What does a good fit between research advisor and student look like?

We should expect research mentors to customize their advising to meet the needs of their students.

I’ve been around long enough to see an innumerable mass of PhD students wind up in laboratories that are a bad fit for their professional development.

Sometimes the student has great capacity for independent work and end up being micromanaged. Or it’s a disorganized student who doesn’t get support for becoming more organized. Or a student who is not fully accepted within the discipline because of their identity and the advisor doesn’t understand this are don’t provide the support needed to struggle with this challenge. You get the idea. The advisor’s approach doesn’t happen to fit what the student needs.

These students often just get screwed over. They might be able to finish up and manage to move on, but not as well as they would have otherwise. Some bail out for a different mentor who is a better “fit,” while others wash out without finishing.

Some folks look at these situations, throw up their hands, and say that it was an unfortunate circumstance. That this wasn’t the right lab for the student. “This lab is suited for [adjective] students but not [adjective] students.” In this line of thinking, the mismatch is a mistake that happened during recruitment and both parties are to blame.

That is a harmful and mistaken perspective. When you deploy the concept of mentor/mentee “fit,” you’re surgically removing the advisor’s responsibility for effective mentorship. Which is their job.

When some see bad advisor/student “fit,” what I see is bad advising.

If you have a student in your lab who is not getting what they need, whose fault is that? It’s simple: it’s you, the advisor. Because you’re not delivering what they need.

This is a hard truth for me to share because it cuts myself rather deeply. Because a whole bunch of my former students are able to give an accurate ledger of my shortcomings that is would be far longer and deeper than it should be.

If you were to ask me what my mentoring “style” is, I’d say that I’d like to think that I customize what I do for the needs of each student, but realistically I’m more hands off than instrusive, and the students who most thrive are the ones who have the confidence to move forward with their work but also are adept at diagnosing when I need to be contacted with questions. In other words, I realize that my low maintenance students do better than the high maintenance ones. This is not a mentoring style, this is a mentoring shortcoming. It is a deficiency on my part. I know this. I take responsibility for it, even if I’ve had the luxury of experiencing little accountability.

When I bring someone into my lab, it’s my job to give them the support they need to thrive. And if I initially misdiagnose these needs and they are greater than I anticipated, then it’s on me to put in the time and emotional labor to fulfill the commitment that I made. If a student made a bad choice of me as an advisor, it’s not their fault, it’s my fault for being the bad advisor. Simple as that. In my preemptive defense, I think some of my biggest screwups as an advisor haven’t been about me not willing to put in the time or concern, but rather about me misunderstanding what the needs are. But of course: impact»intent. If you mostly know me from here, you might have landed at the impression that I’m a very thoughtful and sensitive guy, because I spend so much time advocating for more thoughtfulness and sensitivity. But I know all too well that when it comes to student mentoring, I’ve come up short in these areas. I know this, so I have to own this if I’m going to call out that bad “fit” mentoring is just straight up bad mentoring.

An effective mentor is like this entire wooden box, not just one of the sockets.

To be clear, I recognize that research mentorship is one of those metaphorical two-way streets. Both sides need to clearly express responsibilities and expectations, and these are best formalized in an agreement, or individualized development plan, or whatever you want to call it.

Advisors of research students need to build their plans to meet the needs of the students. And every person is different. While it’s the cookie-cutter approach might be easier for the advisor, who will hope that students are made of the right dough, why should we be taking shortcuts when it comes to cultivating the wellbeing and success of our people? That’s downright exploitative. Yes, it might take more time to individualize mentorship for our students. And yes, some students are more work than others. That’s part of this gig. (You knew that when you took the job, even if it wasn’t part of the training or selection process, but that’s a whole ‘nother post.)

I was done right by my dissertation advisor. In the years after I left grad school, I ended up getting to know subsequent cohorts of students who passed through the same lab. We’d chat on occasion, shared rental houses at conferences, and so on. I quickly learned that the advisor who they knew was different from the one that I knew. And each of them knew a different person from one another to some extent. Why is that? Because we all had different needs, and he gave us different kinds of support. I needed the freedom to run my own project and also the focus to narrow it to the proper scope. I needed to learn how academics interacted with one another and how to be an equitable spouse (and eventually, a responsible parent). And he gave all those things to me, one way or another. Everybody else who went through the lab needed other stuff, and so they got other stuff.

Inflexible mentors with unreasonable expectations come in all flavors, but one thing they all have in common is that they’re not willing to put in the emotional labor to build relationships with all of their students as individuals. This is often a gendered distinction, and is often why women in academic departments end up doing more of the undervalued labor of supporting struggling students when the men who are their advisors aren’t up to the task. It’s getting exhausting to see, in so many different places, women who are left to pick up the messes of their colleagues who are failing at their mentoring responsibilities.

What’s to be done about this? Well, how about we make sure that we all are trained in mentoring. And make sure that mutual expectations between students and faculty are well established in writing. And get rid of linear power hierarchies in student training programs so that one person cannot have the capacity to (inadvertently or otherwise) destroy the progress of any trainee. And what else?

---

Note: I wonder if you’ve gotten to the end and noticed that I haven’t referred to the advisor or the mentor as the “PI.” I did that intentionally. Do you think that our linguistic shift to referring to research advisors as “PIs” is possibly a consequence or cause of our changing perspective on the nature of this relationship? And do you (like me) think this is a little problematic?