How to fail (and sometimes win) at graduate fellowships

The title of this post is rather tongue-in-cheek, but is also meant to be a reality check. In applying to fellowships, you should apply to as many as possible because in all likelihood, you will not get one. You need to try as hard as possible to get one, combing over your essays and reading them aloud before you submit, but keep this reality in mind.

This year I applied to six (6) graduate fellowships and won one, plus a Finalist for one. Last year I also applied to six (6) and won zero, though got Honorable Mention for one. Below is a table of all the fellowships I’ve applied to in the past two years, and their results. If the cell is empty, then I did not apply that year.

In the spirit of sharing, such as those PIs who have publicly posted their funded and unfunded grant proposals, I’m posting ALL my essays. For each year I applied, I’ve put all my documents associated with each application so you can read them and compare the successful and unsuccessful essays. Some are rather personal, but that’s the nature of the essay. Also, the NSF one includes a ratings sheet so you can delve into the mind of the assessors as well. I realize the file sizes are huge but blame Microsoft Word 2011, not me.

After the table, I’ll discuss advice for each fellowship.

Fellowship 2011-2012 2012-2013
DOE Computational Science Grad F’ship no award [16.1MB PDF] no award [18.5MB PDF]
Ford Foundation Predoctoral F’ship no award [94KB PDF]
Hertz Foundation F’ship no award [28.8MB PDF] Finalist [33.2MB PDF]
National Defense Sci&Eng Grad F’ship no award [117KB PDF] Awarded [9.2MB PDF]
National Physical Science Consortium ?? [10.1MB PDF]
NSF Grad Research F’ship Honorable Mention [580KB PDF]
Paul & Daisy Soros F’ship for New Americans no award [13.1MB] no award [661KB PDF]
SMART Scholarship no award [115KB PDF]

Below are the tips I have for each of these fellowships, but keep in mind that I did not win most of them.

DOE Computational Science Graduate Fellowship

I can’t give advice for this one, other than to apply. The first time I around I just talked about science, the second year I emphasized how my research needs lots of awesome computers but none of those strategies seemed to work. Those people also require a course plan, so maybe they though mine wasn’t serious enough. No clue.

Honestly I thought I had a pretty good shot at this, since I got to the Finalist stage for the Hertz Fellowship, but it goes to show you that each of these fellowship associations have subtly different standards.

Ford Foundation Predoctoral Fellowship

This fellowship is aimed at those who are going to increase diversity in the professoriate, whether by being a minority themselves, or by participating in programs that encourage minorities to pursue research and academia. I applied as a member of the latter group.

I have to admit my mistake on this one. My undergrad can submit transcripts electronically, and I didn’t check with them to see if they accepted e-transcripts. I got an email in February saying my application was incomplete, which sent me into a panic and I logged on to the transcripts website and overnighted a transcript to their office. But then I read the email again and saw that they “encourage you to apply and wish you continued success in all of your academic endeavors.” :(

Conclusion: check with the fellowship organization about their accepted transcript formats and whether they have received all your materials

Hertz Foundation Fellowship

This is definitely one of the most prestigious fellowships. The first year I applied to this, I was not very scientifically mature, which you can see from my application. But in my 2012-2013 application, I had a vivid idea of what I would like to see in the future of science, and how we were going to get there.

There are very few resources (save for here, here, and here) for the Hertz so I’ll detail my journey below.

Note that the filename of the fellowship document is Olga.Botvinnik_Hertz_2012-2013_v6.pdf. I did, indeed go through six versions.

The Interviews

Round 1 Interview

On Nov. 16 2012, I received an email indicating I had been selected for a first-round interview! Last year, I received the rejection notice on Nov. 11 2012. Given that their due date is usually Nov 1st, it was nice to at least get a quick turnaround. An excerpt from the “Reject” email:

Thank you for your application to the Hertz Foundation Fellowship Program for graduate study in the applied sciences.  We greatly appreciate your interest in the Hertz Foundation Fellowship Program and we strongly encourage your continuing contributions to science and technology.

We received over 600 applications this year and were able to choose only about 25% of those for personal interviews.  Unfortunately, we were not able to select you for an interview and therefore, you will not be part of the next round of our selection process.  We regret to inform you of this fact, particularly since the Foundation values the pursuit of applied science and engineering of creative young PhD candidates like yourself.

Your application was given serious consideration.  In addition to scholastic excellence, there are many other factors associated with the objectives of the Hertz Foundation that we take into account when selecting candidates for interviews.  Our decision was based on the total assessment.

Thank you for choosing a career in science and technology.  We are sure your efforts will make a difference in the world.

I asked for feedback, especially because one of my recommenders didn’t submit their letter on time, but they said they do not reconsider applications and provided this note from the president of the foundation about the decision process:

We do not provide feedback on individual applications.   In our process, two reviewers read each application.  They look for not just academic excellence, but creativity, early demonstration of research capabilities, motivation and confidence, and the strong endorsement of those traits by referees.  We also look for leverage in the use of our resources.  If you are already well along in graduate school, we question what our contribution to an established research program may bring.

If you were not selected for an interview, it does not mean that there was any defect in your application or that it was not considered and appreciated.  We rarely see an applicant that we would not choose to fund were resources available.   Like Rhodes, Marshall and Churchill, our success ratio is slightly under 3%.  Unlike them, we invest heavily in sequential one-on-one interviews that depend on the volunteer effort of many excellent scientists and engineers.  We must make a hard selection early on to optimize the use of these resources.

And the “Interview” email:

Congratulations! You have been chosen to advance to the interview stage of Hertz Fellowship selection process - as one of slightly more than 20% of this year’s applicants. This involves a formal technical interview with one or more of our interviewers - who are in most cases former Hertz Fellows, themselves. This interview generally lasts 45 to 60 minutes. It is patterned after the PhD oral exam and you may be asked to perform calculations, discuss your previous research work, and to demonstrate the breadth and depth of your technical knowledge. Please bring paper and pen in the event you’re asked to perform calculations during your interview.

If you have new scientific papers that were not submitted with your application, please bring 2 copies for your interviewer. The materials will then be added to your file before it is reviewed for consideration for a second interview.

The interview itself in Los Angeles on Jan 26th 2013, and I carpooled there from San Diego with a couple other people at UCSD who also were invited to interview. I dressed in a suit, since they said it will be a formal interview, and I was crazy nervous. But once I got into the room and started chatting with these crazy smart people, it wasn’t that bad. It just felt like a conversation about science. They asked me random questions like about how does Purell work and why it’s preferred over hand washing, about hydrogen bonding in glass, and some statistical questions. I had a really good time talking to them! I think being relaxed and taking my time with answering their questions made a huge difference.

Round 2 Interview

On Feb 1st 2013, I received an email notifying me of my Finalist status. Here’s some of the email.

We are pleased to notify you that you have been designated a Hertz Fellowship Finalist - one of 50 selected from a pool of over 700 very talented applicants. 

I was so excited! The day that I found out was the first day of my lab’s ski/snowboard retreat at Big Bear Lake/Mountain, and we were driving up. I was refreshing my email on my phone constantly until I found out at 11:55am.

I used Zimride/Craigslist rideshares to get there (I’m having car troubles which is why I’m avoiding driving it long distances). I was a lot more nervous for this one, and I think it showed. They asked me to explain how I “got here,” as in what was my journey in science. The questions where I really faltered were about where I have applied my own creativity in research, and some of the physics/stats questions. Though they were impressed that I had applied two years in a row, and pleased that I completely changed my application, because I guess they get people who submit the same thing. Which doesn’t make sense to me because if it didn’t work the year before, it probably won’t work now…

As they probed me about where I applied my creativity in research, at first I asked if I could talk about a teaching experience because I’ve taken creative approaches to presenting material. But they said no they weren’t really interested because after all it wasn’t like I invented inquiry-based learning (!). So then I racked my brain for where I have done something original in my research, and it dawned on me that most of my research has been following someone else’s instructions. I’ve been creative in the implementation of these instructions and coded up nice solutions, but I haven’t been the originator of the idea. They said “Impress us. This is where we want you to show off your accomplishments.” And.. I didn’t.

After the interview I realized that in my Bioinformatics Algorithms class I formulated a really interesting biological problem as a solid computational problem, and this had never been though of before. So emailed that to them, but maybe it came off as too desperate, or not good enough in the face of the accomplishments of my peers. So in any case, I was in the process of doing something really creative.

Another question they asked was about where I saw the frontier of science being in the future, and I said “non-destructive single cell genomics.” And they they said “Great! So how would you do that?” And I had to invent all kinds of technologies to accomplish this wacky future, on the spot. That was pretty draining.

With the stats questions, I should have got them because they were simple binomial distribution and heat diffusion through spherical volumes problems, but my brain was so dead from feeling inadequate about my lack of creativity, that I took a really long time to answer them, and then I didn’t even get them in the end.

Tips from a Hertz Fellow (not me)

  • Practice some scenarios of questions they may ask of your past research, or a general “tell me about yourself” question. This will help you be less nervous
  • Study the publications of your interviewers and try to find something in common. These are people with PhDs who are curious about the world, so it’s better to get them thinking about your topic than for them to sit and come up with “What can we ask that s/he probably doesn’t know?”-type questions, which are no fun.

Thanks to Christian for the tips!

Conclusions on Hertz Fellowship

  1. You need a strong vision for the future of science, supported by your current research.
  2. You need to have demonstrated creativity in your field by suggesting a new solution, approach, or technique that people go “wow, why didn’t I think of that before?” Simply following instructions to perform other’s research is not enough.
  3. Be quick on your feet with statistics, physics and differential equations.

National Defense Science and Engineering Graduate (NDSEG) Fellowship

Last year I adapted my NSF fellowship for the NDSEG. The NDSEG is more competitive, with ~200 awards out of ~3000 applicants (~6.7%, whereas the NSF is ~2k awards out of 20k applicants, 10%). While by stats the NDSEG is more “prestigious,” it is not as well-known. I’m not quite sure what makes a proposal for the NDSEG good, other than similar criteria to the NSF, so I suggest comparing my successful and unsuccessful fellowship applications. I’ve heard that you should describe potential military applications of your research in the NDSEG applications, but I didn’t do that so… who knows.

Update: Others have informed me that the NDSEG seems to go after recent accomplishments such as publications, whereas the NSF is more about your potential for future research. Thanks to Reid for the tip!

National Physical Science Consortium

This seems like a relatively unknown fellowship because I discovered it randomly. They also have a terrible website where you generate a key instead of a login, and then the key can change… I had to use 3 different keys for this application. And they basically say to use your NSF essays for this application, which is why the header in the file is “NPSC/NSF-Style.”

As for advice, no idea on this one … they still haven’t gotten back to me (?!), even after I emailed them. So I’m guessing this is a “No.”

National Science Foundation Graduate Research Fellowship Program (NSF GFRP)

In case you’re wondering, I didn’t apply to the NSF 2012-2013 because I have a Master’s degree, which makes the NSF GRFP hate me.

As the most well known of these fellowships, there are many fantastic resources for applying to the NSF GFRP (even more here, here, here, and here), so I will just give some general advice.

  1. Research Proposal
    1. Speak fluently about your topic. Sound like you actually know what you are doing. This will come when you have many people read your fellowship app and ask questions about the details. The actual topic doesn’t matter, but you need to convey that you are competent in your field.
    2. Broader Impacts: Your research should be benefitting the scientific community as a whole, whether you are depositing data publicly, contributing to open-source software, or training new people in science.
  2. Personal Statement
    1. Broader Impacts: Make sure you are doing some kind of outreach, either as a TA or a volunteer or something. And don’t just do it for the fellowship because that’s pretty obvious … You should have a history of outreach.
    2. Tell the story of WHO is going to be doing this research. Grad fellowships fund people, not projects.
  3. Previous Research
    1. If you feel “inadequate” because you didn’t do research in undergrad but had classes instead, talk about a class that opened your mind beyond the traditional curriculum and into the unanswered questions of research.
    2. Again, speak fluently about your topic.
    3. Broader Impacts: How have your research experiences benefitted the scientific community, either at large, or a few people? Talk about it.

Thanks to Alex for emphasizing Broader Impacts!

Outside of those three, the “topic” that you apply to is very important, as that determines who reads your application. In Bioinformatics, we’re in a bit of a no-man’s land since if we apply to Computer Science then they won’t understand the biology and say our algorithm is crap, and if we apply to Biology, then they’ll say we don’t understand the biology, and gloss over the computational details. So you have to find a delicate balance.

Update: Finally, the NSF is the only fellowship of these 8 (!) to give feedback, so if you only apply to one thing, apply to the NSF to get a feel for the process, and to receive feedback on your first application. Then in your next application, you will know what’s going on and can embark on a more serious application. Thanks to Reid for pointing this out!

Paul and Daisy Soros Fellowship for New Americans

The PDSoros Fellowship is very difficult to get, and prestigious. It is for immigrants and children of immigrants (I was born in Russia so I count as an immigrant). Like the Hertz Fellowship, they ask you to focus on times when you have applied creativity in your field, and since I’ve had that realization with the Hertz, they probably also saw that I haven’t been that creative. Unlike the Hertz, it is for people pursuing ANY graduate degree, so you are competing with people in law school, medical school, PhD programs, M.A., M.F.A., M.P.H., M.S., M.Eng., etc programs. So it is a very large pool.

Plus I didn’t proofread my essays enough. I looked through them later and saw typos and notes to myself, which is not professional.

Science, Mathematics, and Research for Transformation Scholarship

No idea. Didn’t get any feedback about the first time, only this email:

This message is to inform you that you were not chosen as one of the 130 Round 1 Finalists for this year’s award competition. Please note, a second round of up to 20 additional awards may be issued this month. Your application may be reconsidered for an award at that time. ASEE will keep you apprised of your status as new information becomes available. We appreciate your interest in the SMART Scholarship Program.

But I didn’t get an award for the second round either. I didn’t apply the second time because this fellowship requires a year of service in a Department of Defense lab after you graduate, and I wanted more freedom with my fellowships.

Go Top
comments powered by Disqus