How soon to fire rude and chaotic PhD student?

Question

From experience and from advice I have typically seen, PIs put students through a ‘trial’ phase between a month and a year, to determine if they are a good fit, and if not, they are let go from the lab.

Unfortunately, I recently hired a student who I find rude, disrespectful, abrasive, chaotic, etc., and immediately rebelled against the project idea I had assigned him without a good reason. And it is immediately clear to me I would not want to be around this person for even a short period, let alone the duration of the PhD. This contrasts sharply with other students I have been working with.

This person is coming in with a fairly strong CV. But if he is going to be an a-hole to work with, it doesn’t matter to me whether or not he is a technical genius.

How bad would it be if I told such a student I’m not interested in working with them further after a couple of weeks? I would have no problem doing that if this person was say a prospective undergrad or master’s student, but I am a bit hesitant to do this in such a short timeframe with a new PhD student. Any advice?

Answer 1

I think to meet the expectations of a mentor (this student is still your mentee until they are not), you owe it to them to give a chance for improvement.

That doesn't mean just wait and see how things go. It would involve you explaining to them the reasons that you are considering not continuing as their supervisor, including constructive/positive suggestions for improvement. For example, the feedback should not be "You are chaotic!", the feedback should involve suggestions for how to manage what you perceive as chaotic in a way that would work for you and put them on a path towards successful completion of a PhD. This is likely to be an uncomfortable conversation for both of you, but it's part of the job you signed up for.

As others have recommended, I would brush up on policy to understand what the official steps are because if there is some sort of formal probation status you might want to start that sooner rather than later.

Once you've definitively made up your mind, though, it seems to do everyone involved a disservice to wait any longer.

An exception to my statement that you owe them a chance to improve after feedback would be if they're an active threat or harm to other students. Your description does not seem to put them in that category, but you might make steps to protect others in the meantime.

Answer 2

I would like to challenge the framing a little - this is a relationship problem.

In your question, you say "and immediately rebelled against the project idea I had assigned him without a good reason"

I don't know if this is different between countries, institutions, etc, but to my knowledge, in Australia, a PhD student is fundamentally in control of their PhD research, with the supervisor tasked with providing advice and guidance, rather than instruction.

Obviously, this is influenced by funding questions, and so the PhD student would naturally be constrained to some degree as to their research.

However, I notice your description of the student as rebelling against the project idea you "had assigned him". This sounds less like a PhD student learning to be an independent researcher and more like a PhD student being used as a cheap form of academic labour.

It may have merely been a choice of words, or this may be more typical in your country or institution.

I note that you said in a comment "But my job is not to rehabilitate people into decency, it is to run a well-oiled lab". No, it isn't. You are the student's supervisor for their PhD. In this role, your job is to mentor the student into an effective researcher. If this isn't the role you have in mind, then you shouldn't be supervising PhD students.

Now, don't get me wrong, this doesn't excuse the student being "rude", but it may explain it. The question to ask yourself is, how much reasoning and guidance did you provide to the student regarding the task you were asking them to undertake? How much control did you provide to the student regarding the task?

---

Let me use an invented example. Suppose you're running a chemistry lab, and your student's official project is about finding ways to manufacture a particular compound (this may be simpler than a real PhD in chemistry, but as a mathematician, I don't generally work in a lab). Suppose you have an idea for what could be a really efficient way to manufacture the compound.

Version 1: You go to the student and tell them to test this specific method, with a detailed plan - their role in the lab will be to gather the appropriate components, assemble the experimental apparatus, and run the experiment. When they're done, they give the results to another person to analyse to determine whether it was a success.

Version 2: You provide the student with the idea and task them with exploring it. You answer any questions they might ask, and ask them to provide you with a plan of attack before they put anything into motion. They come up with a variation on what you were thinking, and you give them the okay to proceed with the lab work.

In the first version, you aren't mentoring a student, you're instructing a lab tech. The student isn't a lab tech, and thus they are likely to balk at being treated as such.

In the second version, you are mentoring a student. You are providing oversight and guidance, but it's the student's research being done.

---

This may not be the case, of course - you might be properly trying to supervise the student, and they could be pushing back because they have issues with any kind of influence, or because they've got issues of some sort. That's something for you to figure out, before you make any irreversible decision.

And to answer the explicit question, I would say that, if you were to fire the student without at least making a real effort to guide them to be better (which takes more than a few weeks), then you are likely to gain a reputation for being unreasonable with students - if only because the student is capable of telling their version of the story.

There is no particular length of time you need to wait, but rather an amount of effort you need to put in. If you put in that effort, and the student doesn't respond, then it is likely that the student will come to the same conclusion as you - that it's not working - and you can arrange for either their transfer to a new program with a different supervisor, or for their ending of their candidature.

Answer 3

Letting go of a student is a costly action: you have committed resources (time and funding), and there is also a reputational effect to consider. You can only drop students so many times before people start wondering whether you might be the problem. This can make it harder for you to hire and retain future students; in my institution, bad mentoring performance has promotion and tenure implications as well.

If you drop a student only after a few weeks you better have a very good reason. Of course I don’t know your situation and this student may well be an awful human being, but perhaps it would be useful to have a serious conversation with this student? Something along the lines of:

I’m very excited to mentor you for the next five years, and you clearly have excellent skills in doing X. However, your attitude towards me and other lab members is not acceptable . I know you want to work on Y, but in my lab we work on X. If you realize that Y is what you want to do for your PhD then I strongly suggest that you spend the coming semester looking for a new PhD advisor. If you feel that we need to change how we work together I’m more than happy to hear you out, but do keep in mind that I have certain policies on how I run my lab. If these policies are not a good fit for you, then again I suggest that you seek another advisor.

While I do agree that having a protracted bad relationship with a student hurts you both, I do think that a few weeks is really quite short. Maybe they just need your assurance and support; maybe their partner just dumped them. I would give it at least a semester before making this call.

Answer 4

I would guess that it is better for both of you to do this soonest, rather than later. I'd suggest a face to face meeting with them to let them know that you will drop them within a week if they don't improve their behavior and attitude.

Doing it sooner permits the student to seek someone else before they spend more time in the program.

Have a chat with the department head about this also. You need to know how it will be received overall.

I once had to drop a Tai Chi student in their first class when it became obvious that they wouldn't trust me. I recommended another instructor with whom they might connect. Not the same, I know, except that this situation also requires a long term trusting relationship. Why let it fester.

Answer 5

Unfortunately, I recently hired

I'm guessing that the employer is your university, not you personally - even if you interviewed him and even if you had secured some/all of the funding. You don't have absolute authority over him (nor should you).

a student who ... immediately rebelled against the project idea...

So, you and the junior researcher had agreed to work together: You had discussed some subjects, directions, plan for research, and agreed to pursue them. Then immediately he became intransigent and rebellious? I am somewhat suspicious of this description of events.

immediately rebelled against the project idea I had assigned him

What do you mean, assigned to him? You had just agreed on things. Did you now assign to him something against your agreement? You did not say he went back on the agreement, reneged, denied having agreed, right?

rebelled against the project idea I had assigned him without good reason

Without good reason, or without any reason? You did not tell us, but your phrasing suggests, that there is some "no-good reason". A person going as far as "rebelling" (rather than, say, being unmotivated, or not focusing enough on the task etc.) - typically has a reason which they find very good, enough to risk their whole Ph.D. candidature and relations with their advisor.

Given your attitude as a PI - as much as it comes through in your post - I would not be surprised if your junior mentee actually has a pretty good reason, or at least a legitimate one.

Now, if I were advising your mentee, I would probably say: "Oh, brother, you seem to have stumbled onto a condescending and authoritarian boss, perhaps you should just switch advisors" - just like you are thinking of firing him.

But - perhaps both approaches are wrong, and what you and your mentee (independently and together) should really do is Invest effort in communicating your grievances to one another in a way that is not perceived as mere confrontational hostility; and in receiving, parsing and understanding such communicated grievances from the other side.

If you both understand what each party wants, and is willing to give/compromise/accommodate vis-a-vis the other party - you can make an informed, reasoned decision about whether you should continue your supervision relations.

I find rude, disrespectful, abrasive, chaotic, etc.

It's possible that he is being disrespectful; but remember that respect must be earned. Have you behaved respectfully towards him? Have you acted so as to earn his respect? I wonder.

Also, perhaps other would describe your group as bland and regimented, and that person as being dynamic, spontaneous, passionate, forthcoming...

Answer 6

If there is no good match between the student and you, and you really can't see yourself mentoring the student, then it would seem like there's no real point in letting it go on an instant longer than it needs to, as you're wasting time for both of you that could best be spent in other ways.

That said, how you go about separating is mostly a matter for your department and your graduate program. Also, in some cases, any student might be better than no student, so you might think about whether you'd be able to replace the student in a timely way, whether you'd be burning bridges with a graduate program that may come to think you're too selective, and hesitates to send more students your way ...

My own opinion with respect to the last issue, though, is that planning to keep the student around until you can replace them isn't a particularly good option for the student, and the path should be avoided.

Answer 7

Reactionary personalities are amongst the most difficult to work with. In businesses we usually just fire them, regardless of their technical competency -because their personality gets in the way and they are not efficient anymore, and their technical competency goes to waste, since most work is strongly collaborative.

But a PhD is not your ordinary professional collaboration.

Is doing a PhD seen relatively more as a personal journey (of the student) with a map/compass (the supervisor)? Or is it seen relatively more as a production line of capital goods, where the student is the capital good under construction (designed and destined to be productive itself at a later date), while the supervisor is the manager of the production line, which in turn is part of the factory/University?

It matters how you see it but also how your institution sees it.

If the latter, find a a way to fire them in an appropriate way, and other answers suggested various such approaches.

But if the former, you are the map, you are the compass, not the captain of the ship. A probably useful approach in this case would be to demand (not just ask) from the student to come up with and justify and argue in favor of, their own project idea, if they don't like what you propose. They have to learn, or acknowledge, that if they want to be "their own person", they are obliged to shoulder the responsibility and the heavy burden, that comes with it. This also will put to the test their knowledge and ideas-generating potential.

And then, you will have objective grounds for assessment: if they propose something that you don't consider worthy, you have every legitimate reason to finish the collaboration. But you may like what you see, and then, maybe it will be worth the trouble.

Answer 8

At this point, the goal is just as much about professional development as it is technical competency. While you may not be the best to offer social interaction development, I think part of what your goal is has to be ensuring the student's talent in a field you both care is presented in a manner that fosters teamwork. So "how you fire him" is with documented feedback, given as objectively as you can give it. Maybe even have a 3rd party involved. That way, you can make sure no negative biases or prejudices of yours are influencing the outcome. I'd suggest multiple meetings and then see if the relationship improves. If it doesn't, then let him go. Just make sure its not out of malice on your part. You don't want to drive a bright mind away from your field just because you two can't seem to get along now. Neurodivergence is real, so not every human is going to interact easily. It doesn't mean it is impossible, especially for a greater good.