Influence of Carbon Removal on Mitigation Support

INFLUENCE OF CARBON REMOVAL ON MITIGATION SUPPORT

Electronic Supplementary Material

A. Experimental Condition Vignettes

All conditions

CDR Conditions (all conditions except control condition)

REF Condition

BECCS Condition

DACS Condition

INFLUENCE OF CARBON REMOVAL ON MITIGATION SUPPORT

B. PROCESS Model 1: Testing whether ideology moderates the effect of condition on perceived threat (H3)

Table. S1 Results of PROCESS Model 1 to test whether political ideology moderates the effect of condition on perceived threat

C. ANCOVA: Exploratory analysis testing direct effects of conditions on mitigation policy support

We further tested these direct effects of learning about CDR on support for mitigation (not controlling for perceived threat) using an analysis of covariance (ANCOVA). Age, gender, education, and political ideology were included as covariates. A significant main effect was found for treatment condition, F(4, 975) = 3.38, p < .01). Planned contrasts revealed that reading about reforestation (MREF = 5.65, SE = .08) increased support for mitigation policies compared to reading about CDR in general (MGEN = 5.34, SE = .07, p < .01), BECCS (MBECCS= 5.31, SE = .08, p < .01), or direct air capture (MDAC= 5.40, SE = .08, p < .05). No significant differences were found, however, between the Control (MCONTROL = 5.51, SE = .08) and any of the specific CDR conditions.

D. PROCESS Model 59: Exploratory analysis to test whether ideology moderates the effect of condition on perceived threat and on mitigation policy support

Upon a reviewer’s suggestion, we ran PROCESS Model 59 (Figure S1), which tests whether ideology moderates each link in the causal model, including the direct relationship between CDR condition and mitigation policy support.

Figure S1. PROCESS Model 59

The first step in Model 59 is to test whether ideology moderates the effect of condition on perceived threat. The results are identical to what was previously reported in Table S1 and confirm that ideology does not moderate the effect of learning about CDR on threat. The full moderated-mediation model reveals, however, two modest interactions between CDR condition and political ideology on mitigation policy support. As compared to the control condition, conservatives who read about reforestation or direct air capture were more likely to express support for mitigation policies (see Table S3). No effects were found for moderates or liberals. In line with Model 14, the interaction between perceived climate change threat and political ideology on MPS remained significant. The results also partially replicate the negative indirect effects previously reported in the paper for GEN, BECCS, and DACS. In Model 59, though, the indirect effects are not significant for all political ideology groups (see Table S3).

Table. S2 Results of moderated-mediation analysis (PROCESS Model 59) where political ideology moderates each link in the model

Table. S3 Conditional direct and indirect effects (through perceived climate change threat) of learning about CDR on mitigation support at different levels of political ideology (mean and + 1 SD from the mean), when controlling for Condition x Ideology on Threat (PROCESSModel 59).

2a. CDR General 2b. Reforestation

2c. BECC 2d. DAC

Figure S2. Effect of climate change threat on mitigation support for each CDR condition, as moderated by political ideology.