How to win over reviewers for the "broader impact" criterion of the NSF GRFP?

[structuralBio]

So my NSF GRFP application was denied, which i was expecting, but in the reviewer comments they mentioned that i didn't appeal enough to the broader impacts criterion, both in the benefit of my research proposal and in my personal history. Since I'm in the life sciences, my proposed research would have benefits which would be mostly medical (which is a big no-no for the NSF I was told?) I tried to frame it in the context of designing a new protocol that would expand the scientific tool kit which didn't work well with one reviewer. Additionally, my personal broader impact is pretty weak, i really haven't been involved in and kind of community outreach or tutoring that they love to see. So what can I do to buff this area of the application when I inevitably try again?

[juilletmercredi]

The health-related criterion for the NSF is blurry.  In my research proposal, I made reference to my research helping to prevent a number of health-related disorders - including HIV - and I got the fellowship.  I don't remember the exact language I used, but I also made reference to health and well-being and solving a health disparity.  I'm getting my PhD in public health and social psychology, so it would have been kind of difficult for me not to reference anything health-related at all.  I was really surprised when I got the fellowship as a result, but I think it's because my research was pretty heavily social psychological.

 

But...the NSF makes it a goal to award fellowships to people who actually do engage in broader impact activities.  If you don't have any, then you're not as competitive for the fellowship as others.  It doesn't have to be all saving orphans in Guatemala with the Power of Science; if you've done any kind of science outreach activities, that helps.  Teaching a class at the museum?  Mentoring undergraduates in your lab?  Aside from my volunteer activities I also discussed genuinely enjoying to teach statistics and wanting to de-mystify the subject area for students.  There's a sheet that lists other things that the NSF views as broader impacts beyond saving a bus full of children.  They include disseminating science in innovative ways or trying to increase the reach of science into applied policy areas, so if you've done any applied work or tried to marry policy with research, those could be broader impacts.  Basically, they want to make sure that you're not destined to become a researcher ensconced in the Ivory Tower who will only emerge to teach your once-weekly 11 o'clock graduate seminar.  They're giving you taxpayer money, so they want to see you give back.

 

You can also talk about the things you want to do.  Although they need to be buttressed by things you've already done (otherwise they ring false), NSF does like to see people who want to stay engaged in broader impacts work beyond grad school.  So I talked about wanting to serve as a mentor and role model for diverse students in an effort to try to increase the diversity in science (but also made sure to mention that I view diversity as deeper than just racial/ethnic diversity, and talked about how I wanted to inspire non-minority students to work harmoniously with minority scientists, although not exactly in that language).

 

It's all about how you craft it.  In my personal statement I also discussed how inequalities in the outside world influenced my decision to become a social scientist.

[structuralBio]

thanks for the advice. I think I might try tutoring high school students or something in my year off.

[Eigen]

I would say tutoring high school students is never as good as some more in depth outreach. Not to say tutoring isn't good, but lots of people do it and it doesn't help you stand out as much. 

 

I talked a lot about developing science enrichment programs incorporating hands on demonstrations and field trips at a number of local schools, and that seemed very popular. 

 

I think another under-utilized category for BI is outreach to undergraduates/younger students. I talked a bit about mentoring undergraduates through programs on campus, especially targeted to getting women involved in STEM & leadership. Also, things that help get your work out to a broader audience- I ran an interdisciplinary graduate colloquium series that was great for that, and I started a similar interdisciplinary undergraduate research conference when I was an undergrad. Traditional outreach (ie, tutoring) is great, but there are also lots of other way to show the impacts of your work that you may not think about when writing an NSF proposal. 

 

I would consider developing a toolkit for other researchers something central to the intellectual merit of your proposal, but not broader impacts. Generally, healthcare tangential projects (mine was one) that apply to NSF funding (GRF or other grants) focus on the development of a toolkit for probing basic scientific understanding of the target disease. Enzyme inhibitors become tools for understanding signaling cascades, etc. And then NIH gets the application (can be on pretty much the same project) that discusses the direct healthcare benefits of the system. 

[juilletmercredi]

I had some more time today so I went back to see how prominently the health-related goals featured in my NSF application.  The answer is very prominently, lol.

 

At the time NSF was asking for key words and a hypothesis at the top of the page, so here were mine:

 

Key words: substance use, men who have sex with men, sexual behavior

Hypothesis: Perceived racism acts as a stressor in ethnic minority men who have sex with men, which contributes to substance use and unprotected anal intercourse in this population.

 

I tried to weave the broader impacts of my research throughout my research proposal, but I did write a paragraph that explicitly addressed it.  Here it is:

 

Examining a relationship between perceived racism and substance use has important implications for the public health.  Substance abuse increases risky sexual behavior in MSM; when substance abuse is involved, MSM are two to three times more likely to have unprotected anal sex than when drug use was not reported, which can lead to sexually transmitted infections such as HIV (Wilson, Cook, McGaskey et al., 2008).  Understanding this connection is critical for developing culturally-relevant secondary prevention programs to improve ethnic minority MSM’s psychological and physical health, including HIV prevention.

 

My research work clearly addresses a minority population (Black MSM) and contributes to health-related and even disease-related goals.  I made sure that the research proposal was really strongly theoretically grounded on the basis of advice from a professor, and I think that's what helped me to overcome the technical prohibition against health-related research.  Sure, my research had health related outcomes, but I was specifically interested in how discrimination and prejudice was connected to well-being in people, which is a very social psychological research goal.

 

I'm also staring at 3 glaring-to-me typos in my research proposal, lol.

 

In my personal statement, I did mention in passing that I am a Black woman myself, but not "woe is me, give me money, for I am Black" but more "as a Black woman interested in science, I didn't even know anyone with a bachelor's degree, much less a graduate degree.  That's part of the reason I became so freaking fascinated when I found out people got paid to satisfy their own curiosity."  I also discussed how growing up amid social inequality in NYC had made me start wondering how things could be so different for people from different social/racial/economic backgrounds.

 

For the BI criterion, I started by noting that I realized how important mentorship and role modeling were when I started considering my own research career in early college.  Since I had been so helped by other professors, I mentioned that I wanted to do the same thing for students behind me (Having had these experiences, I realize how crucial it is for students from diverse backgrounds to be able to look up to scholars with backgrounds just as diverse as their own).  But I don't think you have to be an underrepresented minority to say something like that - my primary adviser in undergrad was a white woman.  You can just as easily say that you realize it's important for budding science students - especially those who may have been less encouraged to consider science as a career, like women and minorities - to have professors who attempt to understand their struggles and are willing to work with them and push them towards a career in science.

 

I also discussed how my research - which is community-based - drew me closer to the communities I studied and how I really wanted to be steeped in this reciprocal process with communities, disseminating my research in a way that can help them immediately.  I stated wanting to work for a government agency like the CDC, and have my research used in policy.  I also included a little paragraph towards the end about how the NSF GRFP could help me accomplish these goals (and directly alluded to health research: While at Columbia, I hope to integrate social psychology and sociomedical sciences to develop the skills necessary to look at sexuality and racial identity from a psychosocial health perspective,which can in turn contribute to the designing of effective preventive sexual health interventions.) and ended with one last paragraph about wanting to teach and mentor undergraduate students in science, especially in statistics and methods research.

 

*

 

I also agree with Eigen.  It's not enough to just pick something to slap it on your CV.  It needs to be something that you are deeply invested and passionate about, and something that you can very clearly see and explain the connections to science for.  Tutoring high school students is nice, but some of the examples of things they give are "establish special mentoring programs for high school students, undergrads, grad students, and technicians conducting research", "partner with researchers and educators to develop effective means of incorporating research into learning and education," and "participate in the training and/or development of K-12 science and math teachers."

 

So maybe you can volunteer at an after-school program that's specifically targeted for math/science enrichment, or teach a summer program for underprivileged kids interested in science careers, or partner with a science teacher at a local high school to come in and give a lecture or do a simple experiment with students.  Find a way to judge your city's high school science fair (I know this is possible and desirable in NYC!).  Serve as a mentor for a student's science fair project.  Volunteer in a mentoring program, like Posse Foundation (which is in several cities all over the U.S.)

 

When I was applying the NSF had a broad sheet that explained examples of activities that would satisfy the BI criterion (I found it: it's the first link on this page: http://serc.carleton.edu/research_education/linkage.html).  Obviously a lot of them have to do with working with underrepresented groups in science, since that's an NSF priority.  But others are "enhance infrastructure for research and education," which includes things like discovering new ways to collect and disseminate science.  A new protocol probably isn't big enough for your field, but identifying collaborations between institutions and industry or developing a way that new scientists/students can use big research sites are listed as things that can count.

 

Another is "broad dissemination to enhance scientific and technological understanding," which includes partnering with museums to develop science exhibits, involving the public in research and education activities, give science presentations to broader audiences, publish in non-technical journals, and participating in multidisciplinary conferences.

 

A third is "benefits to society."  You have to give specific examples for how your science research will help the community and society at large.

 

Obviously some of the stuff on the sheet is targeted at more established professors since it's the general broadsheet for all NSF proposals, not just the GRFP, but they want to see you doing some deeply-connected stuff trying to get the word out that SCIENCE IS AWESOME, Y'ALL.