How Does Political Ideology Shape Public Trust in Scientists?

Wheldon, Tallapragada, and Thompson (2025) examine whether political ideology is associated with trust in scientists as sources of cancer information in the United States. They analyze cross-sectional data from the 2024 Health Information National Trends Survey, a nationally representative survey of U.S. adults. They find that overall trust in scientists is high (86%), but it declines as respondents become more politically conservative. Each one-point shift toward conservatism is linked to about a 25% reduction in the odds of reporting high trust (adjusted OR ≈ 0.75). Estimated probabilities of high trust range from roughly 94% among liberals to about 71% among very conservative individuals. 

Why This Article Was Selected for The Policy Scientist

This article addresses a foundational issue in public health: whether trust in scientists varies systematically across political ideology when individuals evaluate cancer information. The topic has broad policy relevance because trust shapes receptivity to prevention guidance, screening recommendations, and treatment innovations. The study is timely given heightened polarization and persistent challenges in science communication. The authors have contributed extensively to research on health communication and trust, and this analysis extends that body of work using recent national data. Drawing on the Health Information National Trends Survey enhances data quality through probability sampling and weighting, supporting population-level inference. The statistical approach—survey-weighted logistic regression—is appropriate for the cross-sectional design, though future work could be strengthened through causal inference strategies or experimental designs to better isolate mechanisms. Generalizability to other jurisdictions is plausible where similar media environments and political dynamics exist. 

Full Citation and Link to Article

Wheldon, C. W., Tallapragada, M., & Thompson, E. L. (2025). Public trust in scientists for cancer information across political ideologies in the US. JAMA Network Open, 8(12), e2546818. https://doi.org/10.1001/jamanetworkopen.2025.46818 

Central Research Question

This study investigates whether political ideology is associated with public trust in scientists as sources of cancer information in the United States. The authors frame the inquiry around a specific hypothesis: individuals with more conservative political orientations will report lower levels of trust in scientists compared with individuals holding more liberal views. The question is positioned within a broader concern about the functioning of science communication in a politically polarized environment. Trust in scientific authorities is treated as a critical intermediate factor that can influence how people interpret risk information, respond to prevention guidance, and evaluate medical recommendations. By focusing explicitly on cancer information—a domain that typically enjoys strong bipartisan salience—the study tests whether ideological divides extend into areas traditionally viewed as less politically contested.

Previous Literature

Prior research consistently shows that scientists rank among the most trusted professional groups in the United States, particularly on issues related to health and medicine. However, emerging evidence suggests that generalized trust in science has become increasingly stratified along political lines. Studies conducted over the past decade document widening ideological gaps in attitudes toward vaccination, climate science, and public health guidance. This literature indicates that conservatives, on average, express greater skepticism toward scientific institutions and regulatory authorities, although trust levels often remain high in absolute terms. The authors situate their analysis within this evolving scholarship by examining trust not as a broad construct but as domain-specific confidence in scientists providing cancer-related information. The study builds conceptually on earlier work using the Health Information National Trends Survey (HINTS), which established baseline patterns of trust in health information sources and highlighted the growing influence of digital media ecosystems. The article also connects indirectly to theoretical perspectives such as source credibility, trust transfer, and motivated reasoning, which explain why perceptions of messengers can shape responses to evidence-based messages. By updating this line of inquiry with recent national data collected during a period of elevated polarization, the study extends understanding of how ideology intersects with trust in a critical public health context.

Data

The analysis draws on data from the 2024 Health Information National Trends Survey, a nationally representative survey administered by the National Cancer Institute. HINTS employs a stratified random sampling design targeting noninstitutionalized U.S. adults aged 18 years or older. Data collection occurred between March and September 2024 using mailed questionnaires. The initial respondent pool included 7,278 individuals, yielding a response rate of 27.3%. After excluding cases with missing information on key variables—trust in scientists and political ideology—the final analytic sample consisted of 6,260 adults. The dataset includes survey weights designed to correct for differential probabilities of selection, nonresponse, and noncoverage, thereby supporting inference to the broader U.S. adult population. The mean respondent age was approximately 48 years, and the sample was demographically diverse across gender, educational attainment, race, and ethnicity. Trust in scientists was measured using a four-point scale and dichotomized into high versus low trust categories. Political ideology was captured using a seven-point scale ranging from very liberal to very conservative. The dataset also contains variables representing socioeconomic status, marital status, personal and family cancer history, and trust in doctors, allowing adjustment for potential confounders. The use of HINTS strengthens the study’s empirical foundation due to its probability-based design, standardized measurement protocols, and established use in peer-reviewed health communication research.

Methods

The authors conduct a cross-sectional secondary analysis using survey-weighted logistic regression models. The dependent variable is a binary indicator of high trust in scientists as sources of cancer information. The principal independent variable is political ideology measured on a continuous seven-point scale. The statistical approach accounts for the complex sampling design through weighting and design-adjusted standard errors. The modeling strategy proceeds in stages, beginning with bivariable analyses followed by multivariable models incorporating control variables that exhibit associations with both trust and ideology. These controls include age, education, marital status, race, family cancer history, and trust in doctors. Missing data for selected demographic variables are addressed using hot-deck imputation. The authors estimate adjusted odds ratios and corresponding 95% confidence intervals. They further compute survey-adjusted predicted probabilities of high trust across ideological categories. Between-group differences are evaluated using design-based omnibus F tests and Šidák-adjusted pairwise comparisons. Sensitivity analyses explore alternative codings of both the trust outcome and the ideology variable, including quadratic specifications and categorical groupings. This methodological framework is consistent with best practices for analyzing nationally representative survey data. While the design does not permit causal inference, the analytic strategy is appropriate for estimating associations and identifying gradients across the ideological spectrum.

Findings/Size Effects

The study reports that trust in scientists for cancer information is high overall, with 86% of respondents indicating “some” or “a lot” of trust. Despite this elevated baseline, political ideology exhibits a statistically significant inverse association with trust. In bivariable analyses, each one-point shift toward greater conservatism is associated with a 27% decrease in the odds of reporting high trust (OR ≈ 0.73). After adjusting for demographic and attitudinal controls, the magnitude of the association remains substantial (adjusted OR ≈ 0.75), indicating a roughly 25% reduction in the odds of high trust per ideological step toward conservatism. Predicted probabilities illustrate a clear gradient: high trust is estimated at approximately 94% among liberals and declines to about 71% among very conservative respondents. The between-group differences are statistically significant. Additional findings show that higher educational attainment is positively associated with trust, while older age is modestly associated with lower trust. Trust in doctors displays a particularly strong relationship with trust in scientists (multivariable OR > 17), suggesting potential overlap or reinforcement between perceptions of medical and scientific authorities. Sensitivity analyses yield directionally consistent results, supporting the robustness of the observed ideological gradient. These size effects indicate that ideology functions as a meaningful correlate of trust even within a domain characterized by strong overall confidence.

Conclusion

The authors conclude that public trust in scientists as providers of cancer information remains resilient but is not uniform across political ideology. The presence of a pronounced ideological gradient implies that polarization influences perceptions even in areas of broadly shared concern. The study contributes to the literature by documenting this pattern using recent nationally representative data and by focusing specifically on cancer communication. The quality of the HINTS dataset supports population-level generalization within the United States, and the consistency of results across model specifications strengthens confidence in the findings. However, the cross-sectional design limits causal interpretation. Future research employing longitudinal designs, causal inference strategies, or experimental methods could clarify whether ideological differences reflect stable predispositions, exposure to distinct information environments, or reactions to specific messaging frames. The findings are potentially informative for other jurisdictions exhibiting similar political and media dynamics. Overall, the article provides empirical evidence that trust in scientific messengers varies systematically with ideology while remaining high in aggregate, highlighting the complexity of contemporary science communication.