Effective grant proposals, Part 2: Feasibility

Today, the second part in my series on writing effective grant proposals. I’ve pointed out the importance of careful thought about what a grant proposal is for. In brief, the function of any grant proposal is to convince its readers of three things:

  • that the work you’re proposing is worth doing;
  • that the work you’re proposing can be done;
  • and that the work you’re proposing can be done by you.

Or (in order): novelty and significance; feasibility; and qualifications.

Having dealt with the first bullet, it’s now time to think about the second: feasibility. A funding agency will want return on its investment in the work, which means that they’ll want to be convinced that the work can actually be done.

There are quite a few pieces to establishing feasibility. Grants generally have very strict length limits, which may constrain the depth of your dive, but it’s wise to at least touch on all of these:

  • Are there standard, well-known methods for the work you propose to do? Are you working, for example, with a model system (like Drosophila or Arabadopsis development) so that you have proven tools to work with?
  • If not, then do you have reason to believe that methods that haven’t yet been applied to your proposed system or question will work? The gold standard for this one, of course, is pilot data – a small dataset from preliminary experiments or observations that establishes that the work can be done, because you’ve already done some of it. (Pilot data are extremely important to the culture of some granting organizations. It’s a running joke, for example, that the purpose of an NSF grant is not to do the work proposed, but rather to gather the pilot data for the next) But if you don’t have pilot data, you should still address this, perhaps drawing on similarity between systems, or previous research by you or others.
  • Once the data are in hand, can you outline an effective way to analyze them? Ideally, you’ll specify both a statistical test and the interpretation you’d give to each possible result of that test – with reference to the question your work will ask. (Note that grants to do theoretical work will be a bit different. They might not use statistical methods – although they might – but they’ll still involve interpreting results.)
  • Given the sample sizes you propose, will your statistical tests have sufficient statistical power to detect the kind of effect you’re interested in? This, in my experience, is a common gap in proposals. Over and over again, I’ve read a proposed number of field sites, samples, or individuals, and wondered whether there’s any reasonable probability that the work would end up with a clear answer to the question being asked – given the strength of the pattern we’d like to discover, and the amount of variability in the system
  • Can the work proposed be done on the budget you’re asking for? (A related issue, of course, is whether the budget you’re asking for is consistent with what the granting agency or program typically awards.)
  • Can the work proposed be done over the duration of the funding period? That’s not just a matter of the time it takes to run the experiments or take the observations; you may need time to hire and train personnel, order or build equipment, analyze data, and more. Your scheduling may need to take into account the nature of the study system (in biology, for example, field seasons and the growth rate or generation time of the organisms you’re studying). It may also need to consider the schedules of collaborators, the cumulative nature of the work (can experiment B be started before experiment A has been completed and its data analyzed?), and more. A common technique here is to provide a Gantt chart showing the proposed schedule of activities. I’m not a fan of Gantt charts, myself, but they at least provide a way to organize your thinking, and writing, around the scheduling issue.
  • Is the work legal and ethical? Sometimes this isn’t much of a concern, but plenty of other research can’t be done without attention to legal and ethics issues. For example, your proposed work might involve endangered species, human-subjects research, radioisotopes, controlled substances, or work in parks or other protected areas. In each case, the funding organization will want to know that the proposed work can proceed (perhaps with appropriate permitting).
  • If some part of the work turns out not to be possible, do you have a backup plan that will still yield publishable results?

I’m sure you can think of other dimensions to feasibility (please use the Replies!). If someone might buy your car, they’ll want to know that it will run. If a funding organization might buy fund your research, they’ll want to know that it will run, too. A test drive works for the car; for a grant proposal, you need to do a little more.

Are we done with effective grantwriting? Nope, sorry – this is just part 2, and there’s a lot more to come. In particular, the feasibility question bleeds into the next one: qualifications. It’s not enough to show that the work can be done; you need to show that it can be done by you.  More about that soon – see you then.

© Stephen Heard  March 29, 2022

Image: Possible, by geralt via Pixabay.com

4 thoughts on “Effective grant proposals, Part 2: Feasibility

  1. Pingback: Como elaborar um projeto de pesquisa – Sobrevivendo na Ciência

  2. Pingback: O que é a justificativa de um projeto? – Sobrevivendo na Ciência

  3. Pingback: Effective grant proposals, Part 3: Qualifications | Scientist Sees Squirrel

  4. Pingback: Effective grant proposals, Part 4: Who are you writing for? | Scientist Sees Squirrel

Comment on this post:

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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.