What I read on my winter vacation

It’s been a while since I posted anything serious — I can’t help but feel that I should apologize, although I’m sure that very few of you sit by your computers waiting breathlessly for the next update. Anyway, my excuse, if I need one, is that much of my spare time has been taken up with graduate program applications. And I thought some of you, especially potential future graduate school applicants, might be interested in a few thoughts I have about what goes into an application, and how it’s read (or at least, how I read it).

There are  four main pieces of information in an application for an American graduate program in biological sciences: GREs, college grades (supplemented by an academic transcript), the personal statement and the letters of recommendation. From my perspective the GREs and grades essentially act as thresholds. If your GREs and grades are above a certain level, I don’t much care how good they are. Academic achievement has some correlation with likely success as an independent scientist, but the correlation is not a straightforward or linear one. So, if your GREs are below a certain level (and you’re a native English speaker), it may raise a suspicion in me that your brain is not especially agile, and I may start looking for other evidence to support this hypothesis in the rest of the package; if your grades are poor, especially in subjects that are very relevant to your proposed area of study (or, especially at a school with rampant grade inflation), I may worry about your ability to assimilate facts and reasoning strategies; but I’m probably not going to draw much of a distinction between very good grades and perfect grades. And if one of your letter-writers explains that your very-good-but-less-than-perfect grades are due to spending too much time in the lab, I may even like you better than I would have if you had a 4.0 GPA.

After the GRE/GPA threshold is passed, I have two pieces of information left: your statement and the letters written on your behalf by people who know something about your abilities. I don’t have much to say about the statement — it should be personal, after all — except this: don’t feel that you have to treat it as an exercise in creative writing. I suspect that there’s a book out there, or more likely a website, that encourages applicants to begin their statements with an eloquently written personal anecdote that conveys their love of and excitement about science. I suspect this because so many of the statements I read start with something along these lines: “Staring into the dusky red depths of the Grand Canyon, I had a vision of someday creating cybernetic wings that would allow human-powered flight. My passionate dedication to a life in science dates from that day.”

If you really did have some kind of emotional revelation when you were 10 years old, fine. I don’t mind if you tell me about it, though please try not to be too dramatic. (I’m English, after all.) But from the perspective of a specialized interdisciplinary program, I’m much more interested in your mature thoughts about why you want to focus on systems biology in particular. It’s not an easy path, and I want to know you’ve thought about it. It’s fine to try and paint a picture of who you are as a person, but the more you can explain why you’re interested in working in this field in particular, with concrete reasons from lab or theory work you have done or courses you have taken, the better.  What you say will give me some sense of how much you know about what you’re getting into, as well as whether you can write a clear English sentence and put two ideas together in a logical order.

What about the letters? IMHO these are the most important part of the package, and what’s hard about that is that the quality of the letter is almost completely out of the applicant’s control. What I look for in letters is the following: evidence that the applicant has taken the initiative to seek out interesting research experiences, and an opinion from the person writing the letter about whether the applicant has the qualities needed to make a go of independent research. Unfortunately, what I often get is a letter that says that the applicant got very good grades in class, which I already know from the transcript. Letters of this kind aren’t helpful; nor are the letters from your swimming coach or saxophone teacher, or the personal friend who’s known you since childhood. These people may all have theoretically useful things to say about your strength of character or your work ethic, but they’re not scientists and I don’t know what “hard-working” means to them. The ideal application package has at least two letters from people who know what it takes to be a good scientist, who can talk about a substantial project that you did in their group. These letters ideally go into significant detail about how well you performed in the context of a real research environment — for most of the faculty on the committee, nothing speaks louder than a statement from an advisor about the research you personally did while working with him or her.

Most heads of research labs at major American universities know just how to write a useful letter. They’ll lead off with a summary sentence on how good you are, then talk about how quickly you assimilated ideas, how independently you worked, whether you had ideas of your own, and how productive you were. They’ll provide some kind of comparative ranking: I’ve trained 100 undergraduates and this person is in the top 30/10/3. If you’re working in a European lab, though, your advisor may not know what’s expected of them. European letters are on average 1/2-3/4 of a page long (while American letters are often 2-3 pages long) and simply don’t provide the kind of detail an American reader is looking for. They say “this person is very bright” but they don’t say how they know that. They say “this person was very productive in my lab” but don’t specify what you produced, or provide any distinction between the hard-working person who needs to have their hand held every step of the way, and the moving-towards-independence person who started to have their own ideas about how the research should be done. It’s frustrating, because we have too many excellent students applying to the program (a nice problem to have, of course) and we can’t interview them all, let alone accept them all. The student with a letter that gives a detailed description of how, during a 6-week summer internship, he or she took the initiative to build their own super-resolution microscope and successfully used it to ask an important question (yes, don’t worry, I’m making this up) is much easier to put at the head of the interview list than the student who has a letter saying “this person did an excellent job on an imaging project “.  It’s not the superlatives that matter, it’s the specifics.  If your letter-writer wants to tell us that you have a certain attribute (intelligence, creativity, etc.), the most helpful and convincing way to convey this is to give us concrete examples of how this attribute manifested itself during your time in the lab.  It’s particularly helpful if they can honestly compare you with other people who’ve come through their lab: e.g. as good as/better than X (who is now on the faculty of a major university) at the equivalent stage. “One of the best I’ve ever seen” is something one might say about lots of students; this statement, though it superficially sounds very positive, is much less helpful than “the 7th best student I’ve ever had”.  If the letter goes on to give the names of the other students on this person’s top 10 list and where they are now, then you know that the writer is really trying hard to give you useful advice.

So, future applicants, if you’re applying for an American program and your main research project took place outside what America considers the mainstream — which I’m afraid means most European or Asian labs, as well as labs in industry — it may be a good idea to have a conversation with your advisors well before the letter deadline to ask them to consider writing an American-style letter. If they’re willing to give it a try, you can probably help them by reminding them of exactly what you accomplished, and especially of any useful ideas you contributed to the research, even if these are fairly minor.  You may also want to suggest that they spend a few sentences introducing themselves and their University or Department, unless that seems really stupid; we don’t know everyone, and it can be really helpful to know that the selection for the students in a particular Department is extraordinarily competitive, for example, or that there’s very little grade inflation in a particular University, or that the person writing the letter was part of the faculty at a major American university for two decades and can confidently compare the applicant to graduate students at that university.  Comparisons to known quantities are useful.

American advisors have their own problems, of course. A few faculty members seem to feel that it’s their job to get their students into any program they apply for, whether the program’s a good fit for the student or not. Letters from these advisors are always long, detailed, full of strategic underlining or bold text. All of their students, apparently, are in the top 1-2%. This may sound like a good trait in an advisor, but the problem is that admissions committees catch on to this quite fast. If you have an extraordinarily positive letter from someone who invariably writes extraordinarily positive letters, you are not much better off than you would be if you had no letter at all. I sympathize with these advisors; it’s much easier to be uniformly enthusiastic than it is to make agonizingly careful distinctions between the outstanding student and the merely excellent one. But what we’re asking for in the letters is a fair evaluation of a student’s potential for research, and in the end these overblown letters don’t serve the community or the student well. We ask each other in committee meetings “are you calibrated on so-and-so?”, meaning, can you interpret this letter in the light of other letters this person has written? If this particular lab head says “best student in 5 years”, do they mean it, or do they use those words in describing at least two students annually?  Some people really do this.  They probably justify it to themselves by believing that the students in their university are much better than the average student.  Possibly true, but the point is that I can’t trust their assessments to be honest and so I find myself doubting everything they say. It’s too bad.

Like any other judgment one has to make between individuals, these are hard choices and we never know whether the answers we come to are the right ones. We do our best, and hope that if we make a mistake in choosing not to accept a particular student some other lucky program will successfully recruit them instead. In the end, perhaps, it all works out.