— 05 July 2020 —
I wanted to write this post for some time now since the longer I wait, the more my viewpoint diverges from the one I had when I was a prospective PhD student. At the moment, my memory of looking for PhD programs is still pretty fresh, but I have also seen the process from the other side, getting to know some other students and whether their decisions played out well or not. The goal of this post is to propose some advice regarding picking a university for your PhD. Note that this is all very subjective (and non complete), meaning that you should take it into account together with information and advice from other sources. Also note that I am doing a PhD in the field of Computer Science, so the advice might not be as relevant to students in other fields.
I often see that prospective PhD students (myself included) heavily weigh their options based on the research interests of faculty members. Event though this is important to factor in your decision, I believe that in the end of the day, it doesn’t correlate that much with satisfaction from the PhD program for two reasons. First of all, it is extremely difficult to correctly assess what are someone’s research interests before you work with them, especially considering that faculty members often don’t maintain their websites. Second, I think that the specifics of your PhD research are often left abstract until the very last moment—even senior PhDs often don’t have a very precise plan of their research agenda.
In contrast, what matters a lot for many people is the daily interaction with their advisors and group. Being content during the everyday struggles of a PhD depends a little on whether you are optimizing a compiler pass or a distributed system, and more on whether you appreciate and enjoy working with your collaborators. If your advisor wants to micromanage you but you prefer a more independent relationship, you might suffer even if you work on your favorite problem. Of course this is mostly based on my own experience, and it might be different for you.
Before moving forward I would like to note that my opinion is biased since I have been fortunate to be surrounded by amazing collaborators, a lot of which are now my dear friends. This includes my advisor, who is extremely supportive, helpful, and flexible, the other faculty members in the group, who are also very helpful and supportive, and finally the other students and researchers, with all of whom I have an amazing relationship.
Based on the above assertion, i.e. that a PhD student’s relationship with their advisor and collaborators is more important than their exact research topic, I would like to give some advice on how prospective PhD students can pick a group to join. Even though the prospective relationship with an advisor is very hard to predict and difficult to assess from brief visits or chats, there are several qualities that an advisor can have and don’t work for most people. Based on this, I think that there are a couple things that one should do to improve the chance that they end up with an advisor that they like working with.
In my opinion the most impactful action by far is asking tons of questions to everyone: present and past PhD students, postdocs, and faculty members. Especially with students and postdocs, it helps if the questions are not abstract, but concrete things that some people might even feel a little uncomfortable answering. Of course the goal is not to make people feel uncomfortable, but rather to get as-honest-as-possible opinions on important matters. The good thing is that after you are accepted in the program, there is nothing that can go wrong, so you have nothing to worry about when asking these questions.
Asking questions could help figure out if there are students that have any issues with the faculty member that you are interested in working with. Possible issues could be (but are certainly not limited to):
Even though it is difficult for people to be completely honest with strangers, there is a lot that one can get out of their responses. In my experience, people tend to positively boost their responses to these types of questions, e.g. if a PhD student is unhappy with their advisor, they won’t express their discontent, but they might just say OK things about them. Spotting a lack of enthusiasm or positivity and poking it with more questions could lead to some important information about possible issues. Furthermore, if more than one person has similar opinions about a faculty member, this strongly indicates that the opinion might indeed be valid.
Since the goal is to figure out the bad aspects of the working environment, it is beneficial talking with past PhD students too, since they might feel more comfortable opening up about issues; in contrast to current students whose future depends a lot on their advisor’s future too.
Here are some concrete questions that one could ask (current and past) PhD students and research assistants:
Finally, it is important to consider universities that have a diverse group of faculty that you would to work with, as well as allow (and ideally encourage) collaboration and experimentation with different faculty members at the beginning of your PhD. As far as I know this is mostly applicable in the US, where prospective students are accepted by the department and are not necessarily tied to a specific professor from the beginning. The reason why this is important is obvious; in the end of the day, it is possible to initially misjudge a person, and only figure out that you don’t totally align with them after the fact. Being in a university where working with different advisors or changing advisor is easy, gives you an opportunity to continue with your PhD without leaving the program. Even better, the ability to work with several people in your first couple years allows you to get more data and then make an educated decision of you want to work with.
To summarize, I believe that when people decide which PhD program to join, they often think more about their research interests, and less about the future relationship with their advisor, assuming that it will be fine by default. However in my experience, PhD student dissatisfaction stems very often from issues related to a student’s interaction with their advisor. An unhealthy student-advisor relationship can not be easily solved, often forcing the student to leave the program. On the other hand, if the relationship is good, a healthy solution can be found to most of the other issues that might arise during the degree.
EDIT 2023-01-25: Stephen Chong gave an amazing talk at PLMW@POPL23 about selecting and working with one’s advisor, which I would highly recommend to any student that is searching for an advisor or already has one!
Hosted on GitHub Pages — Theme by orderedlist