Thoughts on the NSF CAREER application

Or how I spent 7 months writing 15 pages

I am returning to my newsletter after a longer than expected hiatus. After a month-long trip to India (more on that later), I had to knuckle down and submit two NIH proposals on July 16 and my first NSF CAREER application last week. My NIH proposals were resubmissions, which means they weren’t as nerve-wracking to work on as my NSF CAREER application. I want to talk about how the NSF CAREER is unlike any grant I have worked on and what I learned from the experience.

What is the NSF CAREER?

Curiously, CAREER is not an acronym, but it still abbreviates the “Faculty Early Career Development Program”. I suppose FECDP doesn’t roll off the tongue quite as easily as CAREER. The NSF intends this funding mechanism to support tenure track faculty so that they can “serve as academic role models in research and education and to lead advances in the mission of their department or organization” and pursue “activities that build a firm foundation for a lifetime of leadership in integrating education and research.” Unlike a standard NSF proposal, it provides for 5 (rather than 3) years of support. And carries more prestige  than a regular NSF grant. Many academics call the CAREER the “tenure-maker” although this is a bit of a misnomer.  I’ve known the rare CAREER winners who were denied tenure and plenty more that earned tenure without ever winning or even applying for CAREER. You get 3 shots at winning and assuming a 20% chance of winning every time, you have a 48.8%  chance of winning over 3 tries.

Now, I have submitted (and resubmitted) a standard NSF proposal, with the first submission missing out narrowly. The resubmission, if anything, scored far worse after I took all the reviewer comments to heart and overhauled my technical approach. I realized that I was playing the game all wrong. When reviewers tell you they have a problem with the approach, you must not believe them. What they really mean is they found your idea boring and were too polite to say so. Instead, there is a fair deal of nitpicking about experimental details and choice of controls and a thousand other trivial details. In my naivete, I actually took them at their word and tried to bullet-proof my approach when I should have been strengthening my significance and making them fall in love with the “big picture”.

I was obviously heartbroken after two failed attempts at winning an NSF grant; I had been developing some of those ideas since my postdoc. When I told my PO, I was going to resurrect my standard proposal as a CAREER application, she asked me to take a nice long break away from the proposal. And reminded me that the CAREER is a different beast from a regular NSF proposal.

What makes the CAREER so hard to write?

You have to meet three conditions that you don’t need to meet in a standard NSF proposal. First, you need an educational plan that is seamlessly integrated with your research aims and written as persuasively and as rigorously as your research plan. Second, the research and educational activities you propose need to make sense in the context of your department or institution’s overall mission. Finally, the proposal must make the case that you will establish yourself as a leader in your research field and a role model for future generations. Typical grantsmanship strategies  will not help, trust me. Unlike scientific claims, where you provide hard data to make your case, proving that you are a future leader and role model is a lot harder.

So, in addition to a stellar research plan that is the mainstay of any research proposal, there is this additional element of showing reviewers that you are an earnest, thoughtful, and authentic scholar and educator. You have to allow the reviewer to really get to know you and your story over 15 pages, which is hard for most of us. As much as you may hate talking about yourself, you have to weave in your own story and motivation within the overall narrative arc of your scientific story.   

In some sense, the CAREER proposal is a large-scale version of the NSF GRFP. They are not funding the project but the person. I’ve coached a fair number of mentees to successful NSF GRFP applications, but I still found it hard to follow my own advice and be vulnerable and lay myself out in my CAREER application.

How did I go about writing my NSF proposal?

Unlike the first two tries with my standard proposal where I wrote the proposal first and the 1-page in the end, I stuck to the rules and started with the 1-pager. I’m more at home with the NIH format where you write the specific aims first and then expand it into a full proposal. The NSF format doesn’t make it easy for you to expand a 1-pager into a 15-page proposal, but I stayed the course. I sent the 1-pager around to folks for feedback and refined it over 6 or 7 iterations. Then, I sent it to my PO. I worked on it some more after talking to her but not in the way you would think. I have been very fortunate in the POs I have interacted with at both the NSF and the NIH. They are really thoughtful people who have a long-term view of their programmatic priorities. They want me to be true to my ideas and they usually try to find a good home for my proposal.

 I refined the 1-pager sometime in March and started hammering out the full proposal (which I had started outlining as early as November). Like others, I found it easier to write the research plan than the educational plan. The educational plan emerged organically from my lived experiences, values, and my institutional/regional context and I am really proud of it. I sent my full proposal out to senior colleagues in May, and they were extremely generous with their time and thoughtful with their feedback. So, I spent most of May and early June  fixing issues with my proposal that they flagged. Then I did the best thing possible for my proposal; I didn’t work on it for the entire month I was in India and returned to it with fresh eyes in mid-July after pushing out some NIH proposals.

The proposal really started to “click” in that final week and I somehow found the right sentences and words that gave my proposal a heartbeat and made it alive. It was the most intense and exhilarating week of “flow” I have ever experienced as a writer. I usually don’t put off writing to the last minute, but I found unexpected sources of inspiration and creativity when all that adrenaline kicked in. I had somehow written the entire proposal in 11 pt font (stuck in NIH formatting) and in going down to 10.5 pt, I found space to inject all the pieces that had gone missing.  I also tried out a different editing/polishing technique that I had never applied before. With NIH proposals, I apply the same attention and rigor to every page and every sentence. With my CAREER, I applied a 3:2:1 rule. I edited the 1-pager and the first 4 pages of my proposal three times as intensely, the background and research plan twice as intensely, and the last few pages of the proposal with the normal intensity. I figured this would mirror the distribution of reviewer attention levels, who spend a disproportionate amount of time on the 1^st few pages compared to the middle and the end.

How the NSF CAREER changed the way I think about proposals

So that was 7 months of my life. I don’t know whether I will win this cycle or not. In any case, I don’t regret spending the time I did on CAREER because it upped my proposal writing game. I realized that reviewers don’t really care about technical minutiae as long as they are excited by the idea. You should be about as subtle as a blunt axe in hammering your points: why is your problem important? What do we gain by doing it “your way” instead of how it is currently being done? Who are you as a scientist, what is your story, and why are you the best person for this project?  The order in which you present this info is important. You want to answer these three questions right away before you get to the technical meat of the project. If something is left unsaid, reviewers find it convenient to assume the opposite of what you left implied. I guess I had been writing proposals assuming that people devoted equal attention to the entirety of the document instead of tuning out midway through their read. Not true. Reviewer attention (or attention in general) is a scarce resource, and you want to manage it wisely and show them where to look and what to learn from your proposal. I also realized the value of coming to a proposal “cold” and harnessing all the last-minute surge of adrenaline.