Tips for supervising grad students?

In our newest "how can we help you?" thread, a reader writes:

I am an advanced assistant professor getting ready to start supervising both an MA student and a PhD student for the first time. I would love to hear advice about how to do this well.

This is a great query. I haven't supervised grad students myself, so I'm not sure there's much that I can say. But it would be great to hear from readers--not only those who have experience supervising grad students, but also from grad students themselves! Do any of you have any tips or advice on how to supervise MA and PhD students well? Any do's or don'ts?

Comments

[Current Grad]

I think you want to strike a balance between showing what good a good work ethic looks like and not being a mental health drain on your student. A PhD student should know that it is important for them to be working every week on their research. Setting clear expectations and having consistent gentle prods works toward this goal. But ensure that you did not pressure them into the height of the expectations. They don't know what a reasonable amount of work is, and they look to you for guidance on what you think they can achieve.

[UK Postdoc]

One of the most helpful things my thesis supervisor told me the first time we met was this: "Your thesis needs to be x words and you have y days to write it. But you will only keep about half of what you write, so you should aim to write 2x/y words per day on average."

Especially at the start of writing the thesis, when the project is still so very open, this was a helpful guideline to me. Once I really got started, I didn't need it anymore as I know what to work on next.

I think it's also good to discuss with your student (a) your expectations from them and (b) their expectations from you. Perhaps you mainly expect your student to hand in work regularly, which you will then comment on and discuss with them; but your student might expect you to guide them towards relevant literature, or to think through the global issues relevant to their work.

Relatedly, although this depends on your institution, your student's career stage and their career aims, but I would try to think about how much you want to 'coach' them into things like publishing, submitting to conferences, etc. My supervisor never suggested I publish something or submit it for a conference, so I had to figure out these aspects of academia myself. I was self-confident (/over-confident) enough to decide to submit some papers, but I can imagine other students silently waiting for their supervisor to give them the go-ahead.

Two other things my supervisor did that I really appreciated. (1) He made me start by writing a literature review of the five or so most important papers in the field, which was a nice way to get started for someone who didn't know where to get started. (2) He often asked me if I had any reading suggestions for *him*. The latter was especially helpful later on, when my research had specialised a bit more. But it also made me feel more like an equal to my supervisor, and to feel like *I* could set the agenda.

I hope that's somewhat helpful!

[Postdoc]

Writing from the student’s perspective; haven’t advised grad students myself.

First, help them interpret your feedback on their writing or on anything else. I had an advisor who wrote as many pages in comments as the number of pages the paper he was commenting on. This suggested to me that the paper was broken beyond repair. But I later found out it didn’t mean that at all. He thought it was publishable nearly as it was! This was a bad, discouraging, confusing way of giving feedback. So don’t just give them comments on their work, also help them understand what those comments really signify.

Second, this one is probably pretty obvious, but don’t avoid criticizing their work. A different advisor rarely gave negative, critical feedback on their advisees’ work, and the work suffered as a result. Obviously, critical feedback should be respectful and geared toward improvement of one kind or another. But it’s got to be given.

These two suggestions are obviously related. Give students constructive criticism, and help them understand that feedback. Does it mean the project is dead in the water? That it is promising but needs work? That it, like all projects, has its problems but overall is good as is? And so on.

[Michel]

Be timely with your feedback. Establish expectations early and explicitly, and stick to them--tell them at the very beginning that they can expect feedback within two weeks, for example, and stick to that policy. If you won't be able to in a particular instance, tell them so, and tell them when they can expect the feedback instead. Do not *ever* allow them to go for six months or a year or more without feedback. Remember that your comments *save them* six months' worth of work.

Do not make them feel like they're somewhere below undergrads on your priority list (e.g. by scheduling meetings late at night--such as in the 20h00-22h00 range--after the undergrads have been meeting with you since 16h00). You have lots of different priorities, but make sure your grad students are somewhere near the top. They're capable of a lot of independent work, and don't need a ton of herding--but they *do* need *some* direction. Having grad students is a real privilege, and you should behave accordingly.

If you have funding they can use, tell them; don't wait for them to ask you.

*Do* introduce them to people at conferences, invite them to dinner with you, etc.

And take some time to professionalize them--e.g. at the beginning, have them look around at the CVs of people at different career stages in various institutions, and chat with them about their findings. Make sure they sign up for PhilEvents and have a PhilPeople/papers profile. Tell them about the main subfield publication venues and their quirks, the main conferences, etc.