Voelkel, J. G., Ashokkumar, A., Abeles, A. T., Crawford, J., Fuller, K., Redekopp, C., Bongiorno, R., Campbell, T. H., Ecker, U. K. H., Feinberg, M., Hart, P. S., Hornsey, M., Jost, J., Kay, A., Leiserowitz, A., Lewandowsky, S., Maibach, E., Nisbet, E., Pidgeon, N., Spence, A., van der Linden, S., Wolsko, C. V., Willenbring, J. K., neil malhotra, & Willer, R. (2026). A registered report megastudy on the persuasiveness of the most-cited climate messages. Nat. Clim. Chang., 1–12. https://doi.org/10.31235/osf.io/xwceg_v1
Summary
This registered-report megastudy conducts the first systematic, head-to-head comparison of the ten most-cited climate change messaging strategies, testing whether they persuade Americans to hold stronger pro-environmental attitudes and to engage in costly pro-environmental behaviour. Preceded by five pilot replication studies (N=3,216) that largely failed to reproduce three canonical strategies, the main study (N=13,544) found that six messages reliably shifted multiple attitudinal outcomes — but only by 1–4 percentage points, with almost no partisan heterogeneity, and with no positive effect on actual donations. The authors conclude that short-form messages can nudge attitudes but have clear limits for inducing behaviour, and they use megastudy methodology as a corrective to the field’s temporal-validity, power, and comparability problems.
Key Contributions
- First commensurate, head-to-head test of the 10 most-cited climate messaging strategies under identical measures and a large national US sample.
- A preregistered, Stage 1-accepted replication-and-extension that updates the temporal validity of canonical climate communication findings.
- Demonstrates the megastudy design for comparing relative efficacy and detecting small effects.
- Challenges the assumption that Democrats and Republicans need distinct, audience-targeted framings.
- Documents a robust attitude–behaviour divergence, including a scientific-consensus backfire on donations.
- Publicly releases a list of 157 climate messaging papers and their selection metrics.
Methods
Researchers systematically identified 157 climate messaging papers (via meta-analyses, reviews, expert input, and journal searches) and ranked them by three citation-based metrics to select the 10 most-cited strategies. Participants in the main megastudy (national non-probability sample matched to demographic benchmarks) were randomly assigned to one of 10 treatment conditions or a pooled control, with “conceptual” replications permitting original authors to revise treatments while preserving theoretical content. Four primary outcomes (belief, concern, general policy support, political behavioural intentions) plus an incentivized donation behaviour were measured pre- and post-treatment. Analyses used OLS regressions controlling for baseline outcomes, with confirmatory full-sample and within-partisan tests, exploratory relative-efficacy comparisons, Bayesian analyses, and moderation/mediation explorations, powered to detect d ≥ 0.10.
Findings
- Six messages increased belief in climate change; Scientific Consensus 2 had the largest effect (~3.12 pp, ~14% of the Democrat–Republican gap).
- Seven messages raised concern (largest: Purity Framing, 2.47 pp); Warmth Framing showed a small backfire (−0.76 pp).
- Six messages increased general policy support (largest: Purity Framing, 1.65 pp); eight raised political behavioural intentions (largest: System Preservation Framing, 3.67 pp).
- No treatment increased donations; Scientific Consensus 2 backfired (−3.95 pp) and Gains Framing reduced donations among Republicans.
- Effects were highly similar across party lines (only 12 of 90 partisan interactions significant), undermining audience-targeting claims.
- Attitude/intention effects were strongly intercorrelated but uncorrelated (or negatively related) with donation effects — a clear attitude–behaviour gap.
- Pre-treatment concern was the most consistent moderator (26/90 interactions); effective messages typically shifted multiple candidate mediators at once, obscuring mechanisms.
- Megastudy effects exceeded pilot effects, plausibly from greater power, baseline controls, and author-revised treatments (revision extent correlated r=0.29 with effect size).
Connections
This paper sits alongside other large-scale and replication-minded work on climate and misinformation message interventions; its scientific-consensus findings and attitude–behaviour focus connect naturally to van-der-Linden2026-jt and to the broader debate about intervention efficacy raised in Spampatti2026-kx. Its megastudy framing and skepticism toward one-size-fits-partisan messaging also relate methodologically to comparative-efficacy studies such as Szabo2026-rd and Costello2024-bg. The remaining papers under this topic do not appear to bear a direct intellectual relationship based on the information available.
Podcast
A research-radio episode discusses this paper: 🎧 MP3 · Spotify · Apple Podcasts