Wesley Soul Crusher

One down, two to go.

It’s the middle of the grant-writing season for many of us NSF folks, with a number of divisions having deadlines in late October or very early November. Mercifully, one division went to year-round submission, so the third grant is what I will be working on in November.

I am in that unenviable spot where I am exhausted and need to sleep, but my brain is in overdrive, so I don’t sleep well. It’s also the weekend, when I should be getting some rest, because I have kids who are only young once and they need me around, and because I need the rest so I would actually be able to work full steam during the week. Yet I can’t get the rest because my head is on fire and the clock is ticking. At times like these, I completely understand the impetus to start abusing prescription stimulants. Not that I ever would (caffeine is my drug), but I think I understand how someone who chooses to go that route feels when pressed against insurmountable workloads.

I have to say I envy NIH folks on having multiple submission deadlines throughout the year.  I wish I could spread it out a bit and not kill myself every October.
Before you ask why I don’t spread it out and write sooner, I did, it’s just there is so much other stuff, like the actual science with my students, and getting the papers out, that I can’t spend all summer writing grants too. We have to do the work and publish it for the grants we already have. And there are also spring deadlines in other agencies. And, you know, giant classes to teach.

*****

Many people I know now routinely submit two grants per unsolicited window (so once per year). I do it too, and still find it very hard, as I need an intellectual break to do a good job on both, but there is no time for it. Earlier this week I submitted a grant with a new collaborator and it took a lot of time and energy from me, because it’s completely new and it’s always tough trying to follow someone else’s vision and still pull your weight. It took longer than we had agreed upon and thus cut almost a week into the writing on my single-PI grant, the one I am currently working on. It also left me drained and unwilling to work on said single-PI grant, even though I intellectually know that it is much more important to me. On the positive side, the it’s a revision, so I will be able to use some of last year’s material.

*****

I chatted with someone the other day about what it is that makes grant rejections so soul-crushing.

Paper rejections happen and they are nowhere nearly as devastating as grant rejections.

There are several aspects to it, I think. First, there is actually the livelihood of your trainees and staff at stake, and even your own if you are in a soft-money position. The real devastation that comes from not being able to pay the people you are responsible for. A colleague told me today that several of his students would simply be unfunded starting in the spring. He seemed flippant about it, like it wasn’t his problem, which really bothered me, but perhaps I shouldn’t assume anything, because I don’t know what goes on in his head. I know that I often have to do remarkable gymnastics with the funds to make sure everyone is covered at least minimally if a grant doesn’t come through, but I have never just thrown my hands up in the air and said, “No money, you are on your own!” to anyone to whom I have committed as an advisor. This responsibility to keep everyone funded really weighs on me, heavily.

Then there is the fact that grants are written for a hostile audience. When you write a paper, most referees try to be constructive and help you improve the paper. In contrast, during grant review, the goal is to find a reason not to fund you. The comments are not for you, they are advisory to the program director. The comments are not aimed at helping you and they can often seem completely unfounded. Here are my favorites from last cycle, in which none of the scores I received were below a “very good.”One person simply doesn’t believe I can use a couple of techniques even though I have papers and preliminary data and figures showing that I can, and indicates that I should get an expert (read: chaperone) to make sure I do it right. Another person doesn’t comment on what I proposed to do, which is very complicated, and instead complains that I didn’t say I would do something else instead, which is even more complicated and simply so far out there that nobody can get to there from the current state of the art with one student and $300k total over three years, and most definitely not before what I proposed to do is done first.  A third person claimed that I was confused about a technique that I (and others) routinely use and that the technique cannot do what I say it can do (which it can).

So you get these comments that are not only not instructive, but can be misguided, hostile, and often make no sense. And in their senselessness, they convey one simple thing: the panel (or ad hoc reviewers) just didn’t like your proposal and don’t want to see it funded, and there’s usually nothing more than that. The written comments make no rhyme or reason precisely because they are post hoc justification for the fact that the reviewers just didn’t want your work and liked something else (or perhaps someone else) better.

Finally, in contrast to manuscript review, you have no opportunity to respond to the jabs in proposal review. You just have to try to retool the proposal for next time, to be read by a completely new and different set of people. And hope you get 3 or 4 of them who are not cranky or slightly biased against uppity women, and hope at least one of them will actually decide to champion you over someone else. I am not sure the probability of this happening is even 10% for most mere mortals.

(I know, I know. Excuses, excuses. My colleague from above would call this “loser talk.” I know I am whining and should roll up my sleeves and get back to work.)

Young grant writer Wesley Crusher.

Brand new Ensign assistant professor and grant writer Wesley Crusher…

Principal Investigator Wesley Crusher, many, MANY grant submissions later.

… and many, MANY grant submissions later.

6 comments

  1. It’s not “loser talk”. It’s a recognition that the grant system that we’ve got is not working for us. There was a time when lab support was more stable – when an NIH grant renewal that was productive in the previous cycle was nearly always guaranteed to get funded (particularly on the second round – after one had responded to the reviews, which were more like manuscript reviews trying to create better science than the current grant reviews just reporting to program) – or when an NSF grant was icing on the cake, but not critical to lab continued functionality, because there was enough internal money (generally coming from state, internal, and tuition dollars) to ensure that students had TAships or RAships available.

    It sounds like the NSF world is falling into the same pit that NIH is in. Because grants are more like a lottery now, you have to submit several grants each cycle to have a chance at surviving. What that means is (1) a LOT more work each cycle, which is wasted work because grants are not publications and vanish into the ether rather than contributing to the scientific literature, and (2) much more variability in outcome. The problem with variability in outcome is that if 10 people submit 10 grants each because there is a 1 in 10 chance of getting funded, each person doesn’t get one grant. Some get none and starve to death. Others get too much funding to manage reasonably. It’s also (3) a lot of wasted work reviewing all those grants. Which makes the reviews worse, which makes the grant more random, which….

  2. Yeah I agree–I submit NIH R01 grants and many reviews lately are randomly hostile and irrelevant.

    A most recent R01 that I submitted last year didn’t get scored (ie<50 percentile) even though it was a renewal of a successful project that I had tons of new, exciting data on. Sadly, the 2 reviews were both hack jobs–I'm guessing both were inexperienced and they seemed pretty confused overall about the whole process.

    One reviewer's main/only complaint about my application was (I am not kidding) "I didn't like how it was organized, Aim 3 should have been presented as Aim 1 instead because it is more interesting" No comments on the science (except I guess they thought at least part of it was interesting?) but they apparently gave me a bad score based only on their random thought about how they wanted it to be organized. I guess because they would have preferred to read page 6 before page 4…… Pretty Ridiculous.

    I think the NIH is resorting to using inexperienced reviewers because they are getting buried in grant applications. But I really wish program officers would push back a bit more and call out the bad reviewers….

  3. I wish NSF would go for the UK NERC review system, where ad hoc reviews are sent to the PI before a decision is made. The PI is given a chance to write a response, one page per reviewer. Then the PO acts like an editor. Then the whole issue of unfair criticisms and conflicting reviewer comments would be reduced in effect. Or at least the PI would feel like she made her best effort and had her say, and so rejection might be less devastating…

  4. NSF review process is quite irrational. US DOE (some directorates) and ARPA-E both send reviewer comments to the PI, and provide 2-3 pages (and 2-3 days!) for a rebuttal.

  5. a1b1, IME w/DOE, you do get the opportunity to respond, but only if you are close to getting funded (so not everyone is given the option). But yes, their reviews are always very high quality, from people in the field, detailed and thoughtful.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s