As a postdoc in a theoretical STEM field, I have had many wonderful collaborations where all authors contribute non-trivially to the final product. However, there was also many "unbalanced" collaborations: where one author does 90% of the work for example. I am not talking about a situation in which a senior professor is automatically given authorship because of politics. For this question, let's say that we talk about collaborations between researchers of the same academic age/position (e.g. all postdocs).

These unbalanced collaborations are often not a problem per se but can be draining and time-consuming in the long run. To give an example, I have had collaborations where I was almost feeling like a PhD advisor to my collaborator, having to explain/redo everything. In contrast, the balanced collaborations that I experienced were always extremely enriching and efficient in comparison.

As a young and inexperienced researcher, my question is: what is the best approach to choose collaborators?

I see two extreme options:

  • Don't worry about it. Accept and nurture every collaboration as they come. If your collaborators are not useful, take the time to help them and make them grow, even at the expense of your own growth. At some point, you might also be the "useless" collaborator and you will be thankful to having been accepted in the collaboration.
  • Avoid unbalanced collaborations by seeking only collaborators you can benefit from. This is probably the best approach career-wise. However, I feel like it can lead to toxic behaviors which might do more harm than good in the long run.

Even though there is probably not a right answer to the question, this is something that has been bothering me lately, and I would benefit a lot from insights of more experienced researchers.

Many thanks!


5 Answers 5


In any collaboration, some degree of imbalance is inevitable. I've been on both sides of the matter, and the question I've always used to guide my collaboration decision is:

"Are all of the collaborating parties benefiting from the relationship?"

There are a lot of ways to evaluate this question, some very pragmatic and some very personal. Here are some factors one might consider: joy or lack thereof, how the relationship affects future job prospects, impact of the work, access to interesting future projects, opportunity cost for working on the project. It sounds from your description like your answer right now is no, so you should probably stop working with the person in question. Before doing so, ask yourself: "what would change this working relationship for me? Is there a way to make that change happen?" You might find some reasonable requests that encourage continuation, but more importantly this will help you wrap things up better. You do have an ethical obligation to bow out gracefully, though nothing is forcing you to and this may not be possible.


I am dealing with this exact problem right now. It's really annoying, especially in fields where authors are alphabetized.

I think in general you should avoid bad collaborations, both where toxic behavior runs rampant AND where collaborators are unhelpful. Sometimes there's a "fun" combination of both in which a collaborator is doing very little work very slowly, exploits your helpfulness, then takes credit for your work in meetings with senior collaborators. I suggest finishing the projects you have with people like this as quickly as possible (or bowing out/reprioritizing your time to more promising projects if you think it will be too draining) and then not working with them again in the future.

There are a ton of good collaborators out there, and as you grow your network you really shouldn't waste your time on toxic collaborations or bad collaborators. This doesn't have to be toxic in and of itself: if you find people you work well with and collaborations where everyone brings something useful to the table and pulls their weight, you'll generally have better experiences and produce better research. In a sense that's looking for people who will benefit your career, but hopefully the benefits are mutual and overall you have a good time doing science together.

I also think what goes around comes around, especially in author-alphabetized fields. Maybe people who do 10% of the work will get their name on a lot of papers as postdocs, but when it comes to faculty hiring, reputations are more established and faculty really pay attention to any comments about work distributions from your coauthors. People will take pity on a weak researcher and give them a postdoc sometimes if they have networking reasons to do so, but they are not going to extend the same charity when it comes to permanent positions.

  • 2
    wow I had no idea there were fields where authors were alphabetized that seems bad. Basically change your name to Aanthony Aardvaark..
    – neuronet
    Oct 29, 2020 at 20:15
  • 1
    @neuronet there's no point, being first on the list doesn't mean anything. People tend to refer to papers based on the most recognizable name (or name most associated with that line of work). Overall I like it, less grubbing over credit and more collaboration. Economics does it too, and there's an interesting study on implicit bias in that field about how women are severely under-credited bc people assume they didn't do first author level work.
    – Well...
    Oct 30, 2020 at 7:33
  • Yes I guess I am used to the other way where it matters. But no matter how much people think it doesn't matter I guarantee it matters, even if it only because of the effects on people who are not in those subfields. (But not only b/c of that). You are right though it would save a lot of headache people are annoying when discussing these things I've had multiple arguments about 'Who's on first'
    – neuronet
    Oct 30, 2020 at 15:49

This is not a zero-sum game. Including someone as a co-author is almost always mutually beneficial, as long as the person contributes something. Who cares whether it's only a few paragraphs in the discussion and a few comments or a bit of software? They did improve the paper a little bit. In the end, the credit you receive for the paper isn't reduced by including someone else - maybe even on the contrary: intuitively, I value papers with many authors higher, because that means that more knowledgeable people put their thought into it. It will also be useful for future collaborations. So I find it's not helpful to be picky about co-authors, as long as they actually help.

Thus, I would argue that an imbalance regarding the amount of work put into a paper is not a problem at all. What is a problem of course is when someone does not actually contribute or, worse, is slowing down progress. In that case, you should be honest with yourself and the collaborators and quickly find a way to improve the working mode, or get rid of the collaboration (which can also mean simply not actively pursuing it anymore). Note that I'm all for helping collaborators grow! Find canvases for other people to paint on; make them look good. However, these people should also bring something to the table; otherwise, the collaboration is simply not a good fit for the two of you.

In my experience, the checklist for a successful collaboration goes something like this:

  • All collaborators have an intrinsic motivation to actively work on the project. This motivation must be significant enough to bring people to work on this instead of the 856 other interesting projects they have on their desk!
  • There is a clear and mutually agreed-upon assignment of work packages to collaborators. This should (roughly) be clear from the start. Ideally, the collaborators have complementary expertise, making the task distribution obvious.
  • Everybody gets along well with everybody else, and the working relationship is productive. There's no sense in collaborating with people who are stubborn, lazy, arrogant, or simply not fun to be around - even if they are highly respected experts. Interacting with your collaborators should not leave you with a bad feeling.

A caveat regarding "not a zero-sum game": I'm in a STEM/engineering field; this may differ in other fields.

  • If you're competing for jobs and someone does less work but gets the same credit, that definitely hurts you in the competition. Some people who play this game do it deliberately to get their names on more papers than they can support with their own work ethic, making them look artificially more productive than they are, giving them an unfair edge. It gets annoying at the postdoc level when job competition is fierce.
    – Well...
    Oct 30, 2020 at 7:42

Consider a PhD student collaborating with his/her advisor, which is obviously imbalanced. Although the advisor must teach the student, and it takes the student more time to complete a task than it would the advisor, the advisor is often reducing his/her workload by teaching the student instead of performing the work on their own. For collaborating with your peers, you could think of it similarly - you may have to teach them some technical skills, but if it saves you time, that can still be a beneficial collaboration for you.

As an advisor, teaching a student can be a separate reward in addition to doing research. As you said, there can be other benefits in fostering collaborations besides completing papers quickly.

Ultimately, I think the choice is yours to make and depends on your goals and priorities. There's no obligation for you to perform 90% of the work in a collaboration. If your collaborator isn't pulling his/her weight, you could choose to do most of the work to obtain a publication, or instead focus on other research topics.

  • 2
    Consider a PhD student collaborating with his/her advisor, which is obviously imbalanced — Yes, but which way?
    – JeffE
    Oct 29, 2020 at 20:35

Unbalanced is the norm so they are best to be avoided.

Never write with anyone else unless you know for sure that they have something critical you need to use.

If you do collaborate write a contract that spells out who will do what and how much of the total effort. If they come up short then they forfeit all rights to have their name on the paper.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .