Americans experience a false social reality by underestimating popular climate policy support by nearly half

The study investigates whether Americans accurately perceive national concern about climate change and support for major mitigation policies. The context is that threat recognition and coordinated action on climate change are influenced by perceptions of others’ beliefs (second-order beliefs) and social norms. Pluralistic ignorance—shared misperceptions of prevailing norms—can inhibit discussion, coordination, and support for policy by creating pressure to conform to perceived opposition. Given that polls indicate most Americans are concerned about climate change and support various policies, misperceiving public support could help explain limited federal action. The purpose is to quantify the prevalence, magnitude, and breadth of norm misperceptions across the U.S., policies, states, and demographics, and to explore potential sources (false consensus, availability via local norms, and media consumption).

Prior work indicates misperceptions around climate-related beliefs: people in the U.S., China, and Australia underestimate national belief in human-caused climate change. Student samples underestimate peers’ climate concern, and an online convenience sample underestimated support to regulate CO2 and concern about climate change. U.S. congressional staffers underestimated carbon pollution restriction popularity in their districts. In the northeastern U.S., convenience samples underestimated support for offshore wind. Theories such as the conservative bias suggest perceived public opinion often lags actual opinion, especially when policy has not caught up. False consensus effects lead individuals to overestimate agreement with their own views. Availability heuristics imply that local norms can disproportionately shape judgments of national opinion. Media coverage may misrepresent public opinion, with U.S. outlets historically overrepresenting climate denial and portraying conservatives as uniformly oppositional despite divided opinions within conservative electorates.

Design and data sources: Cross-sectional survey of U.S. adults via Ipsos eNation Omnibus, fielded April–May 2021. N = 6,119; stratified with oversampling of smaller states (cap of ~250 per larger state). Ipsos-provided post-stratification weights (gender, age, race/ethnicity, region, education) applied to national analyses. Quality controls removed speeders, straight-liners, and incomplete responses. Power analyses indicated >80% power to detect small national effects (d≈0.04) and small partisan subgroup effects (d≈0.07). Measures: Participants provided free-response estimates of the percentage of Americans who are at least somewhat concerned about climate change and who support: (1) a carbon tax, (2) 100% renewable electricity mandate, (3) siting renewables on public lands, and (4) the Green New Deal (policy descriptions matched YPCCC wording). The “worry” item used “climate change” vs YPCCC’s “global warming.” Participants also estimated these for their own state (concern; carbon tax). Actual opinion benchmarks: Contemporaneous nationally representative polling (e.g., YPCCC) provided actual support/concern levels at national (and where available partisan) and state levels for comparison. Analytic approach: For each item, perceived levels were compared to actual levels using one-sample t-tests against the constant of the benchmarked actual support/concern. Pluralistic ignorance was computed as (actual – perceived) per participant and aggregated for national, state, and subgroup analyses. Mixed-effects models with random intercepts (participant, item) tested differences between state vs national estimation accuracy and explored predictors. Exploratory factor analysis assessed structure of misperceptions across items. Exploratory mixed-effects regressions examined associations with demographics, state-level proxies of local norms (2020 presidential vote margin; logged climate/environmental protests per capita), and media consumption (weekly consumption of outlets; models controlled for demographics including political orientation, education, age, race, income).

- Widespread underestimation: 79–88% of respondents underestimated national climate concern or support for each policy. Average perceived prevalence for concern/support was 37–43%, whereas actual support/concern ranged 66–80%. - Climate worry: Americans estimated national worry at M_est = 43.3% vs actual M_real = 66%; t(6118)=70.9, d=0.92, P<0.001, 95% CI_diff [22.0, 23.3]. - Policies (national; Table 1): • Carbon tax: actual 67% vs perceived 36.6%; t=96.26, d=1.27, 95% CI_diff [29.7, 31.0]. • Siting renewables on public lands: actual 80% vs perceived 43.4%; t=113.10, d=1.48, 95% CI_diff [35.9, 37.2]. • 100% renewable electricity mandate: actual 66% vs perceived 39.5%; t=69.79, d=0.91, 95% CI_diff [21.9, 23.2]. • Green New Deal: actual 66% vs perceived 37.9%; t=89.81, d=1.17, 95% CI_diff [27.5, 28.7]. - Distributional patterns: Perceptions clustered around focal points near 25% and 50%, far below the more accurate 75%. - Breadth and structure: Misperceptions were similar across policies and concern; a single factor emerged in exploratory factor analysis (other eigenvalues <1), indicating a broad, non-specific misestimation of support for climate action. - State vs national estimation: No accuracy difference between estimating state vs national norms for overlapping items (carbon tax, worry); mixed model: t(21762) = -0.94, P = 0.350. - Partisan comparisons: All partisan groups (Democrats, Independents, Republicans) estimated national support/concern below 50%. Republicans provided lower estimates than Democrats by ~5–12%, Independents in between. Even estimates approximated actual Republican support still underestimated true national support for several policies. Underestimation magnitudes were largest for siting renewables (~35–40%), moderate for carbon tax and GND, and smallest (yet substantial) for 100% RE (~20–25%). - Regional/state variation: All states underestimated national worry and policy support; no state’s average error was <20% for policy support. Highest misperceptions tended to appear in southern Gulf states, but liberal states (e.g., CA, NY) also showed large underestimations. - Demographic variation (mixed-effects, weighted): No group was accurate; all were ≥20% off. Political orientation showed a gradient (very liberal ~22% underestimation vs very conservative ~33%). Race differences: white ~25% vs Black ~35% underestimation (others in between). Urban ~29% off; suburban ~26% (rural in between). Education differences were statistically significant but small in absolute terms (e.g., 12 years no diploma ~28% vs doctorate ~27%). - Local norms (state-level predictors; controlling for key demographics): More liberal state ideology and more protests per capita associated with slightly lower misperceptions: state political ideology b = -0.02, P = 0.019 (states highest Biden margin ~25.5% off vs highest Trump margin ~28.1% off); protests b = -0.47, t(39980)=2.32, P = 0.020 (most protests ~25.9% off vs fewest ~28.3% off). - Media consumption: Across outlets, consumers underestimated by ~25–30%. Relatively lower misperceptions among consumers of NPR, major national papers (e.g., New York Times), CNN, and ABC; relatively higher among Fox News, Breitbart, other conservative outlets, and also “other liberal outlets” (noted as least consumed at 15.5%).

Findings show a pervasive false social reality: Americans across parties, states, and demographics substantially underestimate national concern about climate change and support for major mitigation policies, often inverting the true supermajority into a perceived minority. This addresses the research question by demonstrating that pluralistic ignorance is both widespread and large in magnitude across multiple policy types and geographies. The results suggest several contributing factors: false consensus (conservatives, who are less supportive, underestimate more), conservative bias (perceptions lagging actual opinion), availability of local norms (more liberal states and those with more climate protests show slightly smaller misperceptions), and media consumption patterns (conservative outlet consumption associated with greater misperceptions). The societal relevance is significant: misperceived norms likely depress public discourse, organization, and willingness to advocate, while inflating perceived opposition and conformity pressures against climate action. Correcting these misperceptions could remove barriers to collective action and policy advancement.

This study documents a widespread and large-scale pluralistic ignorance regarding climate concern and support for key U.S. climate policies, amounting to a false social reality in which a true supermajority is perceived as a minority. The work contributes comprehensive, representative evidence across policies, states, parties, and demographics, and offers preliminary insights into correlates of misperception (political orientation, local norms, media consumption). Future research should prioritize experimental and causal designs to identify mechanisms (e.g., media effects, social norm updating) and develop interventions that correct norm misperceptions to facilitate climate discourse and support for transformative policies. Translating successful social norm interventions from other domains (e.g., alcohol use, tax compliance, anti-bullying) to the climate policy context is a promising direction.

- Correlational design limits causal inference about contributors such as media consumption and local norms; experimental evidence is needed to establish causality. - Benchmarking relies on contemporaneous polling for “actual” opinion; differences in question wording (e.g., “climate change” vs “global warming”) may affect exact levels of concern and thus the precise magnitude of misperceptions. - Although oversampling enabled state-level precision, residual sampling and weighting limitations may remain. - Exploratory analyses (media, local norms, demographics) may be subject to unmeasured confounding despite controls.