Research Grants at NIH and NSF

The National Institutes of Health and the National Science Foundation are the two largest funders of U.S. academic research, and each runs a different machine. NIH sorts awards by three-character activity code and reviews them in study sections. NSF sorts awards by directorate and scores every proposal on two criteria.

Current figures — verified 2026-08-11

Item	Value	Source
NIH Institutes and Centers	27	NIH
NIH research project grant success rate, FY2025	13.0% — 8,161 competing awards from 62,592 applications	NIH Extramural Nexus
NIH average research project grant, FY2025	$675,426	NIH Extramural Nexus
NIH R21 ceiling	$275,000 total direct costs over two years; no more than $200,000 in any year	NIH R21
NIH R03 ceiling	$50,000 direct costs per year, up to two years	NIH R03
NIH modular budget threshold	$250,000 or less in direct costs per year	NIH Grants Policy Statement 2.3.7
NIH per-investigator application cap	six applications per calendar year (T, R25, R13 excepted)	NIH Grants Policy Statement 2.3.7
NIH resubmission rule	one resubmission, within 37 months of the original	NIH submission policies
NSF CAREER limits	one proposal per annual competition; three competitions per PI	NSF 22-586
NSF decision target	six months from submission	NSF

These figures change. Verify against the linked source before relying on them. Report an outdated figure

Key takeaways

• NIH names awards by activity code; NSF names them by program and directorate.
• NIH runs two-stage review: study section for merit, council for priority.
• NSF scores every proposal on Intellectual Merit and Broader Impacts equally.
• Broader Impacts is judged on the same five elements as the research itself.
• Both agencies publish the governing policy document; read it before writing.

What are NIH and NSF research grants?

NIH and NSF research grants are competitively awarded federal financial assistance for investigator-proposed research, made to institutions rather than to individuals, and selected by peer review conducted largely by working scientists outside the government. Both agencies award grants and cooperative agreements, not procurement contracts, for the bulk of their extramural research. The funding tracks hub maps where each track sits.

The two agencies split the scientific field. NIH funds biomedical, behavioral, and clinical research through 27 Institutes and Centers, “each with a specific research agenda, often focusing on particular diseases or body systems” (NIH). NSF funds non-medical science, engineering, mathematics, and STEM education through directorates including Biological Sciences, Engineering, Geosciences, Mathematical and Physical Sciences, Computer and Information Science and Engineering, Social Behavioral and Economic Sciences, STEM Education, and Technology Innovation and Partnerships (NSF).

Vocabulary differs more than substance. An NIH award is identified by activity code and institute; an NSF award is identified by program and division. Both are assistance instruments under the same statutory framework that separates grants from procurement, covered in grants vs contracts vs cooperative agreements.

What is an NIH activity code, and which ones matter?

An NIH activity code is a three-character identifier that defines what a funding mechanism supports, how long it runs, and what it may cost. The first character signals the family: R for research, K for career development, F for fellowship, T for institutional training, P for program and center awards, U for cooperative agreements (NIH activity codes).

The research codes that carry most of the volume are narrower than they look:

• R01 — Research Project Grant. NIH’s flagship, supporting “a discrete, specified, circumscribed project to be performed by the named investigator(s)” (NIH R01). Renewable, typically three to five years, generally no fixed budget cap.
• R21 — Exploratory/Developmental Grant. Two-year ceiling, capped total direct costs, no preliminary data required, and it “cannot be renewed” (NIH R21).
• R03 — Small Research Grant. Pilot and feasibility work, secondary data analysis, methodology development. Up to two years, non-renewable, and a doctoral student may not use one for dissertation research (NIH R03).
• R15 — Research Enhancement Award. Small-scale projects at institutions “that have not been major recipients of NIH support,” with an undergraduate-focused AREA track and a health-professional-school REAP track (NIH R15).
• K series — career development. K01, K08, K23, and the K99/R00 Pathway to Independence buy protected research time for a named person, usually with a percent-effort commitment and an institutional letter.
• F series — individual fellowships. F30, F31, F32, and F33 support predoctoral and postdoctoral trainees directly under the National Research Service Award authority.
• T and P series. T32 institutional training grants and P01, P30, and P50 center awards go to the institution, which then appoints trainees or allocates core resources.
• U series — cooperative agreements. U01 has the scope of an R01 with substantial NIH programmatic involvement. The practical difference: on an R-series grant NIH funds and monitors; on a U-series agreement NIH funds and participates.

The activity code is not a strategy on its own. Each Institute and Center participates in only a subset, and the notice of funding opportunity always governs.

How does NIH review and fund an application?

NIH review runs in two stages, separated deliberately. Every competing application goes first to the Division of Receipt and Referral inside the Center for Scientific Review, which “assigns each application to a scientific review group (study section) with the expertise to evaluate the scientific and technical merit of the application and to one or more institutes/centers for the second level of review and for funding consideration” (NIH CSR).

Assignment is consequential and partly influenceable. The institute assignment determines whose budget and whose priorities decide funding; the study section determines who reads the science. Applicants may suggest both through the assignment request form, but published study section guidelines drive the outcome.

The study section is “composed primarily of non-federal scientists,” run by a Scientific Review Officer who assigns “at least three reviewers to assess each application” and serves as the designated federal official (NIH first-level review). Scoring uses “a 9-point rating scale (1 = exceptional; 9 = poor) in whole numbers (no decimals).” Applications judged less competitive — typically the bottom half — are not discussed and receive no numerical overall impact score.

Overall impact is not an average of criterion scores. It is each panelist’s holistic judgment of the project’s likely influence on the field, and the final score is “determined by calculating the mean score from the final impact scores from all the eligible members and multiplying the average by 10,” producing a range of 10 to 90.

Council review is the second gate. The advisory council of the assigned Institute or Center considers impact scores, percentiles, and summary statements “in light of the Institute/Center’s priorities,” and the “Institute or Center director makes final Funding Decisions based on staff and Advisory Council/Board advice” (NIH second-level review).

What is the difference between a percentile and a payline at NIH?

A percentile and a payline are different objects that applicants routinely conflate. A percentile is a ranking: it “shows the ranking of your application relative to the other applications reviewed by that study section at its last three meetings,” which normalizes for panels that score generously or harshly (NINDS).

Not all NIH applications are percentiled. Applications scored in a standing Center for Scientific Review study section generally are; applications scored in a special emphasis panel or at an Institute generally are not. An unpercentiled application competes in a materially different way, because paylines are usually expressed as percentiles.

A payline is a funding cutoff that an individual Institute or Center sets for its own fiscal year and applies to percentiled applications. Paylines are set by budget, not by science, and they differ across Institutes reviewing comparable work. Success rate — awards divided by competing applications — is a third number, and it usually sits above the payline because it includes non-percentiled and specially solicited awards.

Resubmission is a normal part of the NIH path rather than an admission of failure. NIH accepts one resubmission of an unfunded application, and the resubmission must arrive within a fixed window measured from the original submission (NIH submission policies). What to do with a summary statement is covered in what to do when a grant is declined.

How does NSF organize funding, and what proposal types exist?

NSF organizes funding by directorate, then division, then program, and a program officer inside that program owns the proposal from receipt through recommendation. Proposal preparation rules live in one document, the Proposal and Award Policies and Procedures Guide, which NSF replaces periodically (NSF PAPPG).

Beyond the standard research proposal, NSF publishes several distinct types worth knowing by name:

• CAREER. NSF’s early-career award, requiring an integrated research and education plan. A principal investigator “may submit only one CAREER proposal per annual competition” and “may not participate in more than three CAREER competitions,” and no co-principal investigators are permitted (NSF 22-586).
• Collaborative proposal. One project across multiple organizations, submitted either as a single proposal with subawards or as separate simultaneous proposals linked by NSF and reviewed together.
• EAGER. Early-concept grants for exploratory research on untested, potentially transformative ideas, reviewed internally rather than by external panel, and requiring program officer contact first.
• RAPID. Rapid-response research where urgency of access to data, facilities, or sites is severe — natural disasters and unanticipated events. Internal review, program officer contact required.
• Conference proposals. Support for meetings, symposia, and workshops, with their own budget rules around participant support costs and an expectation of early consultation.

NSF also distinguishes award instruments from proposal types. A standard grant obligates the full amount up front; a continuing grant obligates an increment and commits to later increments; a cooperative agreement is used where NSF expects substantial involvement.

What are NSF’s two merit review criteria?

NSF evaluates every proposal against two criteria approved by the National Science Board. Intellectual Merit is “the potential for the proposed project to advance knowledge and understanding within its own field or across different fields.” Broader Impacts is “the potential for the proposed project to benefit society and contribute to the achievement of specific, desired societal outcomes” (NSF).

Broader Impacts is scored, not decorated. NSF applies the same five review elements to the Broader Impacts statement that it applies to the science: whether the activity benefits society or advances desired societal outcomes; whether the activities explore “creative, original or potentially transformative concepts”; whether the plan is “well-reasoned, well-organized and based on sound rationale” and incorporates “a mechanism to assess success”; how well qualified the individual, team, or institution is; and whether adequate resources are available (NSF Broader Impacts).

Three consequences follow directly from that element list. A Broader Impacts section with no assessment mechanism is unscoreable against element three. A section promising outreach with no named partner, no qualified personnel, and no budget line fails elements four and five. And NSF states plainly that the work need not be bolted on — a project “can have broader impacts through: your research activities; activities directly related to your research; activities that are supported by, but complementary to, your research activities.”

Reviewer input is advisory. NSF also weighs portfolio factors when assembling awards, including “different approaches to significant research and education questions; potential (with perhaps high risk) for transformational advances in a field; capacity building in a new and promising research area; or achievement of special program objectives.” How reviewers actually read is covered in writing for the reviewer.

How do NIH and NSF research grants differ?

NIH and NSF differ on vocabulary, review structure, and where the decision authority sits. The table below compares the two agencies on the five dimensions that change how you prepare an application.

Agency	Award vocabulary	First-level review	Decision authority	Governing document
National Institutes of Health	Activity code plus Institute (R01, R21, K08, U01)	Study section run by a Scientific Review Officer, 9-point scale	Institute or Center director, after advisory council	NIH Grants Policy Statement
National Science Foundation	Program within a directorate	Ad hoc reviewers, panel, or both, rated Excellent to Poor	Program officer recommends; division director concurs	Proposal and Award Policies and Procedures Guide

Four further divergences matter operationally. NIH runs three council rounds a year on fixed due dates, while many NSF programs have moved away from deadlines entirely. NIH formally structures resubmission with an A1 designation; NSF simply allows a declined proposal to be resubmitted after substantial revision. NIH review is firewalled — the Scientific Review Officer who runs the panel is not the program officer who advocates for funding — while at NSF the program officer selects reviewers and writes the recommendation. And NSF enforces proposal formatting strictly enough that non-compliant proposals are returned without review.

What the two agencies share is more load-bearing than what separates them. Both run peer review by working scientists. Both publish a single governing policy document that supersedes folklore. Both employ program officers who can be contacted before submission and who will tell you whether an idea fits a program, what the portfolio already contains, and which vehicle is right. Neither can be talked into funding a bad fit.

What goes wrong in NIH and NSF applications?

Applications to NIH and NSF fail in patterned ways, most of them structural rather than scientific.

• Wrong assignment at NIH. An application routed to an Institute with no budget line for the disease area, or to a study section without the right methodological expertise, is scored by the wrong people.
• Treating Broader Impacts as boilerplate at NSF. NSF scores it against five explicit elements, and the assessment mechanism is the element applicants skip most often.
• Choosing the activity code by budget size. R21 and R03 are terminal, non-renewable instruments with hard caps, not smaller R01s that grow later.
• Confusing percentile with payline. A strong percentile in a generously scoring panel and a strong percentile against an Institute’s cutoff are different competitive positions.
• Formatting non-compliance at NSF. Font, margin, pagination, and page-limit rules in the PAPPG are enforced, and a returned proposal never reaches a reviewer.
• Submitting before talking to anyone. Both agencies publish funded-award databases and staff contacts, and both expect pre-submission conversation on non-standard proposal types.

The most valuable pre-submission work at either agency is portfolio intelligence: reading what the program has already funded, then confirming fit with a program officer. That conversation is free, and it is the one input reviewers never see but the decision-maker always has.

Frequently asked questions

Can an individual apply for an NIH or NSF research grant?

Generally no for research project grants. NIH R-series and NSF research awards are made to institutions, with a named principal investigator. NIH F-series fellowships are the notable individual mechanism, and even those flow through a sponsoring institution. Eligibility is set in each notice of funding opportunity.

Is an NIH cooperative agreement different from an NIH grant?

Yes. A U-series cooperative agreement carries the same scientific scope as the comparable R-series grant, but NIH staff participate substantively in the work — approving plans, redirecting effort, joining the technical team. The instrument distinction is statutory, not stylistic, and it changes day-to-day autonomy.

What is an NIH summary statement?

A summary statement is the official record of peer review for an application. The Scientific Review Officer prepares it, and it contains the reviewers’ critiques, a resume of the discussion if the application was discussed, the score, and budget and administrative notes (NINDS).

Does NSF weight Intellectual Merit above Broader Impacts?

NSF does not publish a weighting between the two criteria, and does not publish a numeric scoring formula. Reviewers write narrative reviews and assign an overall rating; the program officer synthesizes those with portfolio considerations. Both criteria are assessed against the same five review elements.

How long does an NSF decision take?

NSF states that it “strives to be able to tell applicants whether their proposals have been declined or recommended for funding within six months” of submission (NSF). Business and financial review by NSF’s grants office follows a recommendation and can still result in a decline.

Where are the authoritative rules published?

For NIH, the NIH Grants Policy Statement and the notice of funding opportunity. For NSF, the Proposal and Award Policies and Procedures Guide and the program solicitation (NSF PAPPG). Both change on their own schedules, and the solicitation always overrides the general guide where the two conflict.

Related topics

• Funding Tracks by Funder Type — the hub for how each funding track behaves
• SBIR and STTR Explained
• Federal Grants Explained
• Grants vs Contracts vs Cooperative Agreements
• How to Read a NOFO

Sources

1. National Institutes of Health. List of NIH Institutes, Centers, and Offices. https://www.nih.gov/institutes-nih/list-nih-institutes-centers-offices (accessed 2026-08-11)
2. National Institutes of Health. Activity Codes Search Results. https://grants.nih.gov/funding/activity-codes (accessed 2026-08-11)
3. National Institutes of Health. Research Project (R01). https://grants.nih.gov/funding/activity-codes/R01 (accessed 2026-08-11)
4. National Institutes of Health. Exploratory/Developmental Research Grant (R21). https://grants.nih.gov/funding/activity-codes/r21 (accessed 2026-08-11)
5. National Institutes of Health. Small Research Grants (R03). https://grants.nih.gov/funding/activity-codes/R03 (accessed 2026-08-11)
6. National Institutes of Health. Research Enhancement Awards (R15). https://grants.nih.gov/funding/activity-codes/R15 (accessed 2026-08-11)
7. NIH Center for Scientific Review. The Assignment Process. https://public.csr.nih.gov/ForApplicants/SubmissionAndAssignment/DRR/assignmentprocess (accessed 2026-08-11)
8. National Institutes of Health. First Level: Peer Review. https://grants.nih.gov/grants-process/review/first-level (accessed 2026-08-11)
9. National Institutes of Health. Second Level: Advisory Council Review. https://grants.nih.gov/grants-process/review/second-level (accessed 2026-08-11)
10. National Institutes of Health. Submission Policies. https://grants.nih.gov/grants-process/submit/submission-policies (accessed 2026-08-11)
11. National Institutes of Health. NIH Grants Policy Statement, Section 2.3.7 — Policies Affecting Applications. https://grants.nih.gov/grants/policy/nihgps/HTML5/section_2/2.3.7_policies_affecting_applications.htm (accessed 2026-08-11)
12. National Institutes of Health. Fiscal Year 2025 by the Numbers: Extramural Grant Investments in Research. NIH Extramural Nexus, March 2026. https://grants.nih.gov/news-events/nih-extramural-nexus-news/2026/03/fiscal-year-2025-by-the-numbers-extramural-grant-investments-in-research (accessed 2026-08-11)
13. National Institute of Neurological Disorders and Stroke. Scoring and Summary Statements. https://www.ninds.nih.gov/funding/after-you-submit/understanding-review-process/scoring-and-summary-statements (accessed 2026-08-11)
14. U.S. National Science Foundation. How We Make Funding Decisions. https://www.nsf.gov/funding/merit-review (accessed 2026-08-11)
15. U.S. National Science Foundation. Broader Impacts. https://www.nsf.gov/funding/learn/broader-impacts (accessed 2026-08-11)
16. U.S. National Science Foundation. Proposal and Award Policies and Procedures Guide (PAPPG). https://www.nsf.gov/policies/pappg (accessed 2026-08-11)
17. U.S. National Science Foundation. Directorates and Offices. https://www.nsf.gov/about/directorates-offices (accessed 2026-08-11)
18. U.S. National Science Foundation. Faculty Early Career Development Program (CAREER), NSF 22-586. https://www.nsf.gov/funding/opportunities/career-faculty-early-career-development-program/nsf22-586/solicitation (accessed 2026-08-11)