Growing public reservations against science, including distrust of experts and preference for alternative beliefs, threaten science's ability to inform decisions and address societal challenges. This anti-science sentiment can reduce knowledge uptake, policy compliance, and confidence in scientific endeavors. While previous research suggests a correlation between legacy media (newspapers, TV, radio) use and reduced anti-science attitudes—possibly because these media often portray science positively—this research primarily focuses on Western populations and pro-science attitudes, neglecting crucial global and contextual factors. This study addresses these limitations by using a large-scale global dataset to explore the prevalence of anti-science attitudes and the complex relationship between legacy media use and these attitudes across diverse cultural and political contexts. The study considers several crucial factors that may influence this relationship including freedom of the press, populist political rhetoric, and cultural tendencies towards uncertainty avoidance. The overarching research question is whether the relationship between media consumption and anti-science attitudes is consistently negative across the globe or is shaped by country-level characteristics.

Existing research shows that anti-science attitudes are more common among specific demographic groups (older, less educated, more religious, politically conservative individuals). Studies also indicate that frequent legacy media use is often associated with less anti-science sentiment, potentially due to positive portrayals of science and scientists. However, this research is limited geographically (mostly Western countries), methodologically (often focusing on pro-science attitudes rather than directly on anti-science attitudes), and contextually (ignoring the impact of national-level factors). Some studies highlight the role of framing and cultivation in shaping public perceptions, suggesting that long-term exposure to positive media portrayals of science can cultivate pro-science attitudes. Yet, the influence of country-level factors like freedom of the press, populist rhetoric, and uncertainty avoidance on the relationship between media use and anti-science attitudes remains largely unexplored.

This study utilizes data from the 2017-2020 wave of the World Values Survey (WVS), a large-scale cross-national survey. The analysis includes data from 49 countries (N = 70,867), with inferential analyses using data from 38 countries (N = 54,658) where all necessary variables were measured. Anti-science attitudes were measured using a composite score based on three items assessing rejection of scientific authority in favor of faith, belief that science erodes moral values, and lack of interest in scientific knowledge. Legacy media use was measured by averaging responses indicating frequency of newspaper, TV news, and radio news consumption. Country-level factors—freedom of the press (Freedom House Index), prevalence of populist rhetoric (Global Party Survey), and uncertainty avoidance (Hofstede Index)—were included as covariates. Bayesian multilevel regression models were employed to test the relationship between legacy media use and anti-science attitudes, examining how this relationship varies across countries and how country-level factors moderate it. Bayesian methods were preferred for their ability to directly express uncertainty in estimates and their superior performance in handling random effects compared to frequentist approaches.

Analysis reveals that anti-science attitudes are moderately prevalent globally, but show considerable variation across countries. Latin American nations, in particular, exhibit higher levels of anti-science sentiment compared to many Anglo-American and East Asian countries. Contrary to prior research, the study does not find a universally negative global association between legacy media use and anti-science attitudes. In some countries (Thailand, Ethiopia, Nigeria, US), increased media use is associated with reduced anti-science attitudes. However, in others (Turkey, Bangladesh, Cyprus, Bolivia, Serbia, Peru), more frequent media use is linked to stronger anti-science attitudes. This highlights the importance of contextual factors. A substantial portion (14%) of the global variance in anti-science attitudes is attributable to country-level differences. The analysis further reveals that the prevalence of populist rhetoric significantly interacts with the relationship between media use and anti-science attitudes. In countries with high levels of populist rhetoric, increased legacy media consumption is associated with increased anti-science attitudes, while the opposite is true in countries with low levels of populist rhetoric. Press freedom and uncertainty avoidance do not show a significant moderating effect.

The findings challenge the simplistic assumption that legacy media uniformly reduce anti-science attitudes. The context-dependent nature of the media's effect suggests that the influence of media messages is not universal and is shaped by broader societal and political factors. The strong interaction between media use and populist rhetoric highlights the crucial role of political discourse in shaping public perceptions of science. In countries dominated by populist rhetoric which often frames scientific expertise as elitist and unreliable, legacy media may unintentionally amplify anti-science sentiment by giving a platform to populist viewpoints, or by failing to effectively counter these views. The study's findings on varying levels of anti-science attitudes across different countries suggest that socio-political and cultural factors play a significant role in shaping public trust in science. These factors interact with the message framing within the legacy media, thus further complicating the relationship between media use and public opinion.

This study demonstrates that the relationship between legacy media use and anti-science attitudes is far from straightforward, and that political context is a vital moderating factor. While legacy media can positively influence public perceptions of science in some contexts, their potential is significantly undermined in contexts where populist rhetoric is dominant. Further research should explore the role of social media, refine media use measurement by examining individual media types, and investigate the causal links between media consumption, political context, and anti-science attitudes. Qualitative research exploring public perceptions of science and different types of scientific expertise is also important. Ultimately, a nuanced understanding of these complex dynamics is crucial for effective science communication strategies.

The study relies on cross-sectional data, preventing causal inferences about the relationship between media use and anti-science attitudes. The measurement of legacy media use is broad, potentially obscuring differences in effects across different media types (newspapers vs. TV vs. radio). The country-level indicators used are imperfect proxies for complex phenomena; within-country variation is not explicitly addressed. Finally, the use of pre-existing survey items might introduce measurement limitations.