Thursday, October 10, 2013

Can you DDIG it?

I've been spending most of the fall semester so far writing a proposal for the Doctoral Dissertation Improvement Grant (DDIG). The grant was supposed to be due today, but given that the government is shut down, that didn't happen. It's somewhat ironic since the deadline this year is a month earlier than last year's. So in the meantime while we wait for the government to reboot, I thought I'd write down a few thoughts on the DDIG for others, from my perspective. This is not meant to be a comprehensive guide to writing a DDIG. Please refer to the program solicitation to find out everything you need to do.

What is the DDIG?
The DDIG is one of the big grants for graduate students in the sciences, and is funded through the National Science Foundation (NSF). For the DDIG, basically the main qualification you need is to be a PhD candidate (someone who has passed their qualifying exams). The goal of the program is to fund PhD students to build upon their dissertation (ie. improve it, as suggested by the name).

This grant is not for early graduate students, who should really check out the Graduate Research Fellowship, which graduate students can only apply for in their first year (and second, if they have not exceeded a certain number of credits).

Why write a DDIG proposal?
The DDIG is a fantastic opportunity for PhD candidates in the US to get NSF funding for their research. The DDIG will fund research up to $13,000 of direct costs, which is more than almost all grants that graduate students can apply for that I know of, so you can do a lot with it to further your research and do more science. NSF is pretty broad with what it funds for the sciences, so a lot of different kinds of scientists can apply for it; because of this, the DDIG is quite competitive, but it still manages to have a funding rate of about 30% from what I've heard. Of course, because it is competitive, receiving a grant itself boosts your qualifications. I've also noticed that many of the job candidates that come to interview at my school tend to have gotten a DDIG, which almost seems to elevate it to a minimum requirement for tenure-track research positions. And of course, if your aim is to be a research scientist one day, you will need to be applying for grants in such a position anyway, and so if nothing else, writing a DDIG is good practice. Even though I'm lucky enough to be in a lab where my advisor has been positive about funding my research, I don't think that having available money is a good reason not to apply for a DDIG. After all, we're not in graduate school because it's easy.

What does NSF require for their DDIG?
Among other things, a DDIG needs to include what you've done so far for your dissertation and how it provides a context for the proposed work, explicit hypotheses that you will test with your proposed work, and what you propose to do for your research and how you will accomplish your goals. Additionally, like other NSF grants, a DDIG has an "Intellectual Merits" section, which from what I can tell discusses the scientific importance of your work, and a "Broader Impacts" section, where you talk about how you will expand outside of the lab and promote science education, such as to undergraduate students, primary and secondary schoolchildren, and laypeople. Specific to the DDIG, you need to write up a Context for Improvement, which talks about how your work not only improves upon your dissertation, but is also outside of work that your advisor is doing. I think this is to show NSF that you're not just looking for more money, since NSF may be less likely to fund a project if your advisor can already pay for it. There are also various other ancillary things that need to be included as well, such as proof that you are a doctoral candidate in the form of a signed letter from your department chair, letters of collaboration if your proposal includes collaborators, a data management plan, and others. A full list is provided in the solicitation.

Thoughts on how to write a good grant
So finally here are my thoughts. As I'm not a reviewer and I have been unsuccessful at getting a DDIG so far, perhaps my thoughts on how to write a good one may not be realistically helpful. But these are some of the main things I've been thinking about as I've written my DDIG.
  1. Get previously submitted proposals and the reviewers' comments on them, if you can. This will give you an idea of how to write and how not to write, and give you a basis for the organization. Even though the proposals I referenced were successful, the reviewer comments pointed out weaknesses too, not just strengths. This is one of the most important things, since I think having a reference to base your proposal after definitely helps to get you started, even on unrelated topics.
  2. Preliminary data, preliminary data, preliminary data. This grant is supposed to be an improvement grant, so you need to show what you've done so far and how it leads up to the proposed work. It also shows off your abilities as a scientist, which gives you credibility in your claims you can achieve your work. If you want to be submitting a DDIG in a year, then you will want to be working hard now, so you can have preliminary, or even published articles of your own to cite.
  3. You need to demonstrate that the work is feasible for you to do. Preliminary data can be helpful in proving that you can use the tools that you propose to use. If you don't have such data to show you can do the proposed work, despite the fact we're in graduate school to learn how to be scientists, reviewers may be doubtful of your abilities to do difficult tasks. If you lack preliminary data, other ways you can demonstrate that the work is naming collaborators or mentors that can help you do the work. You can also demonstrate your knowledge from the literature, but I'd imagine this isn't going to be as strong as demonstrating you have practical experience.
    Another potential idea that I thought of to demonstrate knowledge of data analysis was to analyze simulated data; since some data is expensive to procure, you may not have data yet to analyze for real. I'm not sure if reviewers will buy it, but I think something is better than nothing.
  4. Organization is important. There should be a logical flow, and you should be clear and concise. I have a bad habit of being too wordy. Reading out loud can help you trim the fat; reviewers need to read a lot of proposals and you don't want to be rejected just because your proposal was hard to read.
  5. Broader Impacts are important. It will help if you are already engaged in doing broader impact-related stuff already, to prove you have a track record of doing this. So like with your dissertation, actively engage in outreach over time so you will have a good track record to expound upon here; this really is the part that makes candidates stick out as not just researchers, but proactive educators.
  6. Explicit hypothesis testing is the framework that science uses. There is a lot of descriptive and exploratory science as well, but I don't think that reviewers tend to think these are the best to fund. Questions, hypotheses, and objectives are the types of things that need to be laid out clearly. This may differ depending on the field that you're working in.
A couple links
I'm not the only one to write about this stuff. Here are a couple links I've been referencing for writing my DDIG, and there are others around as well. The links below highlight certain things I don't mention or emphasize above, so they're worth a look.

No comments:

Post a Comment