zxcv

To win any fellowship, you need to target your application VERY CLEARLY to the evaluation criteria. In the case of the NSF fellowship, this means the intellectual merit and broader impacts criteria (see the application instructions). There are three long essays, so start preparing early and show drafts to anyone who will take a look. Intellectual merit is more straightforward -- you need an excellent research proposal, and evidence that you can succeed in research (e.g., grades, letter of recommendation, prior publications). Being competitive for broader impacts means is about hitting as many of the evaluation criteria with specific examples as possible. It may seem absurd to devote half your essays (or 1/3, more realistically) to broader impacts, but it has just as much weight as intellectual merit. This is why the standard (in my view excellent) advice is to make your personal statement all about broader impacts. (This spring I won three federal fellowships.)

It's 10k per year more. In return you have to tell the DOE two months in advance of any international travel. Is there any serious question here?

I finally (reluctantly) got around to declining an hour ago (I'll take the DOE SCGF instead). They asked for a response by 3pm EST today. Hopefully you guys will hear soon!

For 10k a year, I think it's worth it to suck it up and make international travel plans in advance, even if you do have to "Fly American" even for personal travel -- which I'm pretty sure (though not certain) is not what that PDF meant. My understanding was that the "Fly America Act" only applied to travel the US government paid for...

Looks like the two month pre-approval for international travel is so that all your international travel plans can get logged into a government database, so counter-intelligence agents can make sure you aren't giving away nuclear secrets to agents of foreign powers. Seriously. Here are more details on the requirements. Apparently anytime we are paid a stipend is "official travel:" www.orau.gov/cdc/Forms/foreign-travel-guidelines.pdf

I think I will take the DOE SCGF since it's so much more money, but there are still a couple of nebulous issues: 1) The bizarre two months in advance international travel authorization requirement. 2) Possible conflicts with a fellowship I have for studying abroad in Taiwan this summer. 3) What exactly does the renewal requirement of "demonstrated commitment to the objectives of the DOE SCGF program" mean? I'm not worried about the teaching limitations, since if you want you are allowed to accept outside appointments as long as you will still spend a minimum of 20 hours/week of research. At this point, I don't plan to teach more than 10 hours/week in the future anyways.

I'm in a physics program, but my research is at the intersection of several subfields, including physical chemistry. My research could certainly be considered as "more fundamental science type energy related things." lightyears -- I'm sure you're being sarcastic, but I really do believe it's a good idea to put effort into writing a proposal. Even if this was your last chance to apply, writing proposals is excellent practice for what you do as a scientist. For me, it was also a useful exercise in refining my research plan and what I wanted to work on over the next several years -- regardless of if I won a fellowship or not. meggo -- Sorry to hear about your bad luck. There is certainly an art to winning fellowships, but also a large degree of randomness. If you guys have the chance to apply next year, certainly do it -- I was discouraged after my failures applying for the second time last year (not even honorable mention for the GRFP) since I put in so much time applying, but it finally paid off this year. My application was certainly stronger this year, but not *that much* stronger (I won NSF, NDSEG, DOE SCGF, all I applied for except the Hertz).

My understanding is that if you're still in the same field and program (e.g. "Neuroscience"), you don't need to get NSF approval. Double check the document with award conditions just the same, but the GRFP is widely understood to provide you the freedom to work on the project of your choice. It certainly doesn't have to be the one in your research proposal.

Here the choice seems pretty obvious -- take both, and use the NSF only if by some unforeseen circumstance you need to (you can use it for funding in years 4 and 5).

anir-dendroica -- I think you have the right idea, but may be confused about the applicability of each award. The DOE, for example, funds far more than simply energy research, including many areas of fundamental physics (even String Theory!). I don't think there is any lack of pure scientific prestige there. Right now their offer is most attractive to me (baring unforseen restrictions in the fine print) because working in a national laboratory aligns with my career goals. And their advantage over the DOD is about the money, not the least including the 5k research stippend. I'll still have an excellent chance for a permanent job or postdoc with NRL when I finish my PhD. After all, I will presumably still be doing the sort of research they are interested in. And of course all the awards still go on your CV anyways. BKMD -- You make an interesting point about full tuition reimbursement with the NDSEG. I will be asking administrators about this difference. If it comes down to it, though, I'm sure my adviser has finding that could make up the difference, which would be at least partially offset by the research stippend anyways. If I thought I was likely to stay in grad school for more than three years, I would decline the DOE for the combined NDSEG/NSF, but that seems unlikely to me right now.

Well, lucky for you guys it looked like I will be one of those turning down the NDSEG. I also won the NSF and new DOE but I think I should be able to graduate in three more years so there is no point in taking two. Not sure how I was so ridiculously lucky this year -- it was my third time applying.

I just got an email with official award notification. There were 150 awards out of 3200 applicants.

Read the fine print for each fellowship carefully, and if it's still not clear, contact program administrators to clarify. Some, like the NSF, can be deferred for a few years while you are supported by another. With the NSF and the NDSEG, you can get up to a combined 5 years of support (you take the NDSEG for the first three years, then the NSF). Others, like the new DOE, state that you cannot concurrently accept another major federal fellowship.

This is when you contact your top choice school and get them to let you in after all. It's been done before, and generally it works.

My online status still has my application listed as "Under Review." The area code was indeed 202.