First authorship dispute between the research leader and the person who did the most important analysis. What is the solution?

Question

I'm involved in a research with a first authorship dispute between the one who lead the research and the one who did the most important part of the analysis.

I give you the facts from a neutral perspective: A had a research idea and gathered a group of people to work on it. In the first meeting he asked to B if they could do some analysis in which the paper would be based on and B agreed. Also asked C to do some other analysis. D, E and F where in charge of minor analysis and to add their expertise on interpret the results.

Once B's results were successful, A started to write the paper with the help of the other coauthors, doing most of the writing work. However, he did not do any of the analysis on which the paper is based. His role was leading the paper and writing.

Once the writing is over, A included the names of the coauthors in order A, B, C, D, E & F. But B complained that he did the most important analysis on which the paper is based, and suggested that the order should be B, A, C, D, E & F. A suggested to B to do it in the former order with the specification that A and B contributed equally, but B replied to do the same but with the latter order. Finally, both want the research to be cited as either A et al. or B et al., since it gives them more prestige.

Now it seems that B wants to withdraw their contribution and analysis to the paper. A wants to continue anyways with the paper without B's contribution since they think that it can be redone with C's analysis. However, the paper will lose a great part of its content. C's analysis support B's findings, but the proper ones are the one that B did. I'm trying to calm them down and to mediate between them, but I would like to know what would be more ethic. For me, it's a new situation, since in my field, usually the one who leads the paper does the most important analysis and writes the first draft.

Supplementary information:

• Both A and B are starting in this topic, so it seems that both want their surnames to be associated with this.
• B argues that at the beginning they didn't know that they would carry the weight of the analysis part. They thought that their part would be equilibrated with other coauthors. That's why they didn't raise their voice before to ask to be the first author, although they recognize that they should have done it before. They are also very shy and maybe that played against them.
• Indeed, A wanted to be the first author from the beginning. So, in that case they should have told B not to participate in this paper.
• B specially complains that A didn't do any analysis and only wants to get credit for others' work

---

Question: What would be the most ethical solution to this problem?

Answer 1

It is field dependent, but in my own experience, the first author ususally goes to the person that did the main investigating / research and the last author goes to the person that took on the supervisory role. The authors in between are normally ordered according to their contribution.

From this viewpoint, I suggest that B should get first author and A last, but, as I said, that is field dependent and might not apply to you.

In any case I think that even though A feels like it would increase their reputation by having another first author paper, if that comes at the cost that B withdraws their contribution and the paper is inferior because of it, A should think long and hard if it really is better for their reputation to have one really good paper with last author or second author with equal contribution than having a bad/mediocre paper with first authorship.

Answer 2

In my experience, there isn't any set criteria for being first author, other than making the largest contribution. Having been on both sides of this in the past, I'm not sure there is an answer, other than maybe to turn back the clock and have an explicit discussion about author order earlier.

I'm of the opinion that author A is in the right. He conceived of the research, delegated tasks, and took the lead on writing. I would consider that a more substantial contribution vs being assigned a piece of the analysis (important as it may be) and taking part in the usual writing and revision process. A also seems to have made a reasonable concession with co-first authorship. This is just based on my own assumptions and field.

Whether or not A is right doesn't really matter though, since a disagreement like this will prevent publication and no-one wins if that is the case. Ultimately, it seems like it is in everyone's best interest to come to some decision, even if someone has to make a concession.

The best thing to do is bring in an outside, impartial person (perhaps a mentor or a more experienced colleague) and agree to go along with their recommendations, whatever they may be. Someone will end up unhappy at the end, but hopefully the paper will go out.

Answer 3

IMHO, both A and B should be reminded that the main purpose of research is advancing the knowledge and not self-promotion (especially by such a meaningless criterion as being the first or the second in the author ordering).

Flip a fair coin and, if everybody is so eager to make his/her particular contribution known to the public, describe it in the introduction. Playing childish vanity games is detrimental to both the science and the public image of a scientist.

Just my two cents.

Answer 4

There is a solution to which the wiser of A or B will agree. It is as follows.

Clearly, there is no clear answer to your question. Had there been one, you wouldn't have posted here. So if you are looking for the solution of your original question, read no further. What I want to do is change your question to a one which can help the situation:

How can we manage this situation in such a way that future collaborations between all coauthors is amicable?

Sit down with both of them individually first to propose the solution that A and B contributed equally, that a distinction of who did more is not readily possible or sensible, so let the order this time be A,B,C... with A and B both contributing equally. Then talk to both of them that the next paper will be B,A,... with both contributing equally (and only if they actually do!). Tell them they are both just starting out and there needs to be more papers.

Answer 5

Assuming there is no workaround the issue, e.g. a sit-down with both authors did not solve the issue, this is how I have seen this solved:

The paper is split into two papers, one paper with authors A, B, the second paper B, A. The fight then moves on to which paper will take the lion's share of the contributions from C, D, E, and F, assuming that A and B will each try to have their respective papers be the bigger one. So C, D, E and F have a conversation without A and B present and split the baby (https://en.wikipedia.org/wiki/Judgement_of_Solomon). Nobody is happy, but everyone moves on.

Both papers are submitted simultaneously with a note to the editor. This requires a compliant editor, which might mean a lower-ranked journal. But that's the price A and B have to pay. I've seen this solution work a couple of times.

But let's assume that they both dig in, neither wants the paper split, and both want you to adjudicate. As @sErISaNo notes, A is in the right.

Answer 6

No clear cut answer but some suggestions.

1. Bring in a senior person and have them talk with A and B (sit in if you want to since you seem to be an impartial party as well and you might help to clarify things when needed as you know the case and both A and B), assuming that you really are impartial in the matter). It is important that nothing gets lost in translation here.
2. Do some research to find what is common in your field (I don't think you mention this). In some fields, senior authorship is as important as first, so if A is willing they might even take that (last) spot and call themselves lead author and corresponding author and give B first (A will then look more senior - only works if this holds standing in your field). As another example, in biomedical research, there are author guidelines and author contribution categories from ICMJE. You can even try to make these contributions quantitative (e.g. for each category, let the authors themselves fill in how much they contributed. As an example: let's say data analysis is a category and the total contribution in this category is 100%. B may think they did 80%, with C, D, E and F each doing 5% (in reality it may turn out to be less than 80% for B after a few back and forth discussions). Same for writing the first draft, where from how it sounds A did 100% (or if others wrote small sections they also get a few percent). Etcetera. In the end, each category needs to add up to 100%. This puts things in a more objective perspective (lots of work though and it requires agreement between authors but this is a good thing).

In general though writing the first draft and getting everything ready for publication as A appears to have done does carry a lot of weight.

Answer 7

A and B should consider a binding determination from a trusted neutral expert

As others have pointed out, this is a lesson in the importance of negotiating authorship order and work responsibilities prior to commencing a research project. Since that ship has sailed, as a suggestion for forward progress, if A and B cannot agree on authorship order (as appears to be the case), they should instead attempt to agree on a process for binding determination of authorship to which they are both willing to be held. It is sometimes the case that parties who are in an insoluble dispute over an issue can nonetheless agree on a reasonable process for binding determination by a neutral third-party.

One sensible option would be to negotiate for a trusted neutral expert (e.g., a senior academic in their field but who is a neutral arbiter on the matter) to review the history of the research agreement and the respective contributions of the parties and make a binding determination on the proper authorship order. Both A and B would have an opportunity to present their "case" to this decision-maker and would agree in advance to be bound by the outcome of the determination. This would allow them to "agree to disagree" but still push forward with the publication under an agreed process for determination of authorship.

Answer 8

I recently read a paper where the authors use an asterisk and a footnote: "Equal contribution, order randomized.": https://arxiv.org/pdf/2112.09332.pdf

Maybe this could be a possible solution.

Answer 9

I faced this problem on two similar occasions, after my long-term coauthor Val and I completed our joint projects. On each of these occasions, we had a lot of new material on our hands. So we could easily afford writing two separate papers, to be published back-to-back in the same issue of a journal. In the first paper, the order of authors was "Michael and Val", in the second -- "Val and Michael".

Do you think you can convince your collaborators to split the paper into two ones?

Things change when the coauthors' job situations are different. Say, when a tenured scientist is collaborating with a junior colleague lacking a permanent position or with a colleague surviving on soft money. In this case, it would be gentlemanly of the tenured coauthor to yield the first authorship to the non-tenured one.

Answer 10

Some fields it is well known the the first author's name is on the paper just to get it out there with the second name being the main contributor. As other have noted name ordering is discipline, institute and country specific. There is no clear ethical order of names as there is no universal agreement to what the name ordering means.

I would echo what others have said that to save confusion and expectation of how the names will be listed should be agreed ahead of time. If conflicts exist then it is over to Human Resources or the Legal Department to start reaching an agreement on what is acceptable or contractually required. Before it becomes an impediment to a project people need to agree to base it on contribution, seniority or lottery.

A possible resolution could be to have all members of the project make a confidential vote on what name order should be. Have a trusted third party organise it and agree the options. e.g. just come up with a ranked voting system where everyone lists their preferred order of names (B, C, D, A) or system for ordering names (random, contributions, lives of report written).

Answer 11

Just randomize order in other arenas, so that each gets the limelight.

OR, you can invent a new designation for authorship. Show principle authors, and then list secondary ones.

Answer 12

The solution is simple and straightforward: List authors alphabetically or in randomized order (and say that you did so in a one-sentence footnote, so people don't get the wrong idea.)

The custom of encoding contribution significance by order of listing of authors is, IMHO, a poor one. More fields should adopt the custom of Mathematics, Theoretical Computer Science and some other fields to list authors in arbitrary order. This dissociates the question of personal aggrandizement / credit / boasting rights from the actual published work. If someone really wants to know who contributed what, they can write and ask; in which case the respondent won't need to provide a ranked list but could describe things in whatever way they saw fit.

Answer 13

In my area (social science) and experience, there is too little information here to say who "deserves" first authorship. Doing the so-called main analysis in a quantitative social science study would be an important factor for author order, assuming it is non-trivial, but hardly sufficient. It would not be uncommon in my area for someone to conceptualize the study, oversee data collection, etc. but delegate statistical analysis to a junior (in terms of author order) collaborator.

It is ultimately about the total intellectual contribution although it is not always cut and dry. When I'm unsure about order, I often give weight to whose participation came first; that is to say, if A and B have roughly equal contribution I would give some weight to the fact A organized the project.