Providing Meaningful Feedback on Grant Reviews
Proposal review is similar to AND different from manuscript review. Proposal review is almost entirely evaluative, almost. Different from manuscript reviewer, your job as a grant reviewer is not only to serve the agency, but also to help the PI ensure that their grant is as strong as possible. We don’t review proposals on behalf of their authors, of course — we are acting on behalf of the funding agency — but it is our responsibility to provide meaningful feedback to help the PI improve their proposal for the next round of funding announcements. That ultimately serves the PI AND the community, because we want agencies to make good decisions. Our job is to tell the agency whether they should fund the project AND to assist the PI with how to make their work better and point the way toward success. Remember: the PIs will see your review, but they are not your primary audience—the review panel and program officers are. Keep this in mind as you evaluate and write.
Meaningful feedback benefits everyone!
- It saves time in the future. If applicants were close to getting funded, they’ll likely resubmit in an upcoming grant round. If they were not close to being funded, then honestly (but gently!) letting them know will save them and you from repeating the process and assessing similarly flawed applications in future grant rounds.
- It increases capacity. Feedback helps applicants learn and thereby increases their capacity, in addition to increasing the granting organization’s (by meeting the organization’s goal to provide funding that advances scholarship) — and eventually the community’s (by providing useful data from the results of the study)— capacity.
- It helps you carry out your mission. Providing useful feedback to applicants will help provide an array of high-quality applications for future rounds, thereby supporting your organization’s mission, and ultimately improving health outcomes
Alignment of Activities with Target Program Outcomes
Most granting agencies have pre-established criteria that should guide your review. Familiarize yourself with the application and those criteria. The project must align with the stated program objectives from the organization. If this is not the case, this should be clearly communicated with the applicant so that they do not repeat the application process with another misaligned submission in the future.
Of key concern is how COMPELLING the proposal is, in terms of the issue the study will address. Is the problem important? Will the proposed work contribute in a substantial way? Is it clearly cutting-edge science, and will this work help move the needle substantively (not incrementally)? There must be a demonstrated need for the project and some anticipated community benefit.
Has the case been made for why this PI/team are perfect to lead this work? Consider their scientific and methodological aptitude, as well as their previous work. Does the applicant have the capacity to successfully carry out the project as described in the application?
If the above points are all favorable, then it’s time to focus on the methods in a detailed manner. If not, then the PI will need to go back and rework their question. The approach will change as the question changes.
Formative vs. Detailed Feedback
Some applicants need substantial formative feedback, and others are ready for detailed feedback. I normally don’t give lots of fine-grained feedback on a proposal that needs substantive shaping through formative feedback. So, decide which is called for and provide the appropriate type of feedback.
To Restate or Not to Restate
Restating is to say (something) again or in a different way. Don’t restate what the applicant said! It takes up space and provides no value to the applicant. Instead, provide detailed and clear statements about what your specific concerns are. For example, “There are design flaws with this proposal” vs. “An assumption is made about XY that is not supported by adequate data.” The second statement provides clear direction to the applicant and informs them that the supporting data was not included or does not exist.