assigning PhD advisors

[bluefooted]

Hi.  Curious as to how your institution/dept reaches PhD application decisions, and how students are assigned advisors? 

My institution seems to have a very low bar for who they admit. And if an applicant lists you as someone they would like for an advisor, then you get them.  I'm finding this very hard to manage (pre-tenure).  Furthermore, we don't guarantee funding, so I'm constantly scrambling to find funding for the students.  In my PhD program, our advisor funded us.  This is not the norm in this dept though.  However, I find it difficult to advise if the students aren't working for me.

With that, any other advice you have for advising?

[Puget]

It would help to know the field.

In my field we accept students directly into labs (or sometimes two labs with an existing collaboration where it makes sense to co-mentor). We would never try to foist a student on a PI who doesn't want them-- if the PI isn't interested, we don't accept them. In my department the admissions committee does control things to the extent that if a PI wanted someone who we didn't think was strong enough, would not give them an offer, but not in the other direction of giving an offer to someone the PI doesn't want.

[Ruralguy]

I know that in my PhD program, which I know now seems like it was in the pre-millenial Dark Ages, we entered without advisers, at least not official ones. We could seek funding from a professor, but typically they didn't want to commit until you passed the qualifying exams. Some did anyway, if they figured you were a shoo-in, which I wasn't. However, our program did guarantee tuition remission for all who could get a teaching assistantship or research assistantship, and unless you really blew it, you'd be given at least a TA load for your first two years. In any case, I sought out advice from a prof with whom I was doing some research and who also taught one of my classes (he turned out to be my PHD adviser later), and the grad adviser and others. Eventually, I was given research assistantships, but if my adviser had gaps in funding, they probably would have raced to get me some sort of TA funding. Fortunately, that didn't happen.

Those who graduated more recently or work at R1's with PhD grad programs could perhaps post their experiences.

[Ruralguy]

Oh yes, I should add that in my program it was very rare to have anyone dropped on an adviser. if it happened, it was probably early on as an experiment, or maybe very late as well, if that other prof had some related funding and the student could do the work. That sort of thing was very rare.

[Parasaurolophus]

In my (humanities) PhD program, it mostly worked that way, too. A student would declare interest in certain areas and be auto-assigned two preliminary supervisors. Over the course of the first few years, as the student refined their interests, they might drift away from one or both and towards someone else, at which point they'd have to approach that person to formalize the relationship.

I know the department tried to balance incoming classes, however, so that it wasn't the case that every year Prof. X got two students but Prof. Y got none for sixteen years in a row, etc. They sort of tried to make it so that everyone had at least one advisee at any given time, and no more than about four. Of course, we did have strings of people with 0 because they either worked in an area that got no applicants, were too difficult, etc. But mostly it seemed to work out OK.

[Caracal]

Oh yes, I should add that in my program it was very rare to have anyone dropped on an adviser. if it happened, it was probably early on as an experiment, or maybe very late as well, if that other prof had some related funding and the student could do the work. That sort of thing was very rare.

Yeah, in my program, you were actually admitted with an advisor or sometimes two. You wouldn't be accepted unless the advisor wanted to work with you.

[ab_grp]

Advisors in my program were chosen post-comps, so after a fair number of courses.  I'm not sure how it worked for everyone, but I just asked the professor with whom I had the most research interests in common (had taken several advanced topics classes with him).  The process could have been different for students who sought funding and ended up TAing or RAing for particular professors during their time there.  Before applying, I had identified a professor whose research I was initially most interested in, but I ended up changing interests over the years, and he wasn't a very active professor.  In any case, I would have to doubt that any professors were forced to take on particular students.

[mleok]

In the R1 math department that I work in, incoming graduate students are assigned an initial advisor in the student's stated area of general interest, and their primary role is to approve their course selection each quarter and to provide general advice, but there is no expectation that this initial advisor becomes the thesis advisor.

Students are generally funded through teaching assistantships, and it's typically only after they have completed at least part of their qualifying examination requirements that they start working more closely with a potential thesis advisor. Typically, the students who end up working with me have taken one or more graduate topics classes with me, and started working on something for the final projects in these classes, and followed up with a reading class or research credits.

For us, thesis advisors are never assigned, nor is there an obligation to accept any student who asks to be your student. At least in my field, where there is no real need for cheap, lowly skilled labor, as we do not run large repetitive experiments in the lab, there is generally no benefit to working with a student who is extremely weak.

In the departments (like engineering) where faculty are expected to support their graduate students with research assistantships, I generally find that a student is only admitted to a graduate program if a faculty member explictly expresses an interest in admitting the student, and commits to funding them.

[Kron3007]

Oh yes, I should add that in my program it was very rare to have anyone dropped on an adviser. if it happened, it was probably early on as an experiment, or maybe very late as well, if that other prof had some related funding and the student could do the work. That sort of thing was very rare.

Yeah, in my program, you were actually admitted with an advisor or sometimes two. You wouldn't be accepted unless the advisor wanted to work with you.

Same here.  I think this is pretty common in most science fields, but I do know it is not universal.  In my field, students usually choose to work with a specific researcher with common interests rather than choosing based on the school (obviously good programs tend to have high quality researchers, so there is some relation).

[arcturus]

In my field, it is normal to admit students without consideration of whom they might choose as an advisor. However, we are still selective regarding qualifications, both in terms of academic record and prior research experience. We have had a few instances of students leaving the program because they could not find a research advisor at the appropriate time (i.e., they worked with various faculty during their first few years, but could never settle on a project/advisor), but most of our students connect with a research advisor during their first year. Our funding is largely through TA positions, so there is not such a concern regarding whether an advisor has funding at the time of the match. We do not *guarantee* funding, but the expectation is that we will fully support all students making good progress.

Keep in mind that various faculty may hold different expectations regarding supporting graduate students (using "support" in both the financial and research contexts here). In my case, one of the senior faculty thought he was being helpful by steering almost all of the incoming students in my direction, even when they specifically stated that they were interested in a completely different subfield. Not unexpectedly, these students eventually left my group and completed their degree with someone else. Until I learned what was going on (and even after...), I thought I was failing at the research advising aspect of my job, since all of these students were leaving my group.  As a junior faculty member, I thought I had to take any student who expressed interest in my research. Now I know that (1) I don't and (2) it is helpful to ask more details about the student's interest before accepting them to my group. If you have a trusted colleague, I would recommend asking whether the department expectation is that you take all students who ask to join your group. If not, you should start to be more selective. That will save you many headaches - both in terms of the time you sink into less productive students, and in terms of making certain that your limited resources are best directed for both your students' success and your own.

[Hibush]

This situation sound weird, so it is a good question.
I'm in a model similar to your training and to what Puget and mleok describe. It cost you several hundred grand to take a student, plus your career success depends on them performing well. Advisors are necessarily choosy.

If you use neither funding nor potential for success as primary screens, what do you use?

Do you take the student you are handed, then if they are no good you tell them to withdraw? (I'm assuming if you don't pay them and don't actively advise them they have to find another advisor or withdraw...but maybe that's ssuming too much.

Do you expect the students to struggle academically and financially, offering them whatever funding scraps you can find, and let attrition take its toll passively?

[mamselle]

People are getting better about saying "in my field" (so, there is learning in fora bipeds! Yea!) but they really need to at least say, "sciences" or "humanities" since those two differ so very widely when it comes to grad work and how it's arranged.  (Having worked as an EA in one and done my grad programs in the other...)

Market segmentation, and all that...

M.

[Caracal]

Oh yes, I should add that in my program it was very rare to have anyone dropped on an adviser. if it happened, it was probably early on as an experiment, or maybe very late as well, if that other prof had some related funding and the student could do the work. That sort of thing was very rare.

Yeah, in my program, you were actually admitted with an advisor or sometimes two. You wouldn't be accepted unless the advisor wanted to work with you.

Same here.  I think this is pretty common in most science fields, but I do know it is not universal.  In my field, students usually choose to work with a specific researcher with common interests rather than choosing based on the school (obviously good programs tend to have high quality researchers, so there is some relation).

That was in humanities. Maybe its a model that is actually more common in sciences? It always seemed like a good plan because it meant that from you avoided the problems I'd hear about sometimes elsewhere where students assumed they would be working with an advisor and then would arrive to find that person wasn't interested in their project, or in taking on new students.

[mleok]

Oh yes, I should add that in my program it was very rare to have anyone dropped on an adviser. if it happened, it was probably early on as an experiment, or maybe very late as well, if that other prof had some related funding and the student could do the work. That sort of thing was very rare.

Yeah, in my program, you were actually admitted with an advisor or sometimes two. You wouldn't be accepted unless the advisor wanted to work with you.

Same here.  I think this is pretty common in most science fields, but I do know it is not universal.  In my field, students usually choose to work with a specific researcher with common interests rather than choosing based on the school (obviously good programs tend to have high quality researchers, so there is some relation).

That was in humanities. Maybe its a model that is actually more common in sciences? It always seemed like a good plan because it meant that from you avoided the problems I'd hear about sometimes elsewhere where students assumed they would be working with an advisor and then would arrive to find that person wasn't interested in their project, or in taking on new students.

The model is much more common in the sciences primarily because of how students are funded. In particular, there is an expectation that faculty in many science fields support their students with research assistantships, so in that situation, securing an early commitment from faculty is highly desirable.

[mythbuster]

In the biomedical sciences, the standard approach is for students to rotate through 3-4 labs over the course of the first year, while taking coursework. The selection of the lab is then a mutual and much more informed decision on the part of both the student and the mentor. Students who do not "find a home" may do extra rotations, or rarely leave the program.
   In the few cases I have seen of students who don't find a home lab, there are significant issues that would prevent the student for succeeding in the program. These can be significant deficits in training, or issues of personality or work ethic.