Monday, November 30, 2009

Broader Impacts R Us

If you are a female PI on a proposal that requires text about "Broader Impacts" (to use the NSF term), which may include the extent to which the proposed research activities increase (broaden) the participation of underrepresented groups, and if you are a member of an underrepresented group in your field, do you explicitly mention in the proposal that you are a broader impact?

Or, if this fact is obvious, do you not mention it and focus instead only on other examples of how your research will fulfill the Broader Impacts criteria?

I get asked about this a lot.

Last year I wrote about how I got blasted by one proposal reviewer who was extremely disgusted by my inclusion, at the end of a list of all my proposed research's broader impacts, that the project would support the research of a female scientist. I don't even know why I mentioned such an obvious fact; I was mostly just being systematic about going through the possibilities.

The NSF program officer put a line through these hostile reviewer comments and said they were ignored, but the overall review, including that reviewer's ranking, was considered. It was the only negative review but it was enough to sink the proposal out of the fundable range.

That was an extreme example, but I have seen cases in which male PIs who write about how they will involve female students in their research get higher marks for broader impacts than female PIs who are broader impacts. Some program officers view as inappropriate the criticism that female PIs are using their gender as a grant-getting tactic, but if one or more reviewers knock their ranking down a notch (or two) in anger about female-PIs-as-broader-impacts, the overall consequences for a proposal can be dire.

Of course there is more to "broader impacts" than involvement of underrepresented groups. And female PIs have to do more than just be passive "broader impacts". As is the case for any PI on an NSF proposal, we need organized and serious plans that recognize the importance of educating and training students and postdocs, that enhance connections with industry or government agencies, that promote the communication of scientific results to the public, and/or that benefit society in any of a number of other important ways. In my research, a significant broader impact typically also involves my close collaborations with international colleagues and students.

I am on board with all that.

I am curious, however, as to whether female PIs (or other members of underrepresented groups) deliberately mention/don't mention themselves as a broader impact. Owing to the lack of women in my field, I seldom review proposals by other women, so I don't know what others typically do. I now leave it off my list of broader impacts in proposals because (1) it's obvious and (2) it might be a magnet for the hostile women-have-an-unfair-advantage reviewers.

26 comments:

zed said...

I don't mention it. It's obvious, and I think it comes across as 'I couldn't come up with anything else to write in this section", regardless of how much else you are actually proposing. Kind of like mentioning that you will be training graduate students. I *do* point that out, but it's more filler than anything else. However, I once wrote a proposal with two other young female scientists, and we mention it in the broader impacts section.

Anonymous said...

I'm female and an ethnic minority, and I have not mentioned myself as a broader impact on NSF proposals. The reason is that I feel the spirit of the NSF broader impacts section is to use those NSF funds to directly improve the situation for trainees, not for PIs. One can always argue that by giving the minority PI a leg up, you are helping them to help similar underrepresented groups of trainees but that is indirect and thus not a good enough justication in my opinion. But who knows, maybe I would have better success with NSF funding if I did include myself as a broader impact?

kelle said...

I've been on a couple NSF panels (and I'm a woman) and when a female PI doesn't explicitly mention herself as a broader impact, I assume that she wants to concentrate on the other avenues her research is having broader impacts. When she does mention it, usually in the context of the larger number of female students she has attracted and/or her mentoring activities, I find it very compelling. Bottom line, if it's important to you and you spend time on it, than you should mention it. Omitting it due to fear of offending someone is directly in conflict with the spirit of the Broader Impacts criterion. Based on my experience with NSF Program Officers, that offended panelist would likely not be asked to serve on any future panels.

also, I like to point people with questions about the broader impacts criterion to this document that provides lots of examples: http://www.nsf.gov/pubs/gpg/broaderimpacts.pdf

Anonymous said...

I slip it in in a very offhand manner at the end of the BI statement. I have also seen a lot of hostility towards female PIs who claim themselves as members of underrepresented groups. On the other hand, I have also seen panelists write their summaries by paraphrasing the BI statement in the proposal, and if the PI doesn't mention being female, then it doesn't go in the panel summary. That's why I slip it in.

And of course, women who have names that are not clearly identifiable as female ...

It outrages me that we have to worry about being trashed because we include information in our BI statement that NSF specifically requests.

Anonymous said...

In my particular subfield, women are less than 20% of the faculty. I don't mention my being female as a broader impact. However, I do give presentations to women in science student groups and serve on committees to increase the representation of women in my field, and I list those activities to stress that it's not just my being female that is a broader impact but rather my work actively promoting women.

Janus Professor said...

I am female and I consider myself disabled. I do not mention in the BI that I am disabled because it seems like I am grade grubbing. I write that "I am passionate about serving as a positive and outgoing role model to women, minorities, and impaired persons." My hope is that the reviewer can do the math in that statement.

John Vidale said...

My impression is that broader impact statements are just a detail that needs to be checked off. If there is no explicit BI statement, that's a problem, but if there is a plausible statement, the details are unimportant. Truly laudable BI efforts help proposals, but they are not common.

Also, for the panels on which I've served, it is the panel vote and not the reviewer vote that matters. Initial ordering from the reviewers can matter, but after ordering, the rest is up to the panel.

Finally, an erratic and nonsensical comment or two in the reviews is normal. The reviewer blast would do more to discredit the reviewer than prejudice the panel against the proposal.

Female Science Professor said...

If only it were so, but some panels only discuss proposals with numerical ranking above a particular limit. A negative review, whether justified or not, can put a proposal below that limit so that there is no discussion unless a panel member requests a discussion.

Ianqui said...

I just sat on an NSF panel, and there were a fair number of proposals by women. None of them mentioned themselves as a broader impact, and only a small number (well, a small number of all PIs) mentioned training opportunities for female students. A bigger deal seemed to be made of minority students and postdocs. Maybe in my field, which has more women than other science fields, it's not as obvious a broader impact.

John Vidale said...

just a detail - if the broader impact blast was the only negative remark in the review, and the rating in that review was fair or poor, and the rest of the reviews were excellent, I'd have expected the program officer to reject that score.

If the rabid reviewer offered a good or very good, and that was enough to sink the proposal out of the bottom of the grey zone into the don't-bother-discussing, then the prospects for funding were unlikely even excluding that review.

Micro-inequities require sufficient density to have much effect, and the attribution of a big effect to one ridiculous criticism in a single review seems implausible to me.

Meadow said...

I didn't think of mentioning myself as an example of broader impacts in action when submitting proposals in the past. Maybe in future....

I know what you mean about one nasty over the top review pushing the proposal into the not-funded group even if the program director claims to discount some of the more absurd statements.

Anonymous said...

hmm.. I don't think I explicitly point out that I am a woman. probably more implicit (talking about mentoring young women, etc). I did however advise my Hispanic graduate student to point out that he is fluent in Spanish and will use that to help mentor underrepresented minorities in his NSF proposal BI's... we'll see how that flies. I have seen a female minority use the same point in her successfully funded proposal.

Kevin said...

I agree with John V that significant "Broader Impacts" statements are fairly rare. Mentioning mentoring and outreach activities is usually good---mentioning one's own gender or ethnicity seems a bit tacky.

I think that whether gender gets mentioned is very field specific---as some fields have huge gender imbalances and others do not.

Anonymous said...

I dont agree that a reviewer should take a dim view if someone decides to include a mention of their gender in the grant proposal. Doesn't every job on offer say that women and minorities are encouraged to apply? If such is the language of the call for applications, it is understandable that such will be the language of the proposals that roll in.

Scientists have let themselves be backed into a corner where gender is a consideration that is taken to be at par or even above scientific merit. We made the bed and we must lie in it. Soon science will join the ranks of the other academic disciplines where you can become a UCLA professor by suggesting that Newton's Principia is a rape manual.

Anonymous said...

I'm a graduate student who just received an DDIG from NSF and had not originally included in my BI that I am female (my name is a dead give-away). I was told by the program manager to include it after it was recommended for funding but before final approval.

LadyScientist said...

On a related note, can you list your blog as a Broader Impact or as outreach that you do? Your blog certainly has positive broad impacts. But yours is anonymous, so either they would have to take your word for it that you have a blog with broad and positive impact, or you would have to give away your identity. Hmmm...

Anonymous said...

I'm a woman and I don't include myself as a Broader Impact for NSF because I interpret the BI mission as being it is the PI's duty to impact others (with the help of the NSF grant). Not that the NSF funds are meant to have an impact the PI. Thus, as a PI the focus of the BI plan should be on others, not on one's self.

EliRabett said...

On single PI grants it is a bit touchy but it can be incorporated most easily by mentioning mentoring to female students, but you can be damn sure it is mentioned in center grants as in "this diverse group of investigators includes....with an optional ...who can provide mentoring to all of our students and recruit....

Anonymous said...

I don't mention myself (female PI) as a broader impact, but I have had reviews that pointed out that I was a broader impact.

I recently reviewed a proposal that mentioned the PI's maternal status ("the PI is a mother with small children") followed by a couple lines about how having children leads to leaky pipelines. I found this quite inappropriate and did make a comment to that effect in my review.

Anonymous said...

I remember explicit directions from the NSF program manager, while we were on a panel, to ignore any claims of impact based on who the PI *is*. (The PM was female.)

Being a woman, minority, disabled, etc. was explicitly NOT a contribution.

The panel members had to examine what the PI has done to improve diversification, not what the PI is.

It sounded pretty fair to me.

Anonymous said...

damn straight I list myself as a BI. And my african project co-director. no apologies for this. If it ever helps me, that will only go a short distance to make up for all the times I got left out, over looked, or ignored b/c of my BI status.

Anonymous said...

I just don't think my being female is a broader impact. I do think that my recruitment of students who are both male and female with a variety of backgrounds is a broader impact. Their dissemination of knowledge and equality is an even further broader impact. My first advisee (I just started last year) is a white male. He's not what I would consider to be an advantaged white male student, because he is a first generation college student (also he is very insecure with himself). Thus, I provide him with opportunities for his professional and personal growth in a similar manner as I do with female students. This is exactly what I needed when my mentors where male, to be treated the same way as the male students.

Anonymous said...

I'm not sure how seriously I would take claims that a PI is a broader impact, especially if there is no evidence that the PI has used their status to encourage and/or recruit. Yes, it's nice to have PIs who look like you...but honestly, as a female student, much of the open misogyny I encountered came from older women.

But if the PI has some evidence that they are a good mentor to people of their type, I don't see why that can't go in. It's part of the formal criteria. (Disclaimer: still a student.)

Kitty said...

I mention neither my gender (female) nor race (non-White) in my BI statement. It's funny -- I am not sure it has ever occurred to me.

Zuska said...

Based on the list of examples in the link that Kelle provided, I would say that listing one's self as a person of the female persuasion who will be supported as a PI on the proposal doesn't exactly qualify as a broader impact. From my reading of the examples, it would seem that NSF has something more expansive in mind than "just" supporting PI's from underrepresented groups. If you look back at the history of how some of diversity efforts have evolved over the years, this is not surprising. I believe things like the POWRE grants - which went to individual women PIs - eventually morphed into the ADVANCE program - which is intended to promote institutional transformation (as opposed to supporting individual researchers here and there, leaving institutional structures unchanged). Broader impacts are supposed to be just that - broader. That is, what effect is this grant going to have beyond this one PI and this one project and this one lab? How is the larger scientific community - including the scientific community-to-be - going to be affected?

Anonymous said...

I'm writing a proposal and I'd been wondering about this myself. In my case it's not necessarily completely obvious. My name is of Asian origin and in past years people on the Internet have assumed I was a man on multiple occasions (generally in a technical context; my field is very male-dominated). Nowadays I have a picture on my homepage, though, which is easily found by googling me. I'm really glad you brought it up, as now I've decided to leave it out (and my proposal has plenty of other broader impact anyway).