Evaluating potential supervisors on the basis of past graduate student results

Question

I am lucky CS senior undergrad who has received offers from a number of good departments in the USA. I am currently going through the usual factors for evaluating departments and supervisors. In the process, I've noticed a common trend in advice: most of the advice that I've gotten has not mentioned looking at where most of the students end up (e.g. quality of the thesis, do they get post-docs, coding jobs, industry research jobs, etc). Most advice (that I have read) emphasizes, not unproductively, the adviser-student relationship and the supervisor's research interests. This observation has led me to several questions:

1. Is it appropriate to ask potential supervisors what most of their students end up doing and why? Is there some other way to talk about this with potential supervisors?
2. How much should this factor into one's decision? Obviously, a lot of what happens to me is up to me and luck, but also obviously my supervisor plays some role in this beyond helping directly with my research.
3. Is there an easy way to figure out this information? Anything better than googling lots of names?
4. Are there more qualitative ways of looking at this question besides just looking at bare results? (Given the relative scarcity of positions over time and the amount of luck involved, I am amenable to the argument that simple stats are useless here.)

Answer 1

It's perfectly acceptable to ask a supervisor what their former students are doing now, and I think the majority of supervisors would be more than happy to discuss it. They might even give you email addresses of their old students -- I myself have been contacted many times by prospective students of my supervisor and am always happy to answer their questions.

How much you factor the information you gain into your decision is really up to you, and depends a lot on your own motivations for doing a PhD and what you want to get out of it. For example, if you want to go into industry afterwards but a supervisor tells you that all of her former students are still in academia, would that influence you away from studying under her? Or if the majority of a supervisor's students are now in industry, how do you know whether they had to leave academia because they couldn't get a postdoc or because they really wanted an industry job instead?

Answer 2

If you are entering a doctoral program in the US direct from a bachelors, this is easier than you think. Your actual doctoral research is still a few years away since the early part of the program will be coursework leading to comprehensive (qualifying) exams. You will have a chance to meet and interact with a number of professors before you need to choose your actual dissertation advisor, though you may have a less formal academic advisor along the way.

So, you will have several chances to sit down in the office of some potential advisor and talk about both research options and where you might wind up if you work hard. They will probably be more willing to discuss other students with you if they can do it anonymously. And they can, themselves, ask former students to contact you so you can explore further.

But, because of the time lag, you will also have the option to discuss various professors with their students, to see how they are getting along. That can be equally important to know before you sign up with someone. Many questions on this site deal with poor advisor relationships.

But this assumes that you are applying to a place with a fairly large faculty and plenty of options. It is a bit harder if you think you need the information just to choose between institutions.

I suspect that most institutions would be willing to provide you with a list of recent graduates and their dissertation titles, though perhaps not with contact information. Privacy and all. But that information and google can get you more information. Many professors list their completed students along with dissertation titles on their own web sites, as I do.

And note that this applies to the US and probably not so much elsewhere.