Fig. 3: The effect of respondents’ inequality perceptions.

Linear random intercept regressions using cluster-robust standard errors. Depicted are point estimates of the effect of behavioral reductions on the willingness to adapt to the lifestyle described in the vignette (rated on scales from −5 to 5)(controlling for gender, age, education, income, climate change denial, climate change worry, climate attitudes, climate change knowledge, self-efficacy, group-efficacy, and the costs described in the vignette) and 95% confidence intervals for all respondents (left panel), and only for the subgroup of those respondents agreeing with the statement of the specific inequality prime they received (right panel). Estimation results are based on 3952 vignette ratings from 1321 respondents overall (number of respondents for each prime and agreeing to the priming statement is depicted in the legend). Percentages on the right indicate the amount of vignettes that would demand behavioral changes from the respondent (for all respondents/agreeing respondents). For the exact estimates, see also the regression tables in Appendix D3.