The newsbeat is, well, beating Michelle and I thanks to proposal defenses, holidays, and our own involvement in politics around the latest tax bill. But we’d like to turn back the news cycle to October and look at a proposal from Senator Rand Paul related to research funding decisions at federal agencies. The BASIC Research Act would require a “taxpayer advocate” on research proposal review panels, the groups of scientists who decide what proposals to fund. Senator Paul deserves credit for seeking to integrate more perspectives into the research funding process, but his vaguely defined taxpayer advocate would fail to better connect research to desired societal outcomes.
Congress has a history of skewering federal agencies and scientists for wasting taxpayer money on research that puts shrimp on tiny treadmills, asks questions of single people on speed dates, studies snail sex1, or examines farm-based tourism in rural China2. Senator Paul regularly complains about NSF, often through poorly written Waste Reports that attract the ire of science advocates. His bill invited similar reactions. For example, one reporter accused Senator Paul of politicizing or misunderstanding science because the bill proposes interventions in the peer review process that federal agencies, such as the National Science Foundation (NSF), use to dole out research money. At the hearing where the bill was introduced, Senator Paul justifiedthe bill as a mechanism to weed out “bad science,” to create a more transparent funding system, and to fight the perceived cronyism of scientists approving funding for the work of friends and colleagues. The bill covers a lot of ground, including requirements to make grant applications to federal research agencies public. The most interesting pieces of the bill, however, pertain to proposal review panels, the groups of scientists who decide which requests get funded. If passed, the bill would forbid researchers from suggesting potential reviewers for their applications and would require agencies to add two people to their review panels: 1) an expert from outside the discipline and outside academia; and 2) someone to serve as a taxpayer advocate.
Senator Paul’s bill centers on long-standing debates about which projects should receive taxpayer funding and who should make those funding decisions. Vannevar Bush, the lauded architect of postwar American science policy, and Senator Harley Kilgore of West Virginia, engaged these questions 70 years ago at the inception of the National Science Foundation3. Senator Kilgore argued for increased federal investment in social sciences, more equity in the geography of federal science investment4, and a shift away from military research. Bush parlayed his success managing the wartime U.S. Office of Scientific Research and Development, which led the Manhattan Project, into a different vision for federal investment in research, one that placed scientific excellence as judged by (mostly physical) scientists at the forefront of funding decisions. Today’s NSF largely resembles Bush’s ideal: Scientists make decisions about which proposals merit funding within individual disciplines based on the scientific merit of those proposals (as opposed to other criteria, like regional economic development or potential technological applications). While Senator Paul’s bill is no present day incarnation of Kilgore’s science policy, it nonetheless attempts to bring non-scientists into the fold of science governance with the aim of improving science5; Bush might be appalled.
That bit of historical context in place, let’s dig through Senator Paul’s proposal for a taxpayer advocate on grant review panels. Frankly, I agree with the spirit of this proposal, though not Senator Paul’s word choice nor his analysis around the topic. Precedent for public involvement in federal decision-making exists in many agencies, often taking the shape of required public comment periods6. The myth that only scientists can govern science, argued by Kilgore, leaves no place for such external input in science funding decisions. But bringing non-scientists to the table could, in theory, better connect federal investment in research to the ‘goods’ scientists and science advocates parade about to justify that investment, such as economic development, new products, new medicines, better health, happier people, and so on. Non-scientists provide broader perspectives, ideas, and critiques of research decisions, forging research robust to conceptual and practical criticism. For example, non-scientists might question troubled research mechanisms like studies of brain health that rely on animal models shown to be disconnected, even irrelevant, from processes in the human brain. And by nature of being external to engrained research pathways, non-scientists could offer constructive insight and critique where institutional inertia has taken hold7.
That said, Senator Paul’s proposal contains many loose ends that cloud assessment of the bill: Who should taxpayer advocates be and how do agencies select them? Are taxpayer advocates on the federal payroll (would love to see Taxpayer Advocate on usajobs.gov) or is this a service expected of people external to the federal government? Are they paid at all (reviewers for NSF, for example, are generally not paid)? What, exactly, does it mean to advocate on behalf of the taxpayer? Is the taxpayer advocate just Rand Paul (we kid…)? Different answers to these questions could lead to very different outcomes. A member of the public could face a steep learning curve to understand different research fields. Yet having someone internal to federal agencies might fail to bring new perspectives to the table or could amount to a rubber stamp on projects8. Further, structural issues with Senator Paul’s proposal pose more questions. Is one person sufficient for establishing taxpayer interests on a review panel? Is having input external to the scientific community most useful at the individual grant level? The bill’s language fails to clarify any of these questions, meaning we might see little change in the role of public value in research funding decisions.
Let’s assume that the taxpayer advocate is someone disconnected from the scientific enterprise: they’re a member of broader society, which ostensibly sounds like a win for both the taxpayer and for research (at least in our view). But the involvement of non-experts brings up more issues.
Non-expert panelists could be too intimidated to question the rest of the panel members due to a lack of topical expertise9. Perhaps the logistics of getting public members to such review panels would mean the ‘usual suspects’ might show up: those already highly engaged in debates around science, like people from science advocacy organizations. Or perhaps other forms of political power would lead to competition for seats on panels, turning them into politically contested arenas; imagine a lobbying firm or an interest group pushing hard to get ‘friendly’ public members on the panels through one mechanism or another. Further, selecting members of the public to sit in as reviewers creates layer upon layer of difficult decisions about recruitment and preparation. Simply put, involving the public in this way is a big challenge, particular for agencies without experience in this type of engagement10.
Given these questions, Senator Paul’s proposal for a taxpayer advocate would do little to clear the air on research funding decisions. Whatever his intentions, Senator Paul’s bill avoids the important, broader question of how to better connect investments in research to public goods and outcomes used to justify research, despite nods to this ideal in his hearing. Perhaps he and other legislators have avoided (and continue to steer clear of) the question because it’s a hard nut to crack and would require substantial institutional changes at federal agencies.
But what can we take away from Senator Paul’s proposal that might be effective? And how could we build on and improve it?
Having diverse input as to which research programs merit federal investment is a laudable goal. Rather than involve the public on individual proposal review panels, non-scientists from the public could be brought into decision-making about federal research earlier in the funding process to set larger goals for a particular research program. At this stage, non-experts can interface with scientists and determine priorities for research investments without digging into the nitty-gritty of a particular method or theory for every proposal considered. Further, public involvement at this level allows for conversations about the ethical and political ramifications of research. Research on public participation in technical decision-making shows that non-expert people from all walks of life can thoughtfully and constructively contribute to technical decisions when given the opportunity. While large-scale involvement might be unduly expensive for any given research program, smaller panels (5-10 people) could spend a few days working with program managers to identify potential issues and societal outcomes and help set research priorities accordingly.
Senator Paul should look elsewhere in the federal government for inspiration; in DARPA, for example, research funding decisions revolve around specific problems and empowered program managers, not peer-review panels (or perhaps he should wait for Michelle’s dissertation to be published regarding the National Park Service). Alternatively, Congress could ask federal agencies to do a bit of experimentation. Invite non-scientists to serve on a few review panels, even if only in an advisory role. Create panels for various research initiatives and ask non-scientists, or better yet diverse members of the public, to help set research agendas and write calls for proposals. In short, play around with different mechanisms appropriate to the agency. Several agencies, including USGS, EPA, NSF, and NASA, support external research, each with distinct institutional shapes and sizes. A little experimentation could go a long way to creating new ways for non-scientists to contribute their expertise and values to funding decisions.
Would taxpayer advocates on review panels better connect research to the outcomes we use to justify funding research? Without more detail, we’re stuck with an ambiguous “maybe”. Perhaps Michelle and I should apply for NSF funding to test citizen review panels for NSF research programs (no really, that would be super cool to study). Or maybe that proposal would end up in Senator Paul’s next Waste Report.
– Nich Weller
- Michelle was part of this research team as an undergraduate.
- The National Science Foundation is used to congressional criticism and does its best to encourage researchers to communicate public values. When I received a small NSF grant for the project examining Chinese farm-based tourism linked above (a relic of a past dissertation avenue), NSF program officers encouraged me to clearly articulate the public value of my proposal in the summary for NSF’s website, ostensibly to keep congressional threats at bay. In our opinion, this does little to promote public value in research but that’s a topic for another day.
- See Gregg Pascal Zachary’s Endless Frontier: Vannevar Bush, Engineer of the American Century for more on Kilgore and Bush’s shared ideas and disputes.
- Notably, West Virginia lacked, and still lacks, major sites of federal science investment like national labs.
- Senator Paul sees external input, which Bush would have detested as interference in scientific excellence, as a path to ‘good’ science, or at least a path away from ‘bad’ science (in the words of the Senator’s testimony).
- Federal public comment periods, however, have glaring problems as 1) existing interest groups dominate the process and 2) they require substantial time, expertise, and resources to participate in.
- Are broader perspectives a panacea for what ails science? Of course not. Believing so is just as detrimental as believing that non-scientists and the public have nothing relevant or wise to say about research.
- An assessment of NSF’s policy requiring ‘broader impact statements’ on proposals might help answer these questions.
- This is a very real phenomena I must account for in my research.
- Agencies could try to do this independently, without a legislative mandate, and “see what works,” which might also provide them political clout if congress passes legislation on the topic.