By Erica Kimmerling, Amanda Vernon, Cate Young (Federation of American Scientists), and Gil Ruiz (Federation of American Scientists)
With draft guidance at the National Science Foundation (NSF) open for public comment, the potential reauthorization of NSF on the horizon, and a new report from the Office of Science and Technology Policy encouraging public impact and portfolio management of publicly funded science, now is the time to consider policy changes that would move NSF research towards greater positive impact. This conversation is valuable now, even in the context of a broader effort to defend grantmaking writ large, in order to be prepared for open policy opportunities as they arise.
The NSF and scientific grantmaking
The NSF has historically achieved notable public impact by making investments in research that generate new knowledge, producing a series of discoveries that, over decades, lay the groundwork for innovations which improve the quality of life of everyday Americans. NSF investments typically have been made via grants, although in recent years, with bipartisan and apolitical encouragement, NSF has started to experiment with additional models (see Regional Innovation Engines and X-Labs), particularly in cases where the goal is to translate foundational research into useful applications. The structure and design of these new models is under active policy debate (recently on X-Labs, see here, here, and here), but there is an underappreciated opportunity to improve how NSF’s grant-based, fundamental research investments could achieve greater public benefit.
Grants for scientific and engineering research are core to the agency’s mission — and NSF represents the third-largest source of federal funding for basic research. The process by which NSF evaluates and funds scientific grants is therefore a strong policy lever for strengthening fundamental research. At a time when there is an active effort to defend the underlying strengths of federal grantmaking, there is also evidence that NSF’s grantmaking processes can, and should, be strengthened to achieve greater public benefit. We provide answers to questions that are at the core of this topic below.
Why should the NSF care about whether their grants for scientific and engineering research might benefit the public?
When the NSF was created by Congress in 1950, it was charged to “promote the progress of science; advance the national health, prosperity and welfare; and secure the national defense; and for other purposes.” Public benefit has therefore been baked in from the very beginning, with merit review as the process for delivering on this mission, and with funds for NSF-supported research coming largely from taxpayer dollars.
Congress requires that, in addition to reviewing for “intellectual merit,” or advancing knowledge, that NSF proposals be evaluated for “broader impacts,” or research benefits beyond knowledge generation. Stated differently, the inclusion of these two criteria for evaluating scientific funding are a clear signal that while knowledge generation is a necessary outcome of research, it is on its own insufficient and must be paired with a full consideration of the funded project’s potential to produce public benefit. Congress has spelled out potential broader impacts:
(1) Increasing the economic competitiveness of the United States.
(2) Advancing of the health and welfare of the American public.
(3) Supporting the national defense of the United States.
(4) Enhancing partnerships between academia and industry in the United States.
(5) Developing an American STEM workforce that is globally competitive through improved pre-kindergarten through grade 12 STEM education and teacher development, and improved undergraduate STEM education and instruction.
(6) Improving public scientific literacy and engagement with science and technology in the United States.
(7) Expanding participation of women and individuals from underrepresented groups in STEM.
These societal benefits cast a wide net, appropriate for an agency that uniquely funds all fields of science and engineering and a range of translational timelines from months to decades. As outlined in the original NSF Act, these impacts can be achieved through the proposed research itself, related activities to the conduct of the project (such as training and mentorship), and through activities that are complementary to the research. A huge diversity of proposals and types of research can have public impact — but those benefits need to be thoughtfully considered by researchers and institutions making proposals, and they need to be managed as a portfolio to maximize return on investment by the federal government.
Historical challenges for NSF’s dual merit review criteria
While policy has strengthened over time and now requires full consideration of both “intellectual merit” and “broader impacts,” societal benefits have not in practice received equal treatment during the review process and the conduct of scientific research (see here and here for more). At its worst, the devaluing of the proposed broader impacts criterion can cause this aspect of proposals to be viewed as a “check box,” a one-off need for an outreach exercise disconnected from the research itself, or “tie-breaker” for otherwise intellectually similar proposals. Although progress has been made, the relatively weak execution in the evaluation and measurement of broader impacts risks perpetuating a negative feedback loop where universities and researchers do not invest adequate resources and attention into the societal benefits aspect of a project because they are not being evaluated for this work in order to get funding.
The end result is a scientific portfolio with significant potential unrealized gains. While NSF has funded research that produced new knowledge of immense public benefit (advanced medical technologies, early development of the internet, and many others), these triumphs have often been a byproduct of excellent research. Fully considering the broader impacts criterion has the potential to help across the research lifecycle: attracting proposals (and researchers) that place public benefit at the fore; selecting proposals with greater potential for high return-on-investment; and supporting improved infrastructure for the use, uptake, and translation of the knowledge that is generated.
A call for change was recently affirmed in December 2025 in a report on merit review from the National Science Board (NSB), the presidentially appointed governing Board of NSF. Some of the NSB’s recommendations would require policy shifts; many could be achieved through culture change and practical execution. Importantly, the Board emphasizes that there should be no changes that would diminish the intellectual merit of NSF-funded research.
So what actions would strengthen NSF grantmaking to support greater public impact?
First, fully embracing the broader impacts criterion as a co-equal review element. As the NSB’s report points out, there’s a need for better consistency within the agency and better understanding in the scientific community. More transparent proposal evaluation, including separate scoring of the two criteria, would help, as would clearer policy and guidance. We need change for the “broader impacts” criterion to be treated seriously in practice; we also need change for the full breadth of potential societal benefits to be considered in the grantmaking process. Once NSF has a more solid evidence base on proposal evaluation and a fuller embrace of the criterion in proposal submissions and evaluations, the agency will be better equipped to make wise decisions, or to experiment, on whether and how to weigh criteria.
Second, approaching public impact from a portfolio management lens at the level of Directorates and the agency overall. Although all proposals should have the potential for societal benefits, when and how that possibility would manifest will vary dramatically across fields and proposals. That’s OK! But it means the agency needs the data infrastructure and analysis capabilities to understand and report to the public more effectively how a portfolio of investments is making a difference – and a strategy that can adapt to grow this impact over time. Applying a portfolio management lens would also help the agency to strategically consider its funding mechanisms overall, beyond grantmaking.
Third, broadening the reviewer pool to bring in the range of perspectives necessary to have relevant expertise across the review criteria. This is more important than ever, given the growth of research activities in industry and amid concerns about declining public trust in science; only just over half of NSF staff surveyed believe the reviewer pool to represent a sufficient range of perspectives. Changing this would help keep NSF relevant and informed, and assessment of a proposal’s potential for broader impacts would benefit from expert perspectives from individuals in a wide range of public impact domains — not just the technical domain of the research.
Why is this topic important right now?
NSF has released draft guidance that is open for public comment until August 24. This is a window for changing merit review guidelines towards greater public impact and for adopting a “portfolio-based approach.”
And the new Office of Science and Technology Policy report, Science: A New Golden Age, underlines the imperative to orient federal science agency funding towards public impact and of adopting a “portfolio-based approach.”
Plus, the next reauthorization of NSF has the potential to come up in the next Congress — another window for shaping merit review of agency proposals.
Does the NSF draft guidance under consideration make these changes?
Though this blog is specifically focused on opportunities to reform merit review in the NSF grantmaking process to improve public benefit, it’s worth noting that the draft guidance would also generally bring NSF in line with the changes proposed in the draft OMB rule under consideration — and we have concerns with the potential negative impact of that rule on the scientific enterprise and at NSF specifically, both in its direct effects and in terms of its potential to erode NSF’s ability to evaluate and support proposals fairly – both for their intellectual merit and their broader impacts.
If you’re interested in reading about other aspects of the draft guidance, see the Union of Concerned Scientists blog post that goes into more detail.
On the question of whether the draft NSF guidance moves merit review process language in the right direction? Yes and no…
Yes: the draft re-affirms the two fundamental review criteria, preserving a rigorous process of expert assessment of research ideas. It removes the potential of “other criteria,” and it endorses recruiting a broad reviewer pool that can assess ideas across fields, contexts, and opportunities for impact.
No: the draft doesn’t bring in other policy language the NSB recommended that would strengthen broader impacts consideration in proposals, such as proposed language that “proposers, reviewers, and NSF staff should equally and transparently address both criteria…as input to the development of balanced award portfolios that address a range of Intellectual Merit and Broader Impacts objectives.”
Interestingly, although it is not in the draft guidance, the Administration’s recently released R&D Priorities Memo does support the portfolio-development goal, saying that “Agencies should adopt a deliberate, portfolio-based approach that matches funding mechanisms to the S&T [science and technology] challenges they seek to address, maximizing Federal return on investment through an explicit mix of modalities, risk profiles, and time horizons.”
OK, I’m interested. How can I get involved on this topic?
If you want to provide input into the public process currently underway at NSF, you can add your comment here. If you want to talk more about this topic, reach out — we welcome ideas on how to strengthen merit review or grantmaking.