific papers (be it due to the ending of the academic course, or to COVID-related restrictions on our other activities). Whatever the reason, I’ve also found myself advising PhD students about the issue of setting the authors for their papers and their order – something that looks obvious to the initiated, but rather baffling to novice PhD students and the general public (how can it be that there are scientific papers signed by 5000+ people??1). A common way out of this recurring dilemma is to use simplistic, half-baked assumptions, often born from fear of displeasing others: “the PI/supervisor should be author in all my papers (even those they have not read)”, “anyone who I’ve ever discussed my ideas with should be an author”, or the other extreme, “only I am the author”; or the classic “authors should be in alphabetical order”, etc.
Which one of those is true? Is any of those true?
As with many things in research, the short answer is: it depends. I’ve seen people go about this in many different ways, which I’ve categorized below for your convenience. Some people say authorship…
- … is local (i.e., every lab or group of researchers creates their own rules). To find this out, we can ask around (to our supervisors and our labmates) how things are done where we are. And I mean to really do it explicitly (i.e., don’t just read between lines from side remarks). Don’t be afraid of looking ignorant. In some labs, they even have written contracts you sign about how authorship should be set. But even in those cases, ask around: sometimes the actual local practice does not match what the written (local) rules say. Find out those local practices… but don’t stop there. There are other things to consider. Read on till the end!
- … is social (i.e., every scientific community has its own rules). This is where the 5000-author paper comes from: in some fields like astrophysics, genomics or particle physics, it is not uncommon to have such hyperauthorship2, since the discoveries take a huge collaborative effort involving dozens of teams across the globe. And there are fields that are the opposite of that, where the PhD student is expected to be the only author of publications. Again, the solution is to ask around, not only to our supervisors and teammates, but also outside, in conferences, doctoral consortia3 or summer schools. Ask respected researchers in your field, people you admire… they probably have seen many labs and workplaces in your area. This is especially important if local practices (from #1 above) look fishy or unfair: although we may have limited power to change things around us, we can at least raise the issue of this divergence between local and field-accepted authorship practices (in a tone of curiosity, not judgement – why do we do things like this over here?). We may be doing a great service to our local research environment.
- … is objective. More and more fields (including top journals and conferences) are slowly veering to the idea that all this patchwork of local practices is not really fair or systematic (and research is all about being systematic!), and advocate for more unified and clear-cut criteria for authorship. One of the most widely cited (and clearest) criteria in this regard was put out by the “Vancouver group”, being thus known as the “Vancouver recommendations”4. While these recommendations were put out by a committee of medical journal editors, they have since been adopted by many, many other fields and national-level ethical committees. The recommendations state that, to be an author, you must meet all of the following criteria (even if minimally):
- Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND
- Drafting the work or revising it critically for important intellectual content; AND
- Final approval of the version to be published; AND
- Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
- … is personal. Although the four criteria in #3 seem quite clear, there are still situations where checking these criteria is not necessarily easy. For instance, if we asked for feedback and received only very few comments, is it because co-authors have not read the paper (criteria #3.2 above), or did they do it and find no fault in it? Besides, the criteria above don’t help us much in setting the author order. While there is no ideal solution to all these issues, the best approach I’ve seen so far (stolen from the field of collaborative learning5) is to ask potential co-authors for a self-evaluation of their contribution: basically, ask each co-author whether they deserve authorship and how much they think they have contributed. Additionally, we could also ask them to rate each other’s contributions (if the context provides enough visibility about what everybody does). While this seems to open the door for all sorts of dishonesty, in my experience researchers tend to be a fairly honest bunch (you know, there are other professions where dishonesty is more handsomely paid 😉).
- … is specified. Another increasing trend in many journals and scientific publications is to ask specifically for the contributions of each of the authors, in a more qualitative manner (sometimes called “contribution disclosures”)6, as in: “X did the data analysis and revised the manuscript for intellectual content”; or “Y performed the data gathering”7. In some journals like Science, they even ask what percentage of each kind of contribution each actor did! Hence, we should be prepared to answer such questions when submitting our paper to a journal.
Which one of the five methods above is the right one? Probably all of them, to some extent. My main advice would be to use multiple methods, and triangulate (as they say in qualitative and mixed-methods research8). To help you do that, I have created a simple questionnaire that synthesizes principles #3-5 above (which I’ve added to the “Happy PhD toolkit”). You can create your own form by copying those questions, or modifying them to your liking, for use in your own paper writing processes.
Just before we take off, I’d like to add three additional things to take into account, sort of “meta-principles” to help us apply the five principles above. Authorship is also…
- … about time. Talk about these issues from the very beginning of the co-writing process (remember that forming the co-author team is one of the first steps of the paper writing process I recommend). Discuss with your co-authors what the conditions for authorship will be, and the criteria for their ordering. That way, no one will feel betrayed down the road. And, of course, do not use otherwise fair criteria like the Vancouver recommendations as a weapon to exclude people from authorship unilaterally: if someone contributed to the research substantially (#3.1), we should give them also the opportunity to contribute to the text (#3.2 and #3.3) and take responsibility (#3.4). If they don’t take those opportunities, that’s their choice – no hard feelings.
- … evolving (and we can help). As mentioned in the principles above, the practices and customs of journals, scientific communities and individual labs, all are evolving, as we learn better ways to handle authorship and make the whole publishing process (hopefully) fairer. It seems that most places are slowly going towards a more objective approach (#3)… and we can be agents of change in our local setting/lab as well, even if we are just novice researchers. We can gently raise the issue, ask questions about the reasons for local authorship habits, and talk about these criteria and good practices that we have read about. Very often, being gentle and curious will work better than being pushy and judgmental. But YMMV.
- … our (i.e., the first author’s) choice. In the end, after all this discussion and criteria, it is the first author who normally has the last word. What seems fair to you? Of course, our present choices (and how fair they are) condition whether people will want to collaborate with us in the future: if we completely disregard or rip people off, we will run out of collaborators very fast (and research is more and more a team sport, remember!); if we hand out authorships too liberally, we (and them!) may run into problems down the road as more and more places move to the objective criteria/paradigm (e.g., what will happen if some error is detected and we have to retract our paper?).
Here is a graphic summarizing the main ideas of the post. I hope those were useful. May you author many papers (and hand out co-authorship judiciously)! 😄
Take care. Stay safe.
Do you have other criteria, heuristic or principle to decide authorship or author order? Have you seen any other good (or bad) practices in deciding these issues? Did you find the example form I provided useful? Let me know in the comments below!
Header image by Paul Downey
Castelvecchi, D. (2015). Physics paper sets record with more than 5,000 authors. Nature, 15. ↩︎
Cronin, B. (2001). Hyperauthorship: A postmodern perversion or evidence of a structural shift in scholarly communication practices? Journal of the American Society for Information Science and Technology, 52(7), 558–569. ↩︎
Gable, G. G., Smyth, R., & Gable, A. (2016). The role of the doctoral consortium: An information systems signature pedagogy? Communications of the Association for Information Systems, 38(1), 33. ↩︎
International Committee of Medical Journal Editors. (2020). Defining the Role of Authors and Contributors. http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html. See also this PDF version (section II.A). ↩︎
De Wever, B., Van Keer, H., Schellens, T., & Valcke, M. (2011). Assessing collaboration in a wiki: The reliability of university students’ peer assessment. The Internet and Higher Education, 14(4), 201–206. ↩︎
Sauermann, H., & Haeussler, C. (2017). Authorship and contribution disclosures. Science Advances, 3(11). https://doi.org/10.1126/sciadv.1700404 ↩︎
A more complete taxonomy of what are typical “kinds of contributions” (e.g., conceptualization, methodology, validation, …) can be found at https://dev.biologists.org/content/author-contributions. ↩︎
Denzin, N. (1970). Strategies of multiple triangulation. In The research act in sociology: A theoretical introduction to sociological method (pp. 297–313). McGraw-Hill New York, NY. ↩︎
Luis P. Prieto
Luis P. is a Ramón y Cajal research fellow at the University of Valladolid (Spain), investigating learning technologies, especially learning analytics. He is also an avid learner about doctoral education and supervision, and he's the main author at the A Happy PhD blog.