Pretty much every successful proposal starts with a variant of the following structure:
1. Topic X is important and interesting.
2. But.
3. This is how we will address “But.”
The rest of the proposal reiterates those three points with enough detail to make it believable.
In a short proposal, the structure fills a paragraph. In a long proposal, it’s three paragraphs, and shouldn’t go past the first page.
The abstract is a 1 paragraph version of the same structure, with the addition of a closing rah rah rah sentence.
If you can’t bludgeon your introduction into this form, you might want to step back and regroup.
November 2nd, 2009 at 2:30 pm
Smoke machine and lots of sand to make mirrors.
November 2nd, 2009 at 2:34 pm
And don’t forget the ponies.
November 2nd, 2009 at 3:35 pm
Unfortunately, these steps are necessary, but not sufficient conditions for a successful proposal. The smoke machines, mirrors and ponies help.
November 2nd, 2009 at 3:47 pm
I must admit that I find it a little funny that people don’t realise that proposal writing is a game, and knowing the rules of the game is a good way to get started.
I often people write the most impenetrable tosh and assume that their inner brilliance will shine through. When ultimately unsuccessful, and given advice on how to play the game, they ignore it and submit a virtually identical piece of impenetrable tosh to be rejected again.
November 2nd, 2009 at 4:50 pm
As you’ve not included a “Broader Impact” section, I’m afraid we’ll have to return your proposal unread (assuming this is apropos of next friday’s deadline). Better luck next year.
November 2nd, 2009 at 6:07 pm
Any chance we will see a more fleshed out article on NSF fellowships? This is the year to apply and I figure I’ll need every edge.
-A tired grad
November 2nd, 2009 at 7:20 pm
I must strongly disagree with your proposal. The real approach is as follows:
1) The person who controls the funding wants an answer to X
2) We will do Y to answer X
November 2nd, 2009 at 9:17 pm
I would definitely second the request from Nathan – any tips on the NSF fellowship application process? Although, I suppose this would technically make it “solicited advice”, but that would be a great addition to the blogging repertoire…
November 2nd, 2009 at 10:33 pm
Nathan and Jenna — Well, my advice is pretty much as above. I’ve been reading proposal after proposal this week (NSF, Hubble, HST, etc) and almost all my substantive revisions keep falling into making the proposal fit the above. Thus, I thought I’d share!
But, to get more specific, a common tendency for junior scientists (such as those writing NSF graduate and postdoc proposals) is to start out a proposal with a “book report”, writing down everything they know about Topic X. The writer is trying to prove that they know their stuff, but the reader probably already knows a fair bit of it, and has no idea yet about why this information is relevant. You’ve got to move quickly to give the reader the information needed to frame what you’re trying to do. You can work the book report stuff in later, if it’s really needed to understand the potential impact of your project.
The other key is to leave absolutely no doubt in the reader’s mind about what you’re actually going to do, and how you’re going to accomplish it. You have to hit this immediately in the first few paragraphs, preferably in italics (”We will make the first measurements of the refractive index of bananas, providing critical constraints needed for the future of fruit-based computing.”). Then, by the end of the proposal, you should have given enough supporting details that it should look like the project is a done deal, if only someone would just give you resources.
Finally, you can never be too clear. Reading the proposal should feel effortless. You accomplish this through taking a pedagogical tone of voice early on, through avoidance of acronyms, and through relentless editing by you and everyone who owes you a favor. If your writing is unclear, it will be interpreted as your thinking as not yet being clear. And that’s not good.
I’ve never served on an NSF graduate or postdoctoral fellowship committee, so I can’t give specific advice for those particular programs. However, if you follow the above, you’re probably going to survive triage (i.e. top 30% of the proposals — truly, 2/3 of the proposals are sunk by not being clear in their intent, importance, or feasibility), giving you a fighting chance at making the final cut.
November 3rd, 2009 at 7:09 am
It is inexpressibly depressing that scientists have to submit to this kind of humiliation. Is that all we are? Glorified beggars?
Here’s how it should work.
“Here are my papers. If you take the time to read them, you will see that I am doing interesting stuff which nobody else is doing. Send me X dollars at the following address. yours sincerely….”
November 3rd, 2009 at 7:59 am
Astronomy has it especially bad. Does any other scientific field have to submit one proposal to get the means to acquire their data and a second proposal to get the money to analyze the data, which are submitted to different sources and often result in one of the two being accepted and the other rejected?
November 3rd, 2009 at 8:12 am
Timon — I totally disagree. A proposal is essentially an experimental design. I should be required to show that I designed a really good experiment that is actually likely to accomplish something useful. I shouldn’t just get money because I wrote a few good papers over the last year (though, of course. doing so does increase the reader’s faith that my assessment of whether an experiment will work is probably valid).
On the other hand, not every single scientific advance was planned, and I do support having small pools of money to seed the kinds of creative project that unfettered smart people can come up with when left alone. But, I think such funding should be a small fraction of the total, and probably biased heavily towards theory.
November 3rd, 2009 at 8:46 am
““Here are my papers. If you take the time to read them, you will see that I am doing interesting stuff which nobody else is doing. Send me X dollars at the following address. yours sincerely….””
In the words of Groucho Marx: “These are my principles. If you don’t like them, I have others.”
November 3rd, 2009 at 11:46 am
[...] http://blogs.discovermagazine.com/cosmicvariance/2009/11/02/unsolicited-advice-x-how-to-frame-a-win... a few seconds ago from xmpp [...]
November 3rd, 2009 at 5:33 pm
A proposal is essentially an experimental design
Except for those of us who are theorists, where Timon’s approach really seems much more reasonable. (I hear the mathematicians especially have fun with this. “I am going to prove the following theorem” doesn’t work so well.)
November 3rd, 2009 at 5:52 pm
> Except for those of us who are theorists, where Timon’s approach really seems much more reasonable. (I hear the mathematicians especially have fun with this. “I am going to prove the following theorem” doesn’t work so well.)
I’m a theorist and I agree with Julianne on this one – you still have to frame the question you are addressing and how you will answer it.
November 3rd, 2009 at 7:17 pm
Astronomy funding IS weird, with a big disconnect between ground-based and space-based: get time on HST, or Sptizer etc, you get funding also, get time on any ground-based telescope (I’m sure I’ll be told if there are exceptions I don’t know about) and THEN you write the other proposal to get the funds to pay (badly) the grad student or post-doc to do the work for you (yes, a tongue is somewhat inserted in-cheek, but only somewhat) or to get 2/9 of your pay to do the work you’re already being paid to do……(There are of course REAL and legitimate other costs such as travel, computers, page charges…)
November 3rd, 2009 at 9:11 pm
@ 6 nathan:
The best way to win an NSF fellowship (or any other fellowship, really) is to talk with folks who have successfully applied for one in the past. Many will be willing to share their successful proposals with you if you ask nicely. I learned a tremendous amount about what makes a good proposal from *reading* good proposals. Talk to the postdocs at your institution, or get in touch with people you know who are a bit older and bit further along the career path.
(This is true, incidentally, at all stages of the game: whether applying to grad school, postdocs, faculty jobs, etc.! )
To give a slight variation on what Julianne has already said, one of the things I was told when I asked for advice from my own elders was that “A good job application is a three-legged stool”. You’ve got to convince your readers that
1) Topic X is interesting, but.
2) Here is how we’ll make progress on that ‘but’.
3) Here is why I, in particular, am the right person to make that progress.
In other words, when you’re applying for a job or fellowship, you need to emphasize that third leg a little bit more than you would for another kind of proposal like for telescope time.
Another piece of advice I would give for postdoc fellowships is to think about the timeline. You’re not just proposing for something you’re going to do next semester, or even next year. You’re generally proposing for three full years of funding! Show your readers that you’re aware of this. Does the size of the project you’re proposing fit properly in a three year period? What parts will you do in year 1, or year 2? What things need to happen along the way for you to progress? How many papers do you think will come out of the project? This doesn’t need to be a long section – but a paragraph or two can show you’ve thought about these matters, thereby demonstrating some scientific maturity.
November 3rd, 2009 at 10:22 pm
And to follow-up on Marshall’s excellent advice, also make sure that the project won’t obviously spill over 3 years either. I’ve read proposals where the fellowship applicant has mapped out a plan that would take him/her through two postdocs and tenure, while claiming they would do it all during their fellowship.
November 3rd, 2009 at 11:31 pm
Dear Mr. Newton.
Your proposal was well-written in Latin. Your Geometrical diagrams are elegant. Your Biblical quotations show deep understanding. However, we don’t see how sitting under an apple tree in Lincolnshire can do anything for understanding the orbit of the Moon, hence we suggest that you re-direct your proposal to some institution devoted to science of apple growing, which we understand is called pomology.
Good luck with your funding requests there.
Sincerely yours,
Royal Kinematics Research Administration.
November 4th, 2009 at 8:34 am
On 18,19, this feels more applicable to experimental proposals than theory ones. Writing theory proposals for fellowships has always made me feel a little dirty. A fellowship is typically for a time period from (3+1)-(5+1) years from when it is written. However the half-life of most theory topics is 6-12 months. So one has to write this nice story about goals to be achieved in years 1,2,3,4 and 5 all the time knowing that you have no detailed idea what you will be working on in 2 years time let alone five years time.
November 4th, 2009 at 8:39 am
Re Theorists and research proposals — I agree that these are hard to write, partly because if I knew what I was going to be doing two or three years from now, I would want to be doing it today.
But on the other hand, some theorists DO write much better proposals than others (which is one of the reasons why it is interesting to sit on grant panels — at least the first time, so as to see what a good, fundable proposal looks like), so this problem is not intractable. Personally, in a multi-year theoretical proposal, I am looking for evidence of “taste” — both in the choice of problem, and in the proposed plan to tackle it.
But it is worth reflecting that grant proposals are typically requests for fairly significant sums of money (at least on the personal scale) — often public money. It is not unreasonable that we should describe our plans for it before the check gets written.
November 4th, 2009 at 9:34 am
Thanks Marshall P, I’m actually referring to the Graduate Fellowships due next week, though. I believe he postdoctoral fellowships were due a week or two ago.
November 4th, 2009 at 3:19 pm
Nathan – the same strategies apply, though! Everything I just said about how I successfully applied for a postdoc fellowship I could equally have said about how I obtained a graduate fellowship (from NASA in my case, not the NSF – but using the same principles and organization, pretty much). Good luck!
November 4th, 2009 at 11:35 pm
OK, Timon, I’m going to give it a try.
Here are my papers.
http://arxiv.org/find/quant-ph/1/au:+barnum/0/1/0/all/0/1
If you take the time to read them, you will see that I am doing interesting stuff which nobody else is doing. Send me any amount of money larger than 10 dollars at the following address. If you are a wealthy individual philanthropist, or a for-profit or not-for-profit corporation, or a government funding agency, make it more than $50,000. Canadian or US is OK; I’m not picky.
Yours sincerely,
Howard Barnum
Perimeter Institute for Theoretical Physics
31 Caroline Street N, Waterloo, Ontario,
Canada N2L 2Y5
November 5th, 2009 at 7:42 pm
OK, Julianne, I’m going to try your advice.