Christopher W. Fletcher

[Graduate School] Fellowships

Created: October 9, 2011
Updated: May 10, 2012

Having a graduate fellowship means different things to different people. A fellowship's importance depends on your adviser, your project, and your school. There are small fellowships (one year fellowships plus others that are worth less than $100,000) and big/long fellowships (typically three or more years, worth more than $100,000). This post is about the big fellowships that are not school-specific. For science and engineering students this means the NSF, NDSEG, and Hertz. I applied for the NSF and the NDSEG, got the NSF, and was rejected for the NDSEG. Update: I reapplied to the NDSEG during the 2012 season and got the award on the second time around. I added a section on this at the bottom.

Frankly, this subject has been beaten to death by other people; here are some good reference links:

I don't want to repeat too much of what they said, except to say that I followed their advice for applying to the NSF and got that award. I did not 'tune' my NDSEG application to match the needs of the armed forces, which may have been my application's downfall (Phil Guo puts a lot of emphasis on doing this). In terms of what this post can do to help you, the reader, attached below are my NSF essays (Jean Yang linked hers as well). I found that reading sample essays was very helpful, especially because of how political the NSF fellowship was and still is (see the "broader impacts" part of the application).

Looking back, the NSF was a really long application. I started my essays about a month and a half before my program deadline. In retrospect, that was the perfect amount of time as each essay needed a lot of going over with other people. Did I say a lot? I mean a lot! My research proposal went through several complete re-writes and about twenty smaller revisions. It helps a lot to get peer feedback for the research proposal as well, and that always takes more time (see here). In terms of the amount of time each essay takes: the statement of purpose took about four times as long to write and revise as the research experience essay. The research proposal took another four times the statement of purpose. I am not joking, the research proposal is a hard essay to write (or at least it was for me). Aside: even though the research proposal was the most time intensive essay (as it was technical), all three essays will require a lot of revision. The only way I got through those essays was with the help of my mom as a third-party reviewer. Find someone who will back you; it makes a world of difference. My essays wouldn't have been close to as strong as they were without her help. Thanks mom! :-)

The NDSEG, on the other hand, is several days of work. There is a single 3000 character essay requirement. Your experience (leadership, teaching, etc) has to be condensed to hundreds of characters (that is not a lot of space). My peers say that it is best to finish the NSF application and then do the NDSEG (shrink your NSF research proposal to meet the NDSEG character limit, basically). To be honest, I found it fairly challenging to shrink my proposal to 3000 characters. 3000 characters is a pittance compared to each NSF essay (my research proposal was almost 7800 characters). In practice, when you fight tooth and nail to squeeze your essay to exactly two pages (for NSF), cutting it down to less than a page usually requires a different pitch altogether. So, don't think of the NDSEG essay as a matter of cutting the NSF essay. Expect to write a new essay that is 3000 characters in length.

NDSEG, Round Two

Note. I wrote this section roughly a year after writing the above section. I left the above section unchanged after getting the NDSEG.

I reapplied to the NDSEG a year after being rejected and got the award the second time through (award year 2012). It really comes to show: if at first you don't succeed, try try again. Aside: this philosophy has helped me before.

The big difference between my application last year (rejected) and this year (accepted) was focus: this year my essay and area of specialization catered to the DoD broad agency announcements (BAAs). Philip Guo hammers this point on his website; I highly recommend reading his article on the topic.

My area is computer systems. Last year, I put Computer systems and architecture as my area of specialization and have two retrospective criticisms to make. First, "Computer systems" is too broad/vague. Second, "[Computer] architecture" was not in any BAA. This year, I used Data processing and heterogeneous computer systems. Both of these were in multiple agency's BAAs, and data processing can be tied to multiple other hot areas like "big data" and throughput computing (which helps when crafting essays).

Here is the 3000 character essay that I used this year:

This is a plain text file to emulate what you will submit on the NDSEG website: you have to copy-paste into a form that counts characters. Contrast this with the NSF, which accepts PDFs and checks your submission by page count. Disclaimer: Despite getting the fellowship, I wasn't very happy with the essay after submitting it. The last paragraph read as if it was tacked on (it was) and there are too many commas throughout. I guess sometimes you just get lucky...