Ask us anything: finding ecology PhD positions, and improving the graduate admissions process

Every year we invite readers to ask us anything! Today’s question (paraphrased and summarized, click through for the original) is a two-parter from Nicole:

Are there any places besides the TAMU boards and Ecolog to look for funded ecology PhD positions?

The current way ecology graduate admissions seem to work (from a grad student perspective) involves either connections your undergraduate adviser has to other labs, or cold-emailing faculty doing research you’re interested in, developing a rapport with the person over email/phone/skype, and having them advocate for your admission. This system seems opaque and bias-prone. What changes do you think could be made to make this process better, both systematically/from an institutional perspective, and individually (on a lab-by-lab basis)?

Jeremy’s answer:

Afraid I don’t know of any centralized location to look for ecology PhD positions in N. America besides Ecolog and the TAMU boards. A few Canadian ones may be on the CSEE website. I know nothing of where to look for ecology PhD positions outside N. America.

Re: opacity and potential for bias in the current ecology grad school admissions process, I have a lot of thoughts and they’re not well-organized. Sorry.

Without wanting to claim that current graduate admissions practices are perfect, it’s worth keeping in mind that there are good reasons why current admissions practices work as they do. As an ecology PhD student, you’re going to be working with your supervisor for 4-5 years or so (in Canada) or 5-7 years or so (in the US). It’s a big commitment for both of you, and it would be bad for both of you if it didn’t work out for any reason (because the two of you didn’t get along, because your interests and goals didn’t align, whatever…). There has to be a mutual fit between student and supervisor. So I do think there needs to be some way for supervisors and prospective grad students to get to know and trust one another sufficiently well, before they make a mutual commitment to one another. For instance, here in my department at Calgary, there’s actually no competition for graduate admissions, at least at present. We don’t get more qualified applicants than we can admit to our graduate program, and so we just evaluate applications on a rolling basis as they come in. So you don’t need your prospective supervisor to advocate for you to be admitted over some other applicant in my department. But even here in my department, we still require you to state in your application who’s going to supervise you. Because admitting someone without a supervisor lined up would just be setting them up to fail.

It’s also not clear to me that the opacity and potential for bias in the current N. American graduate admissions process is due primarily to the expectation that prospective students will contact prospective supervisors in advance of applying. For instance, undergraduate admissions to selective US colleges and universities don’t require applicants to get to know individual faculty. But yet, the undergraduate admissions process to selective US colleges and universities is infamously opaque and bias-prone. Heck, I’d probably argue that graduate admissions in ecology are less opaque and bias-prone than undergraduate admissions at many selective US colleges and universities, precisely because prospective graduate students and supervisors get to know one another. For instance, I know of several cases in my own and others’ labs in which supervisors who got to know prospective students before they applied were able to make the case that those students should be admitted, despite elements of those students’ backgrounds/circumstances that might have seemed disqualifying to someone who didn’t know them. And I know of several cases in which prospective students sought out potential graduate supervisors who’d support them and their specific needs (e.g., mental health needs). The process of supervisors getting to know prospective students isn’t just a way that bias can enter the admissions process–it can also be a way the admissions process can overcome bias.

None of which is to deny that the ecology graduate admissions process can be opaque! One answer to the question, “how do we make the graduate admissions process less opaque?” is “publicize how it works”. Obviously, this is something that individual faculty and departments can do, and that many are already doing, but I’m sure many places could do more on this front. For instance by holding workshops for their undergraduate students about grad school admissions, opportunities to get research experience as undergrads, etc. If your institution has a graduate program, encouraging the TAs to talk to their undergrads about grad school is another, informal way to get the word out about what grad school is like and how to go about pursuing it. Blogs and social media also have a role to play here. If there’s good free information and advice available online where it can be easily googled, hopefully that helps make the process a bit less opaque. I’m thinking for instance of this old post from Jacquelyn Gill and this old one from Meghan. And when I google “how to apply to graduate school ecology” (without the quotes), the first hit I get is this old ESA Bulletin piece from Walt Carson and the second is this piece from Jennifer Miller. The point here is not to criticize prospective grad students who don’t google for information. It’s just to suggest that it probably helps to use various avenues to publicize how grad school admissions work, in order to reach the widest swath of prospective grad students.

In terms of making ecology graduate admissions not just less opaque, but less biased, here is my thinking (which I worry a little is going to get me slammed in some quarters, but that’s an occupational hazard of blogging…). There’s a trend to get rid of the GRE in graduate admissions, on the sensible grounds that it’s not a good predictor of success in graduate school (and some would cite other reasons for getting rid of it). I know some people would like to do away with reference letters on the grounds that they can be biased. As you’ve noted, “getting to know prospective supervisors” could be a bias-prone process too, perhaps selecting for students from middle-class backgrounds who are comfortable corresponding with white professors. Arguably we shouldn’t put too much weight on undergraduate grades because (i) ecology grad school isn’t mostly about coursework, (ii) GPA doesn’t predict many measures of success in grad school (at least among admitted students), (iii) it can be hard to compare grades from different institutions, and (iv) many (not all) prospective grad students have similar, high grades. Research experience is a key factor in ecology graduate student admissions, but I’ve seen arguments that it should be de-emphasized or even ignored because not all undergraduates have equal opportunity to do independent research or serve as summer field assistants. Rather than expecting students to enter grad school with research experience, arguably we should think of grad school as the place where they learn to do research–for the first time, if necessary. And basing graduate admissions on the reputations of the institutions from which the applicants received their bachelor’s degrees has obvious problems. Individually, each of these arguments has some merit (I think some have more merit than others, but they all have some merit.) But here’s the thing: if you grant all these arguments, you’ve run out of sources of information on which to base graduate admissions. And I’m sure nobody thinks that graduate admissions should be determined by a lottery from among whoever applies! (although OTOH some people do think that undergraduate admissions to selective institutions should be determined by lottery, so perhaps somebody thinks that for graduate admissions too…) Arguing “[thing] can be biased, so [thing] shouldn’t be used to inform graduate admissions decisions” is a bad argument, because it’s true for every possible value of [thing]. The right question to ask is “what bits of possibly biased information should be considered, and how, in order to for graduate admissions to be as fair as possible?”

My own feeling, based on my experience with graduate admissions (and on faculty search committees) is that it’s best for admissions committees (and search committees) to look at more sources of information rather than fewer, including relevant contextual information. Committees should then evaluate those sources of information holistically, alert to the potential biases of each source of information. My own experience is that graduate admissions committees (and search committees) absolutely can do this, and that many already do. I also think that an information-rich, holistic approach to evaluating applications should be supplemented by focused efforts on the part of graduate programs to recruit, develop, and mentor graduate students from historically underrepresented groups. Terry McGlynn has some suggestions for what effective recruitment and development efforts look like. And see this recent news piece in Science on the Myerhoff Scholars Program.

There are of course other ways to match graduate students and supervisors besides expecting prospective grad students to contact potential supervisors in advance of applying. The matching can even be done post-admission. As you probably know, many graduate programs in cellular and molecular biology in N. America begin with lab rotations. New graduate students spend their first year or so doing short stints in their choice of 2-4 different labs, at the end of which they pick which of those labs they want to do a PhD in. During their time in each lab, students typically do a small pilot project, or pitch in on someone else’s project. There are a few ecology graduate programs that do rotations, or used to. Duke is one, or used to be. I applied to Duke on a whim and didn’t contact any profs there–and then was very surprised when I was admitted! They told me that if I came I would spend my first year rotating between a few labs and then pick one. I didn’t go to Duke, and so I have no idea how well the rotation system works. I admit that, from my point of view as an ignorant outsider, lab rotations have always seemed like a bit of an awkward fit for ecology. My possibly-stereotypical, overgeneralized impression is that, in cellular and molecular biology, there’s less expectation that graduate students will exercise intellectual independence (developing their own projects, etc.). And many basic lab techniques are common to many cell and molecular labs, so rotating students can get some basic technical training in any lab. And from the perspective of some supervisors, graduate students in cell and molecular labs are fungible “cogs in the machine”, like employees in a corporation. Having a good match between the interests of the student and supervisor isn’t all that important in some cell and molecular labs, or at least isn’t seen as all that important by some supervisors. So lab rotations are a perfectly good way to match up students and supervisors in many cellular and molecular biology graduate programs. But on the other hand, clearly rotations can work in EEB too, because otherwise the EEB graduate programs that have them presumably would’ve gotten rid of them! Hopefully we’ll get some comments from people who’ve gone through lab rotations in EEB.

If you wanted to go in a really radical direction, you could argue for completely revolutionizing ecology graduate school so that graduate students no longer have  supervisors, or even co-supervisors. If graduate students no longer have one or two supervisors with primary responsibility for their training and mentoring, and with whom they work particularly closely on a day-to-day basis over an extended period of time, then there’d no longer be any need for prospective graduate students to line up those (non-existent) supervisors. There’d no longer be any opaque, potentially-biased process of seeking out prospective supervisors and getting to know them before you apply. And there’d be no possibility that those (non-existent) supervisors would ever exercise power over their students’ careers or lives in inappropriate ways. That’s how course-based master’s programs work, and it’s how med school, law school, and business school work. Personally, I instinctively dislike this radical proposal–I’m an evolutionary, not a revolutionary. I find it hard to imagine how most graduate students could still pursue substantial independent research projects without a primary supervisor or a couple of co-supervisors with whom they work particularly closely over an extended period. And “make ecology graduate programs work like professional schools” sounds to me like a recipe for making grad program reputation loom very large in determining the future academic employment prospects of ecology PhD holders. Just as law school reputation looms large in determining the future employment prospects of law school grads. But not everyone agrees with me. Here’s one proposal to reduce the importance of supervisors to ecology grad students, although as I understand it it doesn’t go as far as the more radical proposal I just sketched.

Even more radical still is the argument that grad school–all of it–is rotten to the core with systemic bias and abuse and should be abolished. But if you’re reading this blog I’m guessing you’d regard that argument as throwing the baby out with the bathwater.

I throw these two radical ideas out there not because I endorse them–I don’t–but because it’s useful to think outside the box sometimes.

Brian’s answer:

I don’t have any suggestions for PhD position listings either (except a lot are listed on the paper-based job and message boards at Ecological Society of America meetings but I don’t know how helpful that is to the average undergrad).

It really does all come down to contacting individual advisers.

I think the biggest thing you could do to improve transparency is to just be much more transparent about the fact it all comes to contacting individual advisers. If you read a lot of websites for departments and graduate colleges it will give an impression it is mostly about applying to the department or college. But its not. Every year we get excellent candidates who applied to my biology department but did not contact a professor. Occasionally a professor looking for a student will find one of these applications, but many languish.  When I was returning to grad school after working in business I was quite clueless and didn’t know I needed to make individual connections (so I speak from experience when I say websites are not transparent). I’m kind of amazed I got into grad school looking back. I think we can be much, much more transparent about how things really work.

You mention the problem of clubbiness and undergrad mentors doing the connector of undergrads with graduate advisers. I don’t know how common that is. I don’t think I’ve ever been approached to advise somebody by anybody other than the prospective student. Which is how it should work. (Jeremy adds: same for me.)

I’m not sure that you really can or should fundamentally change the admissions process in terms of connecting with advisers (unless you are prepared to do as Jeremy mentions and blow up the whole notion of advisers). You need to get an adviser committed to your success and we basically have a matching system that does that.  On a certain level the current systems increases the quality of matches – if their biases show before somebody is admitted to work with them that is a bigger problem avoided (not to say efforts shouldn’t be made to eliminate such people on a faculty, but that is another topic). And again without getting rid of the notion of advisers, what are you going to do? Admit people to the department and then force faculty to advise students? I just don’t see that as viable. Very little of grad school is general learning from the department, the vast majority is from an advisor (and maybe one or two other close faculty-student relations). So I think we’re stuck with needing to match up students and advisers to one-on-one relationships that both sides are committed to. And I don’t know how you improve the current system except to be much more transparent that that is the current system.

The best predictors of grad school success are motivation to be there and the ability to be persistent and work hard and be resilient (and probably to write well). I’m not sure how exactly you gauge that. So while acknowledging GREs have severe limitations is good, we shouldn’t kid ourselves that there is an immediate superior alternative. Most PIs I know find it incredibly hard to predict a priori who will be successful or not.

I think probably the other really big factor is who gets the money. In the US PhD RAships are primarily funded under the professor’s control through NSF research grants. In other countries there is much more money invested in giving graduate fellowships to the graduate students. There is no real reason NSF couldn’t do this by pulling a good chunk of money out of grants and putting it into the doctoral (and postdoctoral) fellowships. This totally changes the game when the student brings the money instead of begs to get the money.

So in summary massively more transparent about the need to connect with a prospective adviser and more funding via fellowships direct to students are the two reforms I would favor most.