[Note: Most of the material from these writing workshop blog posts, plus a lot more never-blogged material, is now available in my book, The Writing Workshop: Write More, Write Better, Be Happier in Academia.]
Doing research costs money: Money to pay the research team, to purchase lab equipment and supplies, to travel to data collection sites, to compensate study participants, and so on. The higher you rise in the academic food chain, the less time you spend doing research, and the more time you spend selling research ideas to funders.
This is actually kind of a bummer if you enjoy doing research. I, for example, love listening to kids explain what they think about things. When I was a PhD student, I spent a lot of doing just that, because studying how kids think is a big part of my field of cognitive-developmental psychology. Now that I’m a professor, the data for my studies are mostly collected by undergraduate research assistants, who are directly supervised by PhD students, who are supervised by me. It’s no longer my job to listen to the kids, and I miss it. These days instead of collecting data, I spend a lot of time bringing in money to keep the lab going. Let’s talk about how to do that.
Fellowships, Grants and Awards
A fellowship is money that goes to an individual researcher. It usually includes a salary or living stipend, plus tuition and fees if the fellowship recipient is a student. Many fellowships also include a budget for research expenses, such as research-related travel and supplies. Fellowships that you get as a PhD student are called predoctoral; fellowships that you get after you finish your PhD (but before you get a permanent job) are called postdoctoral. There are also fellowships to help pay for faculty to go away from their home campus for some period of time (usually a few months to a year) to do research somewhere else. A few faculty fellowships exist in the sciences, but they are more common in the arts and humanities.
In the sciences, researchers who are in charge of labs apply for grants. Grants cover a much broader range of expenses than fellowships— usually some salary for the lead researcher (who is called the P.I., for ‘principal investigator’) but also salaries for paid employees such as lab managers, lab technicians, postdocs, graduate and undergraduate student researchers, and statistical consultants. Grants also pay for lab equipment, materials and supplies; for compensating human subjects (e.g., children who participate in our studies get a prize, like a slinky or a t-shirt); and whatever other research expenses the lab has. So a grant proposal is a much bigger, more complicated document than a fellowship application.
Award is a general term for money given out by some institutional body, such as an academic department or a funding agency. When a university department gives $500 to a graduate student to help pay for their travel to a conference, it might be called something like the ‘Exogeology Department Graduate Student Travel Award,’ or even ‘Prof. J.-L. Picard Award,’ which gives the student something snappy to put on their CV, in addition to $500. ‘Award’ is also the word that the big federal funding agencies in the U.S. use for the money they give out. So whether you get $500 from your own Department of Acarology, or $5,000,000 from the National Science Foundation Directorate for Zymology, either way it’s called an award.
Play the Odds
The first and most important thing to keep in mind about applying for funding is that it’s a numbers game. Remember Simonton’s Equal-Odds Rule? It says that the best way to win more fellowships or grants is to apply for more fellowships or grants.
Illustrating this point, Pier et al. (2018) asked each of 43 NIH reviewers to rank the same set of 25 grant applications (all of which had, in reality, been funded by the NIH.) What did they find?
“Results showed no agreement among reviewers regarding the quality of the applications in either their qualitative or quantitative evaluations."
No agreement! For every reviewer who ranked Application A as the best, there was another reviewer who ranked it worst. In the words of the authors,
“It appeared that the outcome of the grant review depended more on the reviewer to whom the grant was assigned than the research proposed in the grant.”
So for heaven’s sake, don’t take it personally when your grant or fellowship application is not funded. It’s no more personal than rolling dice, and a funded grant is double sixes. How do you roll more double sixes? You just keep rolling.
. . . But Play to Win
Do Your Homework
In an earlier post, I wrote that most PhD students start out thinking of themselves as students or employees, when they should think of themselves as entrepreneurs. Many early-career scientists have a similar misunderstanding about grants: They think of grants and fellowships the way some children think of Christmas presents— as rewards for good boys and girls.
Unfortunately, there is no Santa Claus. By which I mean, there is no one waiting in the wings to give you money just for being awesome. You get money in science the same way you get money in any other sector of the economy: By selling something that people want to buy. (In this case, research.) So instead of thinking of funders as Santa Claus, try thinking of yourself as a design-build contractor—a person who designs construction projects and also builds them.
Let’s imagine two design-build contractors with the same skill set: They both know how to build water parks. They’ve spent five years as apprentices at the same water-park building firm, and now they’re each ready to strike out on their own. Let’s call them Alphie and Betty.
Alphie embarks on his new career as an independent contractor by sitting down at a drafting table and designing the water park of his dreams. He decides on the theme Arctic Naval Operations of World War II. The park is divided into four areas named Operation Claymore, Operation Doppelschlag, Operation Gearbox II, and Battle of the North Cape. Alphie spends two months drawing up detailed designs for the park, which will cover 2720 square feet and cost $750,000 to build. He goes to a municipal contractors’ website and finds a list of cities looking to build water parks in the next couple of years. He sends his plan to all the cities, and then sits back and waits for the money to roll in.
Betty takes a different approach. She starts by searching the website for cities building waterparks. She searches each city’s website for information about their budget, timeline, space constraints, and any specific design criteria they have. She makes the following spreadsheet (sorted by city name).
It’s already October 1, so Betty eliminates the Tulsa project— there’s no way she’ll have enough time to prepare a proposal in the next 2 weeks. She divides the remaining projects into two groups (big and small), and sorts them by due date.
The three big projects have similar sizes and budgets, so she decides to create a basic design for a big park and customize it for each of the three cities. She will also design a small park and submit slightly different versions of it to Montgomery and Ottawa. She decides not to submit a design to Quincy, because that project is so much smaller than the others that she would have to create a completely separate design.
Betty adds the projects to her IDP. She will spend the next couple of months designing the big project for Mexico City to be submitted December 1. Then she’ll switch to working on the small-project design for Montgomery, to be submitted Feb. 4. After that, she’ll revise her Mexico City design to make two new designs for big parks: One with a pirate theme for Rock Hill and another with an environmental theme for Escondido, both due around the end of June. Finally, she will revise the smaller design to create the proposal for Ottawa by the end of November.
Compare our two contractors. Alphie is thinking about what he wants to build, and he’s hoping someone will fund it. Betty is thinking not only about what she wants, but also about what the funders want. She’s bringing her skills and expertise to a project that matches the funders’ priorities, space constraints, budget and timeline.
Don’t be an Alphie; be a Betty. Before you start designing your project, take the time to research funders and find out what they want. What kind of questions are they interested in? How long (time-wise) and how big (budget-wise) are the projects they fund? None of this is a secret— most funders openly announce their funding priorities, and their websites include abstracts of projects they’ve funded before. Read the priorities and take them seriously. Read the abstracts. If you have any questions at all, call or email the program officers. At some agencies (including the NIH) program officers are a huge untapped resource—they will give you a lot of great advice if you take the time to ask them.
When you’ve gathered all the relevant information, draft a work timeline that will let you re-submit versions of the same grant to multiple agencies, customizing it for each one. (If the same work is funded by more than one agency, you’ll have to negotiate with the program officers about what to do next, but that’s the best problem to have.) In general, start as early as you can—preferably several months before the submission deadline. This will give you time to draft the whole grant, get feedback from several colleagues (sometimes a program officer will be willing to look at your ‘specific aims’ or other brief outline of the proposal), revise and improve the grant based on the feedback you got, and still submit it by the deadline.
Finally—and this should be obvious, but people don’t always do it, so I’ll say it anyway— Follow the grant or fellowship submission instructions to the letter. If the instructions say to put the proposal narrative in 14-point, purple, italic, Comic Sans font, triple-spaced, right-justified, with 2.25” hanging indents, happy-face emoticons instead of page numbers and every 17th word in Japanese, then do that. The first thing funders typically do when they receive an application is check to make sure it meets all the technical requirements (formatting, length, all the documents complete, etc.) Applications that don’t conform are tossed immediately. So follow the damn directions.
Different fellowship and grant programs ask for different information. For example, the NSF Graduate Research Fellowship Program asks for a lot of information about you, the applicant: They want to know about your background, your future goals, and your research plan for the fellowship period. They typically require letters of reference and transcripts.
Alphie embarks on his new career as an independent contractor by sitting down at a drafting table and designing the water park of his dreams. He decides on the theme Arctic Naval Operations of World War II. The park is divided into four areas named Operation Claymore, Operation Doppelschlag, Operation Gearbox II, and Battle of the North Cape. Alphie spends two months drawing up detailed designs for the park, which will cover 2720 square feet and cost $750,000 to build. He goes to a municipal contractors’ website and finds a list of cities looking to build water parks in the next couple of years. He sends his plan to all the cities, and then sits back and waits for the money to roll in.
Betty takes a different approach. She starts by searching the website for cities building waterparks. She searches each city’s website for information about their budget, timeline, space constraints, and any specific design criteria they have. She makes the following spreadsheet (sorted by city name).
City
|
Due
date
|
Sq ft
|
Budget
|
Notes
|
Escondido, CA
|
July 1, 2019
|
7,500
|
$2,500,000
|
Environment
or conservation theme
|
Mexico City
|
Dec. 1, 2018
|
9,000
|
$2,475,000
|
|
Montgomery, AL
|
Feb. 4, 2019
|
3,000
|
$825,000
|
|
Ottawa, ON ...........................
|
Nov. 30, 2019
|
2,300
|
$644,000
|
|
Quincy, MA
|
April 30, 2019
|
1,500
|
$300,000
|
Separate areas for kids 2-5 and 6-12
|
Rock Hill, SC
|
June 30, 2019
|
10,000
|
$2,500,000
|
Pirate theme
|
BIG
PROJECTS
|
Proposal
due date
|
Park
size (square feet)
|
Budget
|
Design notes or additional
constraints
|
Mexico City
|
Dec. 1, 2018
|
9,000
|
$2,475,000
|
|
Rock Hill, SC
|
June 30, 2019
|
10,000
|
$2,500,000
|
Pirate
theme
|
Escondido, CA
|
July 1, 2019
|
7,500
|
$2,500,000
|
Environment
or conservation theme
|
SMALL
PROJECTS
|
Proposal
due date
|
Park
size (ft2)
|
Budget
|
Design notes or additional
constraints
|
Montgomery, AL
|
Feb. 4, 2019
|
3,200
|
$960,000
|
|
Ottawa, ON
|
Nov. 30, 2019
|
3,500
|
$875,000
|
The three big projects have similar sizes and budgets, so she decides to create a basic design for a big park and customize it for each of the three cities. She will also design a small park and submit slightly different versions of it to Montgomery and Ottawa. She decides not to submit a design to Quincy, because that project is so much smaller than the others that she would have to create a completely separate design.
Betty adds the projects to her IDP. She will spend the next couple of months designing the big project for Mexico City to be submitted December 1. Then she’ll switch to working on the small-project design for Montgomery, to be submitted Feb. 4. After that, she’ll revise her Mexico City design to make two new designs for big parks: One with a pirate theme for Rock Hill and another with an environmental theme for Escondido, both due around the end of June. Finally, she will revise the smaller design to create the proposal for Ottawa by the end of November.
Compare our two contractors. Alphie is thinking about what he wants to build, and he’s hoping someone will fund it. Betty is thinking not only about what she wants, but also about what the funders want. She’s bringing her skills and expertise to a project that matches the funders’ priorities, space constraints, budget and timeline.
Don’t be an Alphie; be a Betty. Before you start designing your project, take the time to research funders and find out what they want. What kind of questions are they interested in? How long (time-wise) and how big (budget-wise) are the projects they fund? None of this is a secret— most funders openly announce their funding priorities, and their websites include abstracts of projects they’ve funded before. Read the priorities and take them seriously. Read the abstracts. If you have any questions at all, call or email the program officers. At some agencies (including the NIH) program officers are a huge untapped resource—they will give you a lot of great advice if you take the time to ask them.
When you’ve gathered all the relevant information, draft a work timeline that will let you re-submit versions of the same grant to multiple agencies, customizing it for each one. (If the same work is funded by more than one agency, you’ll have to negotiate with the program officers about what to do next, but that’s the best problem to have.) In general, start as early as you can—preferably several months before the submission deadline. This will give you time to draft the whole grant, get feedback from several colleagues (sometimes a program officer will be willing to look at your ‘specific aims’ or other brief outline of the proposal), revise and improve the grant based on the feedback you got, and still submit it by the deadline.
Finally—and this should be obvious, but people don’t always do it, so I’ll say it anyway— Follow the grant or fellowship submission instructions to the letter. If the instructions say to put the proposal narrative in 14-point, purple, italic, Comic Sans font, triple-spaced, right-justified, with 2.25” hanging indents, happy-face emoticons instead of page numbers and every 17th word in Japanese, then do that. The first thing funders typically do when they receive an application is check to make sure it meets all the technical requirements (formatting, length, all the documents complete, etc.) Applications that don’t conform are tossed immediately. So follow the damn directions.
Information in a Grant Proposal
Different fellowship and grant programs ask for different information. For example, the NSF Graduate Research Fellowship Program asks for a lot of information about you, the applicant: They want to know about your background, your future goals, and your research plan for the fellowship period. They typically require letters of reference and transcripts.
NIH predoctoral and postdoctoral fellowships also want to know about you, but they ask for more information than NSF does about the training environment: Who will your mentors be? What kind of lab/department/scientific community will you be working in? What is the plan for your training?
When you move on to writing grant proposals, the emphasis is less on you as a researcher (although you still have to show that you are competent), and more about the research you are proposing. For example, a grant proposal to NIH includes a one-page statement of ‘specific aims’ laying out the goals of the project, and a research strategy section explaining the logic of the approach.
Because the information required is different for each funder and each program, the easiest and most useful thing to do is get a copy of a submitted proposal (even better if it was actually funded) to the program where you are applying, and use it as a template.
But no matter where you are applying, you will need to persuade reviewers of several things: (1) There is a problem; (2) You have a plan for research that will address the problem; and (3) You and your team can do this research.
1. There is a problem.
Nobody is going to give you money unless your research addresses some problem they care about. (You can find out which problems funders care about by doing your homework, as described earlier.) If a funder has already identified a problem (e.g., childhood obesity in the U.S.) then you just have to show that your work addresses it. Sometimes this is easy (e.g., ‘85% of children eat lunch at school; we’re going to study the nutrition in school lunch programs’); other times it requires a few steps (e.g., ‘Over the past twenty years, schools have increased homework by 65%. We hypothesize that increases in homework cause increases in obesity by raising children’s cortisol levels and reducing time spent in both outdoor free play and sleep.’’)
For funders that support research on a wide range of topics (e.g., NIH, NSF, other government agencies, the military, etc.) then you will have to identify the problem and tell them why it matters. If you are curing cancer or saving the environment, this is easy. But if you do basic research, where the goal is just to learn about something, and there are no immediate or obvious applications, then you’ll have to convince reviewers that your work is worth the investment.
This is always a challenge. On the positive side, your grant application will be reviewed by other scientists in the same field as you, and they value this work more than the average person does. On the negative side, funders don’t want to spend their money on research that won’t make any difference in the world, so they ask reviewers to pay a lot of attention to the likely impact of the work. So if you are embarking on a career in basic research, start thinking about what real-world problems you can connect your work to.
2. You have a plan for research to address the problem.
In an earlier post, I talked about the parts of a scientific paper. Step 1 (above) is like the very beginning of the paper—the opening and big question—where you say what the research is about and why it’s important. Step 2 is like the rest of the introduction and method section, where you trace a logical path from the big problem to some smaller, specific questions that you can actually answer. Then, importantly, you lay out a plan for how you will answer them.
It’s a bit like writing a preregistration or Stage 1 of a registered report, in that you are talking about research you haven’t done yet. The main difference is that the description of methods in a grant or fellowship proposal is shorter and less detailed than in a preregistration or registered report. That’s because grant proposals have strict word limits or page limits, and authors are usually struggling to cram a lot of information into a small space.
In order for your proposal to get a high score, the reviewers must understand your plan, like it, and believe that you can carry it out. Most proposals will include the following information in some form, with different labels and in different orders, as determined by the grant format.
- A rationale for why this approach is the right way to address the big problem or question. This usually requires a brief literature review, similar to one in the introduction to a scientific paper but often shorter because of the page limits on a grant. A grant is also different from a paper in that grants require you to make a stronger case for why the work is ‘innovative’ or ‘impactful.’ A grant proposal has to sell the research, whereas a paper merely has to report it.
- A method section explaining what you are going to measure and how. This is similar to the method section of a scientific paper but less detailed, again because space is limited. In a grant, the research plan is the real meat of the proposal, and it’s what takes up most of the space in the application.
- A budget explaining how much this work is going to cost. For a big grant, this will include things like salaries, equipment, travel, participant compensation, etc. For a fellowship, the award is usually a fixed amount of money and you may just have to write a paragraph or two saying how you will use it. For a big grant, the budget is a separate spreadsheet that you prepare by working together with an administrative person at your university. (Most grants are technically awarded to the university, not to you.) If you have the opportunity to work with a skilled and experienced administrator to prepare your grant, count yourself lucky and treat that person with the greatest courtesy and respect. They can make your job a whole lot easier and your application much stronger if you just listen to what they have to say.
- A budget justification. This is a document that goes along with the budget for a grant. It explains what each of the items on the budget is and how it fits into the work plan. Again, your administrative support person should be able to help you with this. They might even be able to give you a boilerplate budget justification and just highlight the text that you need to change for this particular grant.
- A timeline sketching out when everything will get done. (This doesn’t take up much space-- maybe a couple of lines at the bottom of a page.)
- Evidence that you have the necessary facilities or environment to do the work. It often consists of just a couple of paragraphs saying that you have an office and a computer and a printer and whatever else will be needed. For example, if your research requires an fMRI machine and a scanning electron microscope, you will have to confirm that you have access to those things.
- A dissemination plan explaining how you plan to share the results of your work with the scientific community, the public, policy makers, or whoever you think should know about it. (You do this by publishing articles, giving talks at conferences, etc.) Publicly-funded agencies are especially interested in this, because their mission is not only to fund research, but also to get the results out there in the world where people can make use of them.
3. You (and your team) can do this research.
Reviewers want to be sure that you will be able to carry out the proposed work. If you are an established scientist with a long history of publishing in this area, then your track record will speak for itself. But let’s assume you don’t have that kind of record built up yet. How do you present yourself in the best light possible?
Highlight your strengths
Just like when you send out resumes for jobs, you want to present the best picture of your qualifications and accomplishments. You are probably too familiar with the information on your own CV to see it objectively, so make sure to draft your biosketch (or whatever abbreviated CV you have to submit for the grant) early, and show it to someone such as your advisor or a senior colleague for feedback. Because other people can look at your information with fresh eyes, they can often suggest improvements or areas for clarification that wouldn’t occur to you.
Put together a team
Science is a team sport; don't try to play every position yourself. Focus instead on assembling a great team. If you are applying for a fellowship where you will get some kind of training, your team will be yourself and the people who mentor you. They should have a strong track record of research themselves, and a strong record of training graduate students or postdocs. You should always work with your proposed advisor(s) to write a fellowship application. (If you are considering working with someone, and you find out that they don’t have time to give you feedback on drafts of your application application, it means they probably don’t have time to mentor you either. In that case, consider working with someone else.)
For a postdoctoral training fellowship, reviewers also need to hear that you will learn something new in the postdoc—something you didn’t get from your PhD training alone. It’s hard to get a postdoctoral fellowship if you propose to stay in the same lab where you did your PhD; it’s easier if you go somewhere else and gain new technical skills, or start a new line of research that complements your existing work.
If you have your own lab and you are applying for grants, then it’s all about the research team. You can’t be an expert on everything, and you shouldn’t try to be. When I was a new assistant professor, I didn’t understand this. I remember having a conversation with an NIH program officer who said that I should include a budget for statistical consulting in my grant proposal. I was offended. Statistical consulting? I fumed. What for? Do the reviewers think I'm incompetent to analyze my own data?
I was offended by the idea of a consultant because I wanted to show that I could do everything myself. The NIH program officer was nudging me to adopt a more mature view. He was implicitly saying, that it wasn't about me. The reviewers don’t care whether I personally do the analyses; they just want to know that the analyses will be done right.
It can also be useful to put together an advisory board of experts in the area—people who won’t actually work on your grant project but are willing to look over your study proposal and give you feedback on it; people you can call when you have questions about how to do the work. I was once told that you shouldn’t invite anyone to be on your advisory board unless they are willing to read a copy of your grant proposal ahead of time and give you comments on it. That’s not a bad rule. People will often say yes to being on your advisory board even if they don’t really have time to spend advising you on the project. Asking them to give feedback on a draft of the proposal is a good way to make sure they are serious about making a commitment to help you.
Some funding agencies ask you to include a letters of cooperation from each person on your advisory board, and you should also include a plan for how the board will advise you. For example, you might put money in the budget to bring all the board members to your campus for a meeting at the beginning of the project, and also plan to hold meetings by teleconference every year while the project is being carried out.
Community partners
If your work requires the help or cooperation of people or institutions outside your university, you should probably include letters of cooperation from them too. For example, in my lab we collect data at preschools and museums. So we include letters of cooperation from those preschools and museums in our grant proposals. The letters are on official letterhead and signed by the director of the school/museum (or other responsible person), and basically say, “We exist and we have a lot of children here, and we are happy to allow the Sarnecka lab to recruit participants and collect data for the proposed project.”
Read the Rest of This Book
Most of what I have to say about writing grant and fellowship proposals is stuff that I'm already saying elsewhere in this book. Grant writing is just writing. For example, planning your writing is even more crucial for grants than articles, because grant programs typically have hard deadlines. If you fall behind on writing a paper, it usually doesn’t matter-- you can always submit it later than you planned. But if you miss the submission deadline for most grant and fellowship proposals, you’re out of luck.
A rejection collection is also very helpful for motivating grant writing. It can be hard to make yourself work on grant applications because you know that most of them don’t get funded, and you start to wonder if you are wasting your time. A rejection collection helps you stay focused on the part of the process that is within your control: Applying. And because it’s natural to resist writing grants given that most of them are rejected, the strategies for managing resistance are also useful.
To get grants, you must describe your work at the disciplinary or public level, as discussed in this post. There, I described sitting on a grant panel and seeing a grant get rejected not because of the proposed work,, but because the authors were unable to describe it in a clear way for a broad audience. Many grants, particularly from early-career scientists, are rejected because the authors get bogged down in subfield-level questions and lab-level methodological details, and don’t sell the reviewers on the big ideas.
Of course there is a balance to strike. When you are a new investigator, you may need to provide a more detailed description of your data collection and analysis plans to reassure reviewers that you know what you’re doing. Once you have a strong publication record, they may be more willing to give you the benefit of the doubt, so you can devote more space to the big ideas and less to explaining exactly how the studies will be done. In general, you want to give enough detail to assure the reviewers of your competence, but not so much that the proposal is boring or confusing to people outside your immediate subfield. (This is one place where getting feedback from experienced colleagues can be especially helpful.)
Later posts in this series (stay tuned) describe principles for writing clearly about complicated material, and for making good figures. Those principles are even more important for grant writing than for article writing, because grant reviewers are usually under more pressure than paper reviewers. For example, the reviewers of a journal article may have just that one paper to read; most grant reviewers are tackling a whole pile of grants. If an article reviewer is too busy or tired to read your paper tonight, they can read it tomorrow or next week; grant reviewers usually have a firm deadline.
Assume that the grant reviewers are reading your proposal on an airplane, on their way to Washington D.C. for the panel meeting. Assume that they are getting motion sick from reading during a flight, but they have to keep reading because all their review scores were supposed to be uploaded the day before yesterday, and they would have uploaded them on time, except that their kid got an ear infection and spiked a fever so, they spent the afternoon at the pediatrician’s office, and then they had to drive around to three different pharmacies to get the prescription filled, and then they had to find someone else to drive their other kid to soccer practice so they could say home with the sick one, and also finish their lecture for the next day that wasn’t totally written yet, and what with one thing and another, they just didn’t have time to read the proposals until now, okay? So stop with the nagging emails, NIH Scientific Review Officer! I am speaking purely hypothetically, of course.
My point is, you never know what calamities have already befallen your poor grant reviewers by the time they read your proposal. The last thing you want to do is add to their misery by making them read a boring, dense, convoluted grant. So follow the instructions in the later posts in this series to make your grant proposals clear, well organized, and pleasant to read. Your reviewers will thank you, and their good feelings will translate into higher scores.
No comments:
Post a Comment