Via Joshua Weitz, I see this morning that the NSF has released the call for applications to the next cohort of its Graduate Research Fellowships. This is an improvement in some respects over last year’s GRF competition—released about two months ahead of an October deadline for applications in the Life Sciences, instead of half that advance notice—but it also continues worrying trends at NSF by dramatically cutting the number of GRFs to be awarded, and dropping support for an important field, STEM Education.

It also makes clear NSF intends to continue the change introduced last year, that applicants enrolled in a graduate program only remain eligible through the end of their first year in their first graduate program. That tightening of the eligibility sets up some bad incentives for students, I think, and I fear it’s going to disadvantage those who are enrolled in master’s programs.

Prior to last year’s GRF competition, for something like a decade, enrolled graduate students could apply for a GRF once while graduate students, with at most one full year of coursework completed. That is to say, they could generally apply in the fall semester of their second year of graduate school. This was a good choice for most students, because they could take a year of graduate-level work to develop experience and expertise, and write a more competitive project proposal. Specifically for the kinds of students who start a graduate career in a master’s program, it was more friendly to the degree timeline.

A typical master’s in Biology is supposed to be two (to three) years of work, frequently with a thesis project that’s largely defined before the student arrives. Contrast with a five- (to six-) year doctoral program, in which the student ideally develops an entirely new project: a doctoral student who spends their first fall semester of grad school writing a GRF application is, at the very least, doing the kind of work they’ll eventually need to start their dissertation. A master’s student doing the same must devise a project proposal that may not match the thesis project they enrolled to complete, both for originality reasons and because the scope of a master’s thesis is necessarily less than a doctoral dissertation—and GRF applications are reviewed to the standard of the latter. Where the doctoral student has spent a big chunk of one semester out of an eventual 10 to 12 on some necessary skills practice and project development, the master’s student has as few as four semesters in their full timeline, and is more likely to be doing work on the GRF application that doesn’t advance their degree.

The GRF program has never been particularly well-tailored to master’s students, to be fair—note my comment above that GRF applications are reviewed as proposals for doctoral research, regardless of the program in which an applicant is enrolled. But at least under the old eligibility timeline a master’s student could write a GRF application after a year of thesis project work, at which point they’d be better prepared to propose a project that builds on that thesis. The new timeline forces them to choose between fully focusing on starting up their thesis and devoting a lot of their first semester to a high-stakes, highly competitive fellowship application that may not meaningfully advance their progress to degree. I mentored two master’s students through the latter choice last year when the shorter timeline was announced with about a month’s notice, and I’m proud of the work they both did—but in retrospect I’m not at all sure it was the wisest use of that time, and I don’t know that I’ll advise future students to try it.

But then again, this year’s competition specifies that graduate students are eligible once, in the first year of their first degree program. That means a master’s student who hopes to go on to a doctorate can either apply in that first fall semester, or not at all. A student who wants to try for a GRF, then, has to either accept suboptimal conditions to apply at the start of a master’s program; or go directly to a doctorate. The net effect, I think, will be to further reduce the small—but by no means nonexistent—number of master’s students awarded GRF support, both by setting enrolled master’s students up for less competitive applications, and by nudging more ambitious students directly into doctoral programs with timelines and goals that are better aligned with the GRF.

I say this, admittedly, as faculty in a program that only offers master’s degrees, but I really think the master’s degree has an important role in science education. It offers a lower-stakes onramp into research careers for students who may not feel ready to tackle a doctorate, or indeed who wouldn’t be competitive for doctoral programs, or who want to do work that doesn’t need a doctorate. I’ve seen many graduates of our master’s program go on to doctoral programs, and permanent jobs, that would not have been accessible with the CV they originally brought us. I’ve also seen a lot of them earn GRFs—but the way to that distinction is narrower than it was two years ago. The new GRF eligibility window raises an already-high hurdle for students like ours, and it will ultimately deprive the nation of the future discoveries they could make, if given the chance.