Applying to the NSF GRFP: The Research Statement
In a previous post, I gave an overview of the NSF GRFP, including some quick advice regarding the Graduate Research Plan Statement. Here, I share some more about my experience writing the research plan, similar to what I did for the personal statement.
Before I begin, a word of caution: I applied for the NSF GRFP after I had graduated from undergrad but before I went to grad school, so while the general advice should be the same, how you select a topic could very well be different if you are applying as a current grad student.
Regardless of your current status when applying for the NSF GRFP, it is important to remember that your research plan is not a contract; you don't have to actually carry out the project you outline in your plan. So if you have yet to join a research group, relax. This research plan is non-binding.
The application components page of the GRFP website states that your research plan should "present an original research topic that you would like to pursue in graduate school" and that you mut "describe the research idea, your general approach, as well as any unique resources that may be needed for accomplishing the research goal." Basically, the NSF wants to assess your problem-solving skills, something that is crucial for successful science. I'll break apart these three pieces, citing examples from my own statement.
Describe a research idea
Applying for fellowships is a lot of work, so pick a research topic that you are passionate about. I'm passionate about astrochemistry and particularly interested in how complex organic molecules (COMs) form in the interstellar medium, so I wrote about that.
I started out with an introduction section that talked about the gaps in our understanding about the origins of interstellar complex organics before going into a research objectives section that gave a list of what I thought to be three key steps toward understanding how such molecules form. (I totally forgot until just now that this is what I wrote about... turns out, it is very similar to what I'm doing now!) To make it clear exactly what gap I wanted to close, I used bold text to emphasize my main objective: "I propose to investigate COM formation pathways by coupling studies of site-specific reactions using stable isotopes with radio astronomical measurements." That way, if my reviewer was exhausted from reading numerous statements, it would be clear the key points they should pay attention to.
I also recommend having a short paragraph about intellectual merit and broader impacts (mine is at the end of my statement, just before the references). These two criteria are essential to your application, so including them in both your personal statement and research plan will show the reviewer you take these criteria seriously. In my research plan, for instance, I include a sentence repeating the importance of my work in the field of astrochemistry which was mentioned in the introduction. I also mention outreach opportunities I anticipate to engage in, naming Project Scientist (a summer academy for elementary and middle school girls) as a specific group I wanted to partner with.
Your general approach
While you don't need to (and shouldn't) provide a super-detailed outline of the methods you intend to apply in your proposed project, you should include enough detail to demonstrate that you have thought about what would be required to address your proposed research question. Space doesn't allow for a whole lot of detail, and painting a big picture of your research ideas is more important than details here.
In my research plan, I mention that I would look at gas-phase reactions observed under interstellar conditions, specifically very low pressures. I give the name of one such molecule I might want to study (aminoacetonitrile) and relate it back to glycine, which I had mentioned in my introduction.
I also mention radio telescope observations to observe the reactions I simulate in the lab and see how well they match up to what is actually in the interstellar medium. I listed three telescopes that are commonly used (again, just enough to show that I have thought about what would be needed to carry out such observations).
Another approach that might be useful, if applicable, is to share "preliminary results" based on previous research experiences. In my statement, I highlight some things related to my proposed research area that I had learned from my summer at the Harvard-Smithsonian Center for Astrophysics and my Fulbright with the Cologne Laboratory Astrophysics Group. I also bolded specifically what I did (it can't hurt to emphasize that you have practice doing research) and the key "results" from this previous work as they applied to my research plan.
Unique resources
I'm not totally sure what the NSF means by "unique resources" that you would need, but my guess is that they want to assess whether you have thought about if you could even carry out your proposed research in grad school. I approached this by naming Geoff Blake from Caltech (who I ended up working with) as an example of a PI who had the laboratory equipment required for the experiments I might carry out. For radio telescope observations, I mentioned that all PIs I was considering for my grad school advisor "frequently secure telescope time; hence, I would have access to guidance from experienced observers when determining telescope settings." Finally, I described needing access to molecular line databases and that, if I had any questions about the databases I was using, I could either head up the street to JPL (who maintains one of the most complete databases) or reach out to my old research group in Cologne (who runs another big-name database). This also served to say, "Hey look! I have a network that I can use to collaborate... perhaps even internationally!" Where possible (and realistic), describe how you might collaborate with other groups and what connections you already have in place.
General advice
Bold importmant statements highlighting your objectives and any related past experiences that make you particularly qualified for the proposed work.
Write for a general science audience and avoid jargon. While your application will be reviewed by people who are "in-field," they still might not be close enough to your own proposed research area to follow along with the jargon. For instance, my application was submitted for review by physical chemists, but a theorist or spectroscopist probably wouldn't know that "complex" is used to describe molecules with six or more atoms in the context of astrochemistry.
Include at least one figure to illustrate something you explain in the proposal.
Discuss briefly intellectual merit and broader impacts, being as specific as possible to show that you have looked into how you might address these criteria as a fellow. For instance, anyone can say "outreach," but not everyone will list specific groups with whom they might engage the public.
