Monday, February 28, 2011

What am I looking for when hiring a new graduate student


February 28, 2011

For students who are entering graduate school with the intention of earning a PhD, the choice of an advisor is truly a career-determining decision.  Your thesis topic is certainly not the only thing that you will ever work on in your career, but your choice of a research group will determine at least what your scientific focus will be for the next 5 years of your life.  Students seek out faculty to serve as their advisors by any number of criteria: Are they famous?  Are they well-funded?  Are they working in the technical area that is likely to result in future faculty positions?  Is their research group spinning out companies with a lot of potential?  Are they the same nationality?  Are they at a prestigious university?  Do students seem to be graduating in a reasonable amount of time?

While you might be considering all these questions about potential advisors, let’s think about how things look from the opposite point of view: What are professors looking for when considering which new students to bring into their research group.

For a faculty member, this is an extremely serious question.  Excellent students will be able to work independently, collaborate on development of creative ideas, work hard, write lots of papers, and give presentations that reflect upon the whole research group.  Poor students will not make much progress in the lab, will need precise directions for every small task, will never think for themselves, never conclude an experiment, write poorly organized papers with no attention to grammar or spelling, and mumble incoherently when up in front of an audience at a conference.  Hardly any students start off graduate school in the “excellent” category in all respects.  While part of the process of learning in graduate school is to develop, over time, the skills and approaches that are more like the “excellent” student than the “poor” one, a professor is looking at a potential new graduate student and trying to decide if they have the makings of future greatness.  The ability to recognize great students has consequences for the professor: a group loaded with poor students is going to accomplish very little, resulting in lower numbers of high quality publications, little research getting done in the lab, and bad prospects for tenure and promotion.  Certainly, a professor can try to micromanage students, write every word of every paper, and take total responsibility for every tiny aspect of the group – but with a big cost in time, effort, and potential accomplishments.   It is far better, and less frustrating, to hire great students, to help them to develop their full potential, and to work with them as colleagues.

The relationship between a professor and their graduate student is kind of like the Jedi/Apprentice relationship in Star Wars.  The two people will be spending a lot of time together for several years, and most faculty members can only effectively train a handful of students at any one time.  A professor may also be responsible for providing funding support for a student for several years – representing a big commitment in successful proposal-writing.  Further, the student will be the person working in the lab to achieve the proposal’s goals, and thus will determine to some extent whether the “customer” (funding source) will be happy or not – which in turn has an impact on the professor’s ability to obtain future funding.

So selection of a student by a faculty member means a lot.  I do not know how typical my research group is, but in an average year, I may hire between 1-3 new students.  I am comfortable serving as the advisor for no more than ~12 students at any one time, so at this point my group has reached a kind of steady state.  Mostly, I look to replace students who are graduating (and try to arrange some overlap between the new ones and the graduating ones), or to staff a new research grant that has been recently awarded.  Certainly there are research “mega-groups” with a famous faculty member who oversees 30 postdocs and 50+ graduate students, but in groups like that it is most likely the postdocs rather than the professor who you will see most often.  In such groups, I have heard that the professor “does not even know your name” until you have prepared a first-author journal manuscript for a major journal.

With no more than 3 positions available per year, I receive literally hundreds of requests per year from students who are interested in joining my group.  How do I choose?

As I mentioned, hiring a new student is a risk.  Like most people, I would like to minimize my risk to some extent.  Therefore, the students most likely to be hired as students are those who I already know.  Fortunately, the University of Illinois is one of the top engineering schools in the US and in the world, and the population of undergraduate students is extremely good.  I teach undergraduate courses every year, and get to observe directly the classroom performance of nearly all seniors and juniors who are interested in MEMS and biosensors.  I can tell which students are taking the time to prepare for exams, doing a meticulous job on their homework assignments, asking good questions, and thinking about interesting topics for their final projects.  When those students apply to the University of Illinois for graduate school and indicate that they are interested in my area of research, those are among the students that I will consider first.  I also host several undergraduate students in my research group who are working on research projects for course credit.  I meet with them individually several times during their internship and can see what progress they are making.  The undergraduate students work with graduate students in my group who serve as their mentors.  I ask my graduate students about who is a hard worker, a fast learner, a careful experimentalist, and a good collaborative personality.  If a great undergraduate student who is working in either my lab or another faculty member’s lab shows interest in graduate school and joining my group, I am sure to talk with them, and they are at the very top of my list.

So it seems like undergraduate students at the University of Illinois have the best shot at joining my group.  This is true, but they are not the only students who I consider as top candidates.  When students from other universities apply to graduate school at Illinois, they can indicate on their application which research groups they are interested in.  When it is time to consider graduate school applications, I look at all the applicants who indicated interest in my area, but only if their GPA is greater than 3.5/4.0.  Grades are not everything in life, but they do say something about a student (see a future blog entry: “What do your grades say about you?”).  A student with very strong grades at an engineering program that I know to be a rigorous one also goes to the top of my list.  I look most closely at the applications of students with GPA>3.8/4.0 because these grades represent a long-term dedication to excellent performance and some ability to master a wide variety of course material.  Students from top-10 engineering schools with excellent grades also stand a strong chance of getting support from a Fellowship, which I can facilitate in some cases by writing recommendation letters or by working together to define a research plan.  Grades are not everything.  I also carefully read students’ essays to see if they can communicate effectively, to understand their motivation for going to graduate school, and to see if they articulate a research interest that matches mine.  Finally, I also carefully read a student’s recommendation letters.  I am looking for evidence of intellectual curiosity, problem solving ability, good interpersonal skills, and willingness to take on challenging problems.  If I happen to know the faculty member who wrote a recommendation letter, it carries extra weight with me, and I might follow up by writing a note asking about the student.  If all this is good, THEN I will write an email to the student to arrange a phone interview.

Basically, I will not hire someone without having an opportunity to meet them - at least by phone.  How a student handles the interview can either make me excited about hiring them, or totally kill my enthusiasm.   Some students do not seem to understand that interviewing a faculty member for a Research Assistant position is exactly like interviewing for a job, and fail to take it seriously.  Before speaking with a professor, you should know something about their research (at least from their web page) and you should have at least browsed through a couple of recent papers.  Being able to ask a couple of good questions shows that you have at least thought about what the research group does.  A student who can articulate a specific desire to work in some area is much more impressive than one who just says that they want to work on “something” or “anything having to do with nanotechnology.”  I want to hire someone who is committed to making an impact on the world with their research.  A student who wants to “cure cancer” is much better than one who wants to “think about it for a while and then see what clicks.” I have heard both things.

I have hired several students who worked for another advisor (at Illinois or elsewhere) and left their former group for some reason.  For example, if a faculty member moves to another school, or if a faculty member is notorious for poor relationships with their students (no names mentioned here), their semi-trained students often work out extremely well in a new environment with a new research topic.  These also are students who I can know a lot about from having them in my class, or by asking other professors what their opinions are.  I will also meet with them, sometimes several times, to see if there is a good fit.  If a student has been let go by their previous advisor for poor performance, however, then I am very unlikely to consider them unless the circumstances were exceptional.

Students from foreign countries have a more difficult path to joining my research group, however, I have served as the advisor for many very excellent foreign students.  Some of the top students from foreign countries are awarded fellowships that enable them to train in the US.  If these students have top grades from a great school AND financial support for at least part of their graduate education, it significantly lowers the risk to hiring them.  If they get here and turn out not to be well-suited to a PhD program, occasionally they may leave Illinois with a M.S. degree.  If the student gets here and turns out to be great, I am happy to support them for their Ph.D. work.  So in essence, their financial support helped move them closer to the top of the list – even though I still carefully read their application materials and talk with them before recommending admission.

For students who are not in any of these categories, this “system” may seem unfair.  The system is not totally different from the way that college basketball teams all try to get the best players from the top schools.  College basketball recruiters look at the numbers (like points per game, wins/losses), the quality of the basketball program at the previous school, the strength of the training provided by other coaches who they know, the opinions of other people, and their own instincts about who will develop greatness.  The top college basketball teams are able to pick from the top players, and there is a definite stratification that is used to differentiate the top from the rest.  The top research universities are able to likewise draw from the best undergraduate programs worldwide, and the competition is very tough.