Wednesday, December 17, 2008

The Perfect Number of Proposals

A colleague recently asked me to participate in a proposal. I'm happy to do so, even though I know that I was his second choice. This colleague originally asked an assistant professor who ended up declining because he didn't feel he could manage working on two proposals within a few months of each other. I respect the assistant professor's decision, as I suspect that there are some complicated things going on in his life right now. He made his decision based on what he felt he could and could not handle in the near future in terms of proposal-writing, and he may well have made the right decision.

Or not. He made his decision without talking to his official faculty mentor. That's fine, but it raises the general question: How do you decide what you can and should take on in terms of proposals and projects early in your career? How do you balance what you want to do vs. what you should do vs. what you can realistically do?

This episode brought back memories of my early decisions about how many proposals to submit as an assistant professor. My first NSF grant, lo these many years ago, was a modest one that helped me get my research program started but that wasn't enough to build a research group and tackle some of the big ideas I wanted to pursue.

My second grant allowed me to do some of this, and that was great, but when the grant was awarded, the relevant program officer (PO) told me not to submit any more proposals for most of the duration of my grant and that I should do a good job with this one grant and not be 'too ambitious' early on.

At the time, I was very annoyed by that. The PO seemed to think that the best way for me get started was by proving myself one project at a time, but I felt I should and could do more. If it had been the general practice to give a young investigator one big grant for the first 4-5 years, even if it was a CAREER award, that would have been fine, but that wasn't the situation at the time.

It's not as if I wanted to write 17 proposals, or even 4 or 5. And I wasn't planning on hurling vast numbers of proposals at NSF in the hopes that one or more would be funded. I had some very specific ideas about what research I wanted to do, why I wanted to do it, and how I was going to do it.

I was impatient. I didn't want to wait 4 years to submit another proposal, and I wanted to advise more than 1-2 grad students at a time. So I ignored the well-meaning program director's advice, submitted proposals to other programs, sought other sources of funding, developed collaborations, and built my research program.

I have no regrets about that, even though I can now appreciate that I wasn't being patronized by the program officer, I was being mentored.

But I still wonder: do we have to be so cautious? The practical short answer is yes, in part because funding for basic science research is limited and not everyone is going to get as many grants as they would like. For the sake of discussion, though, I'm going to ignore funding budget constraints. In this discussion, the only issue is how to advise early career faculty about what the optimal number of proposals/projects is, and whether early career faculty should be advised to go slow and not be 'too ambitious'.

I can see how it would be bad advice to say to an early career person: Sure, go ahead, write 3 or 5 grant proposals this year and take on 8 grad students, sink or swim, good luck. And it is perhaps rare for a mentor (a program officer or a faculty colleague) to know enough about their mentee to gauge how much of a research program they can reasonably manage and sustain in the first few years of their career.

Perhaps it is therefore better to err on the side of caution and advise a gradual building of a research program. Along with preparing classes for the first time and figuring out how much service work to take on, not to mention having a life outside of work, we can keep plenty busy even with one research project and a couple of graduate students.

Even so, if a young colleague wanted to try to do more, I would not discourage them unless I had specific reason -- based on their record, not on a one-size-fits-all philosophy -- to counsel caution.

It would be nice if there were an equation that would calculate the Perfect Proposal Number for us, based on all the relevant variables involving work and life. Lacking such an equation, perhaps the best thing that assistant professors like my young colleague can do is to talk with their mentors and discuss the issue. The mentor might not give the best advice, and maybe the advice will be (and should be) ignored, but perhaps the discussion itself can put the issue into better focus for all concerned.


Anonymous said...

Are you trying to raise my blood pressure, FSP?

Female Science Professor said...

That specific goal had not occurred to me, but it does have some appeal, now that you mention it.

JaneB said...

3 or 5 proposals = 8 grad students? FSP, can I come sit at your feet and learn how to get that sort of success rate at funding???

In my experience/field (the muddy end of science) very few grant applications are actually funded - success rates run around 5-10%. In addition, for many sources of funding, revising and resubmitting projects to the same funding stream is generally not allowed. Each proposal is a one-off. So... how does one balance making many applications (and coping with many rejections) against the remote possibility of getting multiple projects funded and ending up with a lot of people (who all need accomodation etc.)? It's tough enough now after being faculty for 11 years and getting a little used to rejection (and the system, and having a few successes along the way too)...

Anonymous said...

very timely post.. I am starting out and have one proposal in and another one in the pre-submission stage. i have been debating if i should continue the grant writing blaze or try to use my startup and nurture these projects. likely they WON'T get funded on first try (just being realistic in our lovely constraint filled time), so it's tempting to keep pushing for something that hits.. thanks for the reminder to seek out mentor(s) - I think the (s) is a key point as no one knows enough about me, my funding opprtunities, etc all in one package.

Anonymous said...

You may have misinterpreted your NSF PO.

In my area, funding is so scarce, we're very, very heavily discouraged from having more than 2 grants at a time, and even that's a rare exception! If we want additional funding, we really have to submit to another program.

Perhaps the PO was not suggesting that you not seek other funding, but simply that you don't seek other funding from him/her? I'd imagine most POs won't come out and openly say, "You just won't get a 3rd grant from me, we have to spread the money around" because that would probably violate some NSF rule, although it's implicitly very clear (in my setting) that that's what's going on.

-- an MSF

Average Professor said...

In my little discipline, the proposal success rate is around 15%. If we assume I could have, at the very beginning of my career, managed one major project, that would mean that I would have been justified in writing 6 or 7 proposals a year until something hit.

In reality, for the first few years I was involved probably more proposals than was really necessary. But I had zero proposal-writing experience and so I justified a good bit of the proposaling by considering that I needed to work on that skill; it wasn't all about the actual research that might ensue.

I also said yes to (almost) every offer to be a co-I on a proposal because I had no network here and needed to build one. Not being the kind of person who knocked on doors around campus just to introduce myself and my work, contributing to proposal writing was the easiest way for me to let other people get to know me and my research.

So, that said . . . I find what the asst prof did to be really weird. BUT, I think if a new person can be successful by writing not very many proposals, that would be super.

Anonymous said...

"I can see how it would be bad advice to say to an early career person: Sure, go ahead, write 3 or 5 grant proposals this year and take on 8 grad students, sink or swim, good luck"

Writing 3 or 5 grant proposals per year cand lead to you to ZERO grants awarded and ZERO graduate students. The funding rate is somewhere between 8-10%, 15% is optimistic. I am in my forth year on the tenure track and in my third year I submitted on average ONE PROPOSAL PER MONTH. I am not sure what the situation of your young colleague is, but I would definitely encourage any assistant professor to submit as many proposals as possible, taking into account the funding enviroment today. I would never say no to writing another proposal, even three within a couple of months. I couldn't afford that.

There is a word of caution though regarding "don't be too amitious". If your hit rate is above average despite the current situation, you may end up with more money than the senior profs. in your dept. and than you become "too ambitious" and attract envy :)) I remember reading in the book "Ms. Mentor's advice for women in academia" her saying something like "do well, but not too well":)) And we know she says that "collegiality" is actually the main criteria for tenure. :))

Female Science Professor said...

The 3-5 proposals/8 grad students comment was not a real suggestion. I picked a large but not too absurd (e.g. 57 proposals) number, so there's no point in focusing on these numbers. They are not impossible to attain, but my point was to mention something larger than 1 grant/1 student, which is what I was initially being restricted to, even though none of my peers were.

Anonymous said...

Perhaps one of the issues here is time management and learning to accurately estimate how long projects take. I am a postdoc and this is one of the skills I struggle with most- I tend to be overoptimistic, even when I try not to be. So it is relatively easy to end up with too much to do and too little time.

Anonymous said...

I still don't understand. You were advised to only have one grant and one graduate student for four years?? And if you had more than one grant, and maybe 2 or 3 students that would be a bad thing?
I can only understand that if that particular grant paid your academic year salary, or a significant part of it. I am now in such a situation, with 75% of my academic year salary being covered by one grant and I therefore shouldn't have too many other grants and I cannot apply for grants for a while. Having too many grants would put me above 100% effort. But in any other situation I don't see why anybody would advise someone not to apply for grants. The Department usually wants money. A PO can advise that you shouldn't apply for another grant from his sections, but if you got a grant from his NSF section, you can apply for DOE for example and I don't see how anybody can stop you.
Coming back to your colleaugue, tell him to stop being moody and apply for grants.

Anonymous said...

I wonder how "real" those 15% funding rates really are. By which I mean:

1. I hear that many, many proposals are rejected without review simply for failing to meet the stated requirements. If you're at an R1 university with a grants office and you're even a moderately serious person, this shouldn't be a problem.

2. Then there are the proposals that are properly formatted but submitted by unserious people (i.e. people who have no ability to deal with federal funds, people who are cranks, etc). I'm sure this is a small but not-insignificant fraction.

3. Then there are the proposals that are properly formatted and submitted by serious people, but are not really related to the program. Based on my experience with workshop and conference organization, I'd think this is a not-insignificant fraction as well.

4. Finally you have the proposals that are properly formatted, by serious people, are on-topic, but are crap. I'm sure this is a not-insignificant fraction also.

By the time you've gotten past all the obvious rejects, surely the acceptance rate would be 30% or higher, no?

I have no inside info whatsoever, so maybe I'm way off. My main reason for thinking I'm not is that I'm in CS, where conferences are the main venue for dissemination, and where the top conferences have lower than 20% acceptance rates---yet my own acceptance rate is in the 70s or 80s at these same conferences. And I'm good, but I'm not a superstar.

Drugmonkey said...

Anon @8:22, when it comes to NIH apps the answer is that the numbers are quite real. The notion that half the apps are from goofallons is a fatal mistake

Average Professor said...

I've sat on a number of grant review panels for another federal agency, and 10-15% is about right within those programs, AFTER the random ones that didn't meet the minimum criteria were weeded out.

For the "serious" grant programs I doubt there are many unserious proposals, since putting one together takes a considerable amount of collaboration, not to mention time and effort.

Anonymous said...

DrugMonkey: But remember that NIH counts computes their success rate per proposal, which can be submitted multiple times. NSF counts their success rate per submission, where there can be an infinite number of submissions until you piss off the program officer and panel, at which time they'll tell you not to submit that particular dog any more. So NIH success rates are really lower than what they appear to be at first glance, compared to NSF.

And though I agree that the advice to go slow is stupid, it *IS* important not to annoy your program officer and every potential reviewer by flooding them with sucky or unnecessary proposals.

Drugmonkey said...


Ms.PhD said...

Actually, I think the PO was giving you sexism. I think women are usually counseled to "not be too ambitious", when in fact we're usually quite good at time management and multi-tasking.

8 is about the size group that one manager can effectively manage. But it's wayyy too many for a new assistant professor.

EliRabett said...

The correct answer is submit as many grants applications as possible in your first year. The lab is not ready, equipment has not arrived, you have no students, and you can use your set up money for cost sharing which does impress reviewers and is sometimes necessary. As a matter of fact, you might be able to leverage some colleagues who are looking for MRI cost sharing to agree to letting you take the lead if they get to use the equipment a bit.

You are going to need those grants to support the grad students who are interested in working with you and will start to show up towards the end of the first year.

You also want to work with the chair/mentors to support your young faculty grants (some of them have to be submitted by the chair)