Finding a research advisor

by Michael Ernst (
November, 2002

Undergraduate students who get no research experience are ill-served by college, for they are missing a major benefit of their education. Graduate students who get no research experience simply don't graduate. Thus, all students should be motivated to find an advisor to supervise research.

You needn't be intimidated by the prospect of finding a research advisor. However, you must be proactive; a research position will not fall into your lap. When seeking a research advisor, you should think about two things:

Given a topic, you can translate that into names by surfing the webpages of research advisors.

Now, you should have a list of names of people who might make good supervisors. Read their webpages to ensure that they are appropriate for you to contact. (But remember that webpages are sometimes out of date, especially for busy people.) Send them email, say why you are interested in that particular research group, describe your qualifications, and ask for an appointment. (It is not effective to send a generic message to lots of potential advisors; you will only alienate yourself for being a spammer. Focus your search and contact a smaller set.) Always attach a resume to this message. You won't hear back from some potential advisors; they are busy people. Others already supervise so many students that it would be a disservice to everyone for them to take on more. Others may be willing to met with you.

The interview is a time for you to learn more about potential projects, and for you to describe your interests. (Do not go to the interview in a state of ignorance. Have done your homework by learning about the project and the faculty member. Also consider talking to other students: they will give you the skinny regarding what it is like to work in the group.)

Perhaps you will discover that the research isn't as interesting as you expected. Perhaps you won't have good personal chemistry with the potential advisor. (This is less uncommon than you might expect. Plenty of well-meaning and smart people just think and act in different ways, so they don't have good synergy; it doesn't reflect poorly on either one.) In the best case, you will like the research and the advisor.

You should definitely meet with multiple potential advisors. Even if your first meeting goes swimmingly, and you decide the project would be lots of fun, you should talk to other potential advisors so that you can be sure that they wouldn't be even better. The additional meetings will also reinforce your certainty and prevent "buyer's regret". And you won't look like you don't prepare well when the potential advisor asks you whom else you interviewing with.

Many research positions are advertised through public mailing lists. However, you should not restrict yourself to such positions. Most faculty members can always make space for a truly outstanding student, and many don't know about, or don't bother to advertise on, the jobs list. So feel free to contact every faculty member whose research is of interest to you, not just those who have posted looking for research assistants. But don't spam every professor in a department, and make sure your letter indicates why you are interested in that particular professor's research.

All other things being equal, I encourage cmputer science undergraduates to do undergraduate research with a computer science faculty member. Those tend to have a larger research component (they aren't just using you as computer talent, for instance to maintain a website or write a simple program), they are much more likely to grow into a thesis, and the advisors can give better mentoring and write a much more compelling letter of recommendation.

It takes at least a year to do a quality undergraduate or Master's thesis. Thus, when people approach me in the fall of their graduation year, I tell them that, if they wish to work with me, they will graduate late. (Sometimes people who start earlier but don't make good progress graduate late; but most people do graduate on time.) Don't let this happen to you: nail down a thesis advisor no later than the academic year before you graduate. That will give you time to finish quality research, or if things go terribly badly, it will give you a chance to find another advisor and still have a year to work on your new project.

Back to Advice compiled by Michael Ernst.

Michael Ernst