The political divide between the left and the right in the US has become particularly stark in recent years, and it seems this gap is even apparent in how the two major political parties use science in their policies. According to a new analysis of a massive database of policy documents, there are systematic differences in the amount, content, and character of science cited in policy between partisan factions. Science has become an important part of our society, and therefore, the policy-making world as well. It provides decision makers with the primary sources of evidence that inform potential solutions to various issues and offers them authority. Its role has only become more crucial as pressures facing society today, such as climate change, public health, and the advancement of technology, have become more acute. These issues are not only significant for society; they are also intractably tied up with scientific progress. However, as the political climate in the US has become increasingly polarized, some researchers have asked a vital question: Is science used differently by policymakers in the different parties? To assess this, a team of researchers led by Alexander Furnas, a political scientist at Northwestern University, examined nearly 50,000 congressional committee policy documents produced between 1995 and 2021. They accessed this information from the government-policy database Overton, the world’s largest policy and grey paper database. They also examined over 190,000 reports from 121 US-based ideological think tanks for a similar period (starting in 1999). In total, these reports contained 424,199 scientific references, which were then matched to a secondary dataset from Dimensions, a massive publication and citation database that captures 122 million scientific publications across disciplines. Across the time period being studied, both Republican and Democratic-controlled committees have referenced scientific papers more often; however, the patterns of growth differ between the parties, and the gap has increased over time. For instance, the team’s statistical analysis showed that for every two-year congressional cycle, Democratic-controlled committees had a higher probability of citing scientific literature than Republican-controlled committees. “Estimating the overall partisan differences with year and committee fixed effects, we find that policy documents from Democratic-controlled committees are nearly 1.8 times more likely to cite science than those from Republican-controlled committees,” the team writes in their paper. “These results are robust when accounting for indicator variables for chamber, document type, and whether there is a copartisan president.” The team found that this effect was most obvious in the House and in committee meeting documents, and does not appear to vary between committee versus subcommittee status. However, the difference between science citations was most evident in reports produced by ideological think tanks. According to Furnas and colleagues, these groups are “key resources” for partisan policymakers, as they help extend party networks and provide “legislative subsidies”, set agendas, and develop policy alternatives. Here, the researchers found that policy documents from left-wing think tanks were five times more likely to cite scientific research than their right-wing counterparts, with this difference being widespread across fields and policy issues. Moreover, the results showed that Democrats and Republican-leaning policymakers drew from different science as well, as only 5 to 6 percent of scientific citations were shared by both groups. This difference is further demonstrated by the issues the respective parties sought research over. According to Furnas and colleagues, the topics cited by the US House Committee Energy and Commerce varied considerably when controlled by Democrats or Republicans. For instance, when the former was in control, science citations tended to include abortion, drunk driving, youth and e-cigarettes, energy production, infrastructure, and guns, violence, and mental health. In contrast, Republican-controlled committees cited science covering health care, insurance costs, air pollution, opioids, and high-school athletic injuries, among others. These results clearly demonstrate the level of difference in how science is cited across the political divide, and this raises concerns. As the team writes in their conclusion: “Ultimately, science is a crucial public good and depends on both sustained public support and long-term commitments. By contrast, the American political landscape is characterized by inherent volatility and periodic shifts in political control. Despite recent instances of bipartisan support for science, the uncovered partisan differences in the use and trust of science highlight a profound tension at the nexus of science and politics.” The study is published in Science.