If you have a bad advisor in grad school


A couple weeks ago, I emphasized that most PhD advisors are really good.

In a haphazardly conducted poll, one in four people reported their PhD advisor that was not caring or helpful. Crappy advisors may not be the norm, but we still have 1 in 4 too many.

I’ve seen a variety of situations, choices, and outcomes over the years, and would like to share some thoughts with grad students who are experiencing a bad PI. I’m hoping those of you who have gone through nasty experiences might be able share insights as well. I’ve just been a bystander, and there should be many more voices than my own.

When dealing with a bad PI, I think there are two big questions:

  1. What can be done to ameliorate the situation?
  2. When should you bail on your PI and move to a new lab or even a new institution?

How important is your PI to your academic career? Well, usually, very important. But everybody gets different things out of grad school. Here’s a list of good things that grad students might get from advisors:

  1. Mentorship
  2. Collaboration
  3. Connections to influential people
  4. Academic ownership of your research agenda
  5. Letters of recommendation
  6. A supportive community in the lab
  7. Middle authorship opportunities
  8. Freedom to work on what you want to work on
  9. Access to proprietary stuff (data, methods, field sites)
  10. Access to specialized and expensive stuff (facilities, instruments)
  11. Funded research assistantship
  12. Funds for research
  13. Support for life balance, such as scheduling for family matters
  14. Editing
  15. Continued advocacy to advance your career
  16. Connection to a prestigious academic lineage
  17. Someone that picks up the tab when dining out

I don’t think everybody gets all that stuff from their PhD advisor. I hope everybody can get what they need. (In my opinion, items 1-7 are necessary ingredients for all grad student experiences, though they don’t always have to come from the PI.) Depending on your discipline, some items on the list might be more important or less important.

What can do you do to help minimize the negative effects of Bad PI Syndrome? I’ve seen grad students disadvantaged or screwed over by their PIs in a variety of creative ways, so this isn’t a one-size-fits-all kind of issue.  It might help to take a look at the list of items above — and also the things I’ve failed to include — and identify what you’re not getting and consider how you might get them.

For example, if you don’t have a supportive group of peers in your lab, are you connecting well with nearby labs? If you’re not collaborating with other people in your own lab, size up the potential for building collaborations elsewhere (which is important regardless). Is your PI supportive in spirit but not an effective mentor? Can you find others more experienced in your life who can take an interest in your career? Senior grad students and postdocs are often more important mentors than PIs, anyway. If you’re not going to be getting an amazing and detailed letter of recommendation from your PI, are there other people on your committee who you know you can count on at least?

The first step to liberating yourself from a bad lab is to diagnose that you are being treated badly and unfairly. I can’t tell you how many grad students I’ve met who are experiencing horribly manipulative conditions, but have decided that this is the norm in academia. Having a manipulative PhD advisor is not normal. Having PI that takes advantage of you is not acceptable. Having a PI that puts their career advancement ahead of your own professional development is not something that you or any other grad student should have to tolerate.

You don’t have to suffer in silence. You have constructive courses of action available to you. One place to start is the chair of your graduate program (assuming it’s not your PI), and also any other trusted professors in the department. Of course, some of these people might be friends with your PI, so this might not work out — but this is clearly something to consider. Your Graduate School also has Deans and Associate Deans whose job is to deal with these kinds of problems.

Of course, there is a power differential, in which the PI has authority and influence. However, universities are — at least in theory — designed to deal with the abuses that come with this power differential. Usually universities side with professors unless a student can clearly demonstrate an incontrovertible abuse of power. Keep in mind that you’re probably not the first one, also that horrible PIs often have a long history of being horrible. If you talk to chairs and deans about your concerns, they should recognize your situation and also be in a position to intervene in a way that protects and supports you.

Before you take major steps, it’s a good idea to discuss these things with a trusted mentor. There are consequences for being a whistleblower, but it’s also a different matter to find a way to repair or bail out of your own situation with a clean landing. Every situation is different, so that’s why I’m making a point of advising any specific course of action. I just want people to be aware that there are people out there in every university whose job it is to protect grad students from bad advisors. These aren’t easy things to deal with, I realize.

Grad school is a springboard. If you’re continuing in academia after you finish, then having a quality experience in grad school is critical to land opportunities for further success. If grad school is a poor quality experience, then it will be much, much harder to uncover opportunities once you finish.

If you feel like your time in grad school isn’t working out specifically because of your PI (meaning = you don’t have access to items on the top of the above list), then aiming to survive until you graduate is probably not a plan for success. Grad school isn’t something that you should merely survive — it’s a training academy and how you emerge from the training really matters. A solid launch from a PhD program is necessary for what happens afterwards. If it doesn’t kill you, maybe it’ll make you a stronger human being but it won’t necessarily help your academic career.

If you’re not planning to try to stay in academia, then having a supportive PI is, maybe, even more important. Because you need to train for your non-academic career plans while in grad school. Preparing for a career in sales, patent law, writing, industry, K-12 teaching, the FBI, whatever — you can’t just jump into this after you finish. Having a PI who is on board with your plans is huge, if you want to be able to build an “alternative” set of skills while doing your dissertation.

Should you stay or should you go? I can’t answer that question for you, obviously. I’ve seen several people stick it out, only to finish and then discover that they probably shouldn’t have stuck it out. Is that you? I have no idea, I don’t even know you, but this is a question to consider.

I suggest that the criterion for “stay” should not be whether you limp out with a PhD diploma in hand. In my opinion, a more useful criterion is whether you will have the ability to successfully move to your next step when you finish. If you are staying, and you’re not getting what you need, be sure that you have strong mentors to work with you to diagnose precisely what you do need.

Beware Toxic PIs. Having a suboptimal PI is perhaps tolerable, but Toxic PIs cannibalize their grad students. You know who these people are. They may initially support and encourage them, but as work progresses, the PI becomes less and less cooperative as the student becomes viewed as an academic competitor instead of a protégé. The students — if they ever are allowed to finish — are deprived of ownership of their own work. Then they graduate without a good recommendation from their own PI, disputed ownership of their own work, and probably not a strong set of publications.

I’ve known a few Toxic PIs. In a high-functioning department, prospective students are steered away from these labs before they even start. (I was the beneficiary of this kind of information once, and now I see that the students who were in that lab before and after me went through some horrible things.) Students in a Toxic lab often skid out of the ride with a PhD, but emerged lucky to finish rather than ready to roll.

I know one person who is traditionally “successful” in academia, who did his dissertation with a Toxic PI. Right before finishing, as things went south, he chose an alternative course. (He may or may not have fully chosen this course — his PI probably wouldn’t have signed his thesis if he submitted it.) Instead of limping out with a dissertation that would keep him in academic shackles — he started over. Brand new. He went to a new university and found a new PI, and did a new and spectacular dissertation. The papers from the first dissertation never came out, at least not with his name on them. On one hand, it was five years of his career that were just gone — poof. There’s no real evidence in his professional life that he had that experience. On the other hand, he had a huge amount of knowledge, savvy and skill that he could apply to his second dissertation — the one that he used to build a very successful line of research.

Do you have any stories, with good or bad ending, or in progress, that you’d like to share?

18 thoughts on “If you have a bad advisor in grad school

  1. I had a toxic advisor. I vowed never to burn any bridges on my way out of that situation, so I’ll avoid too many details. Of the 17 items on your list, I got numbers 9, 10, and 12, and none of the others. By the time I figured out what was going on, and that my (horrible) experience was far from the norm, I was ready to drop my attempt at my PhD. I tell people that the most important factor was a personality conflict between myself and my advisor – we simply did not get along.

    Your mention of “the student becomes viewed as an academic competitor instead of a protégé.” rings very true, though in my case that impression came and went. Sometimes I felt like I was viewed as a competitor (which is ludicrous), sometimes like a mistake my advisor was now stuck with (which was at least plausible, if deeply unpleasant to contemplate).

    As far as support from other people, I don’t feel like I got much of that. Maybe the occassional boosting comment now and then, but at no time did I feel like a member of the faculty – the Chair of the department, or somebody with “graduate student issues” in their job description – would have ever taken my side over my advisor in a conflict or even a simple disagreement. The power imbalance is simply too great, because the university has invested so much more in a professor – even a new prof, pre-tenure, as my advisor at that time was – than in a graduate student. I could complain all I wanted, and get plenty of sympathy from other graduate students, but nobody was in a position to actually do anything. For example, how do you think it would go if another professor, hearing my complaints, privately approached my advisor? The next lab meeting I would have been taken aside and been subjected to his anger; I could have forgotten about any request (time on equipment, orders of supplies, access to data) I might have had on my mind. If that continued as a pattern, where would it end? “Happy graduation” is not the answer to that question.

    That might be a false impression generated by where I was and how I was feeling at that time, but I remain convinced that my options were quite limited and that help from any other person would have to be built on actions I initiated myself. I’m not talking about “find this person and talk to them” as an action I would initiate that would count. I mean something more like “File a formal complaint, with evidence, to upper administration or agencies beyond the university, such as the government of the province”. Lawsuit-level actions, which would obviously be huge and probably disproportionate and certain to leave a trail of distrust and career self-destruction in their wake. There was no option between “tough it out, somehow” and “quit”; every other option I could see was “quit +”, like “quit and sue” or “quit and take somebody with me”. All I could hope for was that there was some consequence happening behind me as I walked out the door, that something would change in that department as a result of my exit.

    Quitting that PhD was a hard decision but ultimately the right one for me. Obviously, every situation is unique, but I can recommend to others in similar situations that the exit is not the end of an academic career. I started over in a different department at a different university – I tell people it’s been a long, strange trip through graduate school – through some contacts I’d made during that time. I finished my PhD at the beginning of this year and now I am a post-doc and I feel like pretty much everything is pretty much on track.

    I really like your list – I got (and continue to receive, as appropriate) ALL of those things from my second (successful) PhD advisor. Take 2 was much, MUCH better. One item I might add is “Income”. How many graduate students have been sunk by disastrous personal finance? I don’t mean “bad with money”, I mean “Oops, the university forgot to pay you, not my problem, get back to work”.

    I say more on old posts on my blog. I’ve included the URL in the “details below” field, maybe somebody will stimulate me to write more on my blog (about this or other things).

  2. Two comments to add nuance:
    (1) When searching for an external mentor (which is invaluable when you can get one!), try to find one who you can trust to tell you if you are part of the problem. External mentors should not be echo chambers, although it’s not always easy to tell the difference. Grad school does not always teach easy lessons, by design.
    (2) As a PI, there have been times that I have been a bad adviser. Fortunately, my group has been willing to constructively deal with those instances so that I can overcome my lapses and learn from them, something that I am tremendously grateful for. Obviously, this is a different kind of issue than the “bad adviser full stop”, but I think it’s important to point out that there will always be stumbles. It’s how they’re fixed that differentiates a good PI from a bad one.

  3. One issue not addressed is whether you go straight for the PhD or stop along the way for an MS. If you enter into an MS, you can use the transition from MS to PhD to get out of a bad situation without burning bridges because the expectation is that you came there for an MS degree and will be leaving to pursue a PhD. At many schools, you could transition into the PhD at the same institution if you are enjoying that experience. But once in a PhD Program, many students feel trapped or locked in. More and more students are skipping the MS degree but it can be an important stopover on the way to a PhD in terms of research experience, fostering knowledge of what a good final step would be, fostering skills in dealing with the ups and downs of academia, etc.

  4. Advisors come in all shapes and sizes. I surmise that few advisors are universally “bad,” but that there are many more that are mismatched with their specific grad students. This can also happen when the advisor has even a positive preconception about the student (e.g., in one case a professor I knew had a very high opinion of the coding and mathematical skills of Asian grad students, and burned out his Chinese PhD student with a type of workload that she could not handle).

    Many, perhaps most, mismatches occur when the student and the advisor do not understand or appreciate each other’s aspirations, and a significant number also happen because there is are differences in the desired type of advising (the advisor is okay with a once-a-semester update but the student wants regular weekly meetings, or vice versa).

    To the extent that many doctoral students are themselves seeking academic or research lab positions and thus want to build good publication records, perhaps the most objectionable behavior (short of outright criminality) I have seen in advisors is to get the student to do all the work and give them the results so that they can write the paper and give the student a footnote, if that. This is of course a gray area because research assistants do not always deserve co-authorship, but some advisors, even at R1 universities, are definitely skating on thin ice ethically.

  5. Just discovered your blog and I love how honest and straightforward you guys are! Is there an email I can use to ask you guys specific questions?

    • Someone, if you’d like to discuss a certain post relevant to your circumstances, feel free to comment. My contact info is in the ‘about’ section. I should add that haven’t been able to do real advising or career counseling for people who cold contact me from reading the site. But an open-source conversation here or on Twitter probably could be more helpful.

  6. I’m wondering if you can say a bit more on (7) Opportunities for middle authorship which is considered necessary, but not necessarily by the PI (or do you mean necessary from the PI?) Who is first author on these papers? Can you say a bit more about what the PIs role is here?

    • I’ve noticed that one element of being competitive for (some) jobs is to have a bomb-ass publication list and metrics by being middle author on a bunch of publications. If you leave grad school only being first author on your thesis papers, then your CV isn’t going to look as strong as if you worked your way into other collaborations. I think a strong PI supporting their students would make sure that students have these opportunities, especially if they’re looking for a career in academia. It could be that this would be from collaborating with other members of the same lab, but it’s also possible that the PI could arrange for a student to develop a skill set and opportunity to collaborate with folks in a different lab, somewhere in the world. If the PI’s not doing this, then the student can try to create these relationships on their own, which is totally normal. But ideally, as a student is getting deep into grad school, they’re using their techniques/analyses/expertise in a project that they are not leading.

      • There is no denying that middle authorship is proliferating, I’m just not sure that as a PI I should be feeding into this, and if its a good thing for science in general. Should we focus on doing good science or ‘play the game’ and what are we teaching our students in the process, not to mention the selection for a particular mentorship model that becomes prevalent at the faculty level in the future. I was recently a reviewer on a paper that claimed 7 people had equal first authorship. Stephen Heard has discussed whether two different students can claim the same paper as part of their thesis. Isn’t only our own ethics that stop us from taking this too far? Hypothetically, speaking couldn’t all the grad students from one lab all be co-first authors on the same papers and use them all for their thesis?

        • I think experience being part of multiple collaborative teams is a valuable experience. I think the point is not just the product but also the training that goes with it. People who work with others in grad school to the extent they have earned middle authorship can demonstrate how their skills are useful more broadly and that they’re capable of fitting into different kinds of teams.

Leave a Reply