• Only 57% of Americans say science has had a mostly positive effect on society, down from 73% before the pandemic. Only 45% think scientists are good communicators.
• Focus groups before Brexit found that people shown data were more likely to support leaving the EU, while those told human stories were more likely to oppose it — a reminder that facts alone don’t persuade.
• Congress has so far defended federal science funding on a bipartisan basis, but maintaining that support requires scientists and their institutions to connect with communities, not just publish findings.
In fall 1985, the Royal Society published a study that changed how scientists thought about their jobs.
“A basic thesis of this report is that better public understanding of science can be a major element in promoting national prosperity, in raising the quality of public and private decision-making and in enriching the life of the individual,” stated the Bodmer Report, named for its influential geneticist-coauthor Walter Bodmer.
“All scientists need, therefore, to learn about the media and their constraints and learn how to explain science simply, without jargon and without being condescending.”
The lesson is that how you communicate uncertainty matters as much as what you communicate.
The report helped launch a field: science communication. A movement followed to expand public understanding: press outreach, university programs, media training, events. April as “Citizen Science Month,” a campaign led by Philadelphia-founded SciStarter, draws its intellectual lineage from the Bodmer Report.
Forty years later, that infrastructure is being stress-tested like never before.
At a symposium hosted by the University System of Maryland last month, researchers, communicators and journalists gathered to confront an uncomfortable question: Why, despite decades of investment in science communication, has public trust in science declined so sharply?
The symposium was organized by Michele Masucci, Maryland’s vice chancellor for research and economic development. Her work is focused on how brainy science in a lab affects development on a local street. Citizen Science Month and SciStarter are led by Darlene Cavalier, the onetime Sixers cheerleader turned longtime science communicator. Cavalier has spent a decade encouraging residents to contribute to real scientific research, from transcribing historic records to analyzing brain images for Alzheimer’s studies.
Both, and many others, have done plenty for participation in science. But the trend has only gone against them.
Only 57% of Americans now say science has had a mostly positive effect on society, down from 73% before the COVID pandemic, per Pew Research Center presented at the symposium by Brian Kennedy, a director of science and society research. The share expressing “a great deal” of confidence in scientists has dropped from 39% in 2020 to 23% in 2023. And only 45% of Americans describe scientists as good communicators.
The partisan divide is stark. Among Republicans, just 47% view science positively, down from 70% in 2019. Among Democrats, the figure is 69%. I asked Kennedy if socioeconomic factors predict views of science, but he said race is a better predictor: More than 70% of Asian Americans say science has had a “mostly positive” impact on society, compared to fewer than half of Black Americans (48%). White and Hispanic are closer (65% and 57% respectively). Nearly 90% of left-leaning white Americans view science positively, compared to half of right-leaning Hispanic Americans.
Kennedy, of Pew, noted that for most Americans, their primary encounter with science is through their own health — or through spectacles like a NASA launch. They’re not reading journals or attending conferences and, at least as critically, they may well not know a scientist personally. That’s something Cavalier’s SciStarter movement intended to address. She identified elite athletes and cheerleaders with serious science training to familiarize the concepts.
If these science champions have won any battles, they’re surely losing the war.
Evidence alone is no longer enough
I joined a panel at Masucci’s symposium in Baltimore at the University of Maryland’s gleaming new 4MLK building, alongside Megan Nicholson, a senior editor at Issues in Science and Technology, and Heath Kelsey, director of the Integration and Application Network at the University of Maryland Center for Environmental Science. The moderator, Michael Sandler, the UM system’s vice chancellor for communications and marketing, opened with a simple premise: Research that nobody knows about cannot have impact.
What followed was a conversation about how to make research matter — and why that task has become so difficult.
Nicholson, whose magazine publishes long-form arguments aimed at policymakers, emphasized that a good research story requires an awareness of what people are already experiencing in their media environment.
“You need to know what you want people to get out of it at the start,” she said.
Kelsey, whose team produces the annual Chesapeake Bay report card, made a different point. The old model of science communication — a researcher presenting findings to a passive audience — is obsolete.
“Now it’s much more of a dialogue,” he said. “And I think having that kind of dialogue and back and forth will also help us.”
His team has seen this work in practice. On a project assessing the Orinoco River Basin in Colombia, researchers engaged local communities from the beginning: environmental groups, engineers, economists, even the cattlemen’s association. By the time the research was complete, those communities had bought in. One pristine watershed was designated a protected Ramsar wetland as a direct result.
“When people in the community are engaged from the beginning of the research design all the way through the end,” Kelsey said, “that community will be trusting in the science that you’re putting out there.”
But trust is not built only through process. It is built through narrative.
In his 2019 book Nervous States, author William Davies describes focus group research conducted before the Brexit referendum. Participants who were shown data about immigration’s economic effects were more likely to support leaving the European Union. Participants who were told stories about individual immigrants were more likely to oppose it.
The finding is counterintuitive for anyone trained to believe that evidence persuades. But it reflects a deeper truth about human cognition. We do not first evaluate claims and then decide whether to trust the messenger. We decide whether we trust the messenger, and then evaluate the claims. Scientists sensibly spend as much time as possible thinking about what the science tells them. They have not always considered that a wider public might draw entirely different conclusions from the same information.
This has implications for how research institutions approach communication. A university communications office that chases New York Times coverage may be serving prestige, but it may not be serving trust in an especially polarized world. The neighbor who sees a scientist quoted in the local paper, or featured on a community podcast, is more likely to see that scientist as a person instead of an emissary from an ivory tower.
The ivory tower framing is dangerous because it is partly deserved.
When public health authorities reversed guidance on masks early in the pandemic, trust eroded. Not because the science changed, but because the communication failed. A PhD told me: “Masks work, but mask mandates don’t.” One expert later described the CDC’s mask messaging as “the worst example of risk communication in the modern era.” Research on Twitter sentiment found that the public responded negatively to both the initial mask recommendation and its later relaxation. Trust in the CDC fell notably from the pandemic’s start to spring 2022, and has not recovered.
Being a good communicator has become existential for scientists
The lesson is not that scientists should avoid changing their conclusions when evidence changes. The lesson is that how you communicate uncertainty matters as much as what you communicate.
When institutions oversell certainty, or reverse course without explaining why, they invite the accusation that science is just another political actor.
That accusation has become central to the current political environment. The second Trump administration has cancelled more than $8 billion in already-approved federal research grants, targeting projects whose descriptions included terms like “climate justice” or “gender minority.” The National Institutes of Health and National Science Foundation have seen the largest cuts in a generation.
And yet: Congress has so far defended federal science funding on a bipartisan basis, restoring many, though not all, of the Trump administration’s most draconian cuts. Why? Because science, at its core, is not partisan. Republicans and Democrats both get sick, marvel at technology and occasionally lose themselves in the wonder of the stars.
If the worst outcomes were stopped, the broader risks have not abated. Worryingly, specific research grants are still lagging, “grim” as one researcher called it: “the very real and ongoing dismantlement of federal cancer research.”
The Bodmer Report was right that scientists have a duty to communicate. What has changed is the stakes.
In 1985, the argument was about civic engagement and scientific literacy. In 2026, the argument is about whether research institutions can maintain the public support they need to exist. It is about whether evidence will shape policy or be dismissed as ideology. It is about whether the communities most affected by scientific questions — on health, on climate, on technology — will see scientists as partners or as adversaries.
The need from scientists is clear enough. Engage communities early. Tell human stories, not just data. Meet people where they consume information, including platforms and formats that feel unfamiliar. Acknowledge uncertainty honestly. And remember that trust, once lost, is very hard to regain.
“If modern science is our greatest cultural achievement, then it is one of which most members of our culture are largely ignorant,” The Bodmer Report said.
“Our most direct and urgent message must be to the scientists themselves: Learn to communicate with the public, be willing to do so and consider it your duty to do so.”
