Reflections on the NSF-GRF

A few days ago the NSF-GRFP (National Science Foundation Graduate Research Fellowship Program, which awards a 3-year fellowship that includes a fairly luxurious living stipend, a tuition allowance, XSEDE time, a travel allowance, and access to other resources, and is valued around ~150k USD) announced that they would be limiting graduate students to applying only once while in grad school, either during their first or second years. Specifically, their letter states:

NSF 16-050

Dear Colleague Letter: Change in Eligibility to the NSF Graduate Research Fellowship Program (GRFP)

March 7, 2016

Dear Colleague:

With this Dear Colleague Letter, NSF is providing advance notice of a change in the eligibility of prospective applicants to the Graduate Research Fellowship Program (GRFP) to allow the community to plan accordingly. Effective as of the 2017 competition (Fall 2016 deadlines), NSF will limit graduate students to only one application to the GRFP, submitted either in the first year or in the second year of graduate school.1 No change is made to the eligibility of undergraduates, of bachelor’s degree holders without any graduate study, or of individuals who have had an interruption in graduate study of at least two consecutive years.2 GRFP continues to identify and to inspire the diverse scientists and engineers of the future, and especially encourages women, members of underrepresented minority groups, persons with disabilities, and veterans to apply. This change in eligibility should result in more individuals applying as undergraduate students who have not yet made the commitment to go to graduate school. This is a more diverse population than admitted graduate students.

NSF’s strong commitment to supporting graduate students through this program since 1952 has resulted in awarding close to 57,000 graduate fellowships to U.S. individuals pursuing graduate degrees in science, technology, engineering, and mathematics (STEM) and in STEM education. Through its Graduate Research Fellowship awards, NSF hopes to encourage outstanding individuals from across all groups to attend graduate school.

Please see NSF 16-051, Frequently Asked Questions (FAQs), for more information.

Joan Ferrini-Mundy
Assistant Director
Education and Human Resources Directorate

1 First-year graduate students in Fall 2015 who applied to the 2016 GRFP competition will be allowed to apply a second time in Fall 2016, if they are otherwise eligible. All other graduate students are subject to the new eligibility requirements.


A lot of discussion followed this announcement, and I’ll try to summarize my thoughts below.

First off, the ostensible purpose of this change is to increase diversity in the application pool by increasing the number of undergraduate applicants. I’m not so sure this will work. I suppose they could be referring only to the relative number or proportion of undergrad applicants, which would go up if fewer grad students are allowed to apply? Thus allowing the NSF to potentially fund more undergrads and devote more reviewer-time to each application. And maybe senior undergrads, seeing that they’ll only have one chance to apply after they graduate, will be more inclined to apply knowing they won’t have as many opportunities to in the future (and maybe undergrad PIs will stronger stress the benefits of applying). And since undergrads represent an inherently more diverse applicant pool (because they’ve not yet been filtered by the grad school admission process, which tends to homogenize, I’m guessing), this will increase diversity in who receives the award.

If that’s the hopeful mechanism, I guess this change being effectively free and effortless on the side of the NSF would fall in its support. Though I suspect it could easily have the opposite effect. Why? Grad school admissions might filter out diversity in itself, but they very well might improve diversity at the level of being aware of and applying for the GRF. If primarily prestigious, fancy, well-funded undergraduate programs and labs are pushing students to apply, having (relatively or absolutely) more undergrads will actually reduce diversity in the applicant pool, whereas if all grad programs push students to apply, having more grad students will increase diversity (provided admissions isn’t too great a filter). For this change in graduate eligibility to have the desired effect, targeted outreach to raise awareness in underrepresented undergraduate populations would be needed.

Second, and mostly out of sympathy for incoming grad students who were expecting to be able to apply twice, this seems kinda rough! The oof of getting rejected my first time applying was certainly dulled by the knowledge that I’d have a second chance and could respond to reviewer suggestions. Giving first year grad students a choice on whether to apply now or continue polishing their applications to apply next year will probably lead to a fair bit of stress! And imagine getting rejected after having applied in your first year and playing what-could-have-been if you’d just held off! Ouch! Some might argue (as in, some have argued) that stress is good for grad students, and those that can’t handle the heat should step out of the fire! It’s a dog-eat-dog, survival-of-the-fittest world out there, and if you’re weeded out by anxiety then you deserve to be! But I don’t agree. I only care about making people anxious insofar as it helps them improve (in the sense of eustress), and not as a hazing or weeding mechanism. This is because I don’t think a more stressful application is targeting the trait of interest, i.e. ability and potential to produce quality scientific research. People can be great scientists and do excellent work while still being susceptible to anxiety or other mental health disorders, and a needlessly stressful atmosphere doesn’t seem conducive for maximizing research excellence. In any case, students have the entirety of grad school and beyond to grow tougher skin, and a more nurturing environment might help them more in their early years as junior scientists.

Allowing students to apply twice in grad school may help to even the playing field, in that I don’t think the quality of your application — at least, at first — necessarily reflects your quality or potential as a researcher, and getting feedback is helpful in improving the former. Indeed, I think your chances of success depend a lot on how much input your advisor gives you (e.g. if they, as a seasoned researcher, spoon feed you a project, it’ll sound a lot better than one you create yourself) and how familiar you are with what the NSF wants, which strongly hinges on what sort of guidance you receive and your chance exposure to NSF desiderata in years prior. Letting students revise their application and respond to feedback could help them to fix the easy fixes and let their “true” potentials shine.

Third, this change might have the unforeseen (?*) consequence of incentivizing students to enter grad school immediately after undergrad, instead of “testing the waters” and exploring their compatibility with research and academia by joining a lab as a research assistant or getting a Master’s before committing to a PhD program. This is incredibly common advice, especially in light of the PhD surplus relative to a dearth of faculty positions — your really oughta be sure you know what you want and what you’re getting yourself into before jumping in the deep end (or so says conventional wisdom). Moreover, lots of students nearing the end of undergrad aren’t thinking of grad school at all. Rather, they’re entering the workforce (for a variety of reasons, including socioeconomic disadvantage. If they need to support themselves or any dependents, a TAship stipend might be insufficient from the start, and even if they’d manage well enough with NSF money, an acceptance rate in the teens likely represents too great a risk to pin hopes on). Reducing those students’ opportunities to apply for the GRF in grad school could have a penalizing effect on them for not starting a PhD immediately after undergrad.

Fourth, this would also be a good opportunity to raise awareness of other hefty, broadly inclusive incoming/recent grad student fellowships and scholarships, including but not limited to the DOE SCGF/CSGF, FORD, Hertz, NDSEG, NPSC, and SMART, as well as more specific ones depending on students’ interests or backgrounds like the NIH F31, GEM, PDSoros, and NASA GSRP, as well as even more specific ones that are relevant only to very particular projects or backgrounds (not to mention internal fellowships, like the many offered by UC-Davis). The NSF-GRF isn’t the only game in town, though I think it is the most popular. Some of the above are targeted specifically at underrepresented groups (e.g. to apply for the GEM, you need to be American Indian/Native, African American/Black, or Hispanic American/Latino; to be eligible for the PDSoros, you need to be a New American — i.e. the child of immigrants or an immigrant yourself). It’s all well and good to advertise the NSF-GRFP to disadvantaged groups, but how many people (especially those who would benefit the most from them) are aware of the above opportunities (and the many I didn’t list)? My impression is not many.

Fifth, others who’ve thought about this issue a lot more than I have from differtent perspectives have commented on this change in eligibility, too, so take a look at what they have to say.

(In all this I will note my personal bias — I’d never really heard of the GRF until after application season ended my senior year, at least in part because I was still debating between grad school, industry, and government work. Then I somewhat switched fields going from undergrad to grad – from geology/biology to anthropology. My first year of grad school I cobbled together a rather muddled proposal and didn’t get anything that round of application with “fair” reviews. Over the next year, I took some classes in my new field — classes that were relevant to my proposed research — and was awarded the GRF my second time applying with “excellent” reviews and a much clearer project idea and a better understanding of what the NSF wanted. So under the new system, I wouldn’t have gotten the fellowship. Then again, I probably wouldn’t have applied my first year in grad school, so who knows)

(header image ruthlessly adapted from the NSF-GRFP website. Why somebody is using a stirring rod to poke a map of the Americas in a Petri dish, I dunno, but it does look cool)

*So clearly the NSF is aware that undergrads getting the GRF are immediately entering PhD programs. That’s the whole point of letting them apply for it in the first place. I think the hope here is that undergrads who otherwise wouldn’t have gotten into a grad program now have a pretty effective bargaining chip in the form of a phat pile of moola courtesy of the NSF, which overwhelms admissions committees’, PIs’, and society’s potential biases against the disadvantaged with clouds of $$$ and the NSF’s weighty seal of approval.  But how often are those disadvantaged undergrads actually getting the fellowship? It’s been argued not many.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s