Climate change belief systems across political groups in the United States

and addresses. The respondents completed the survey questionnaires in a web-based environ- ment. Individuals that did not have internet access were provided with computers and internet access. This study utilized fifteen waves of survey data that each include fifteen common variables that were used to construct the climate change belief systems. The total number of respondents in the selected survey datasets was 16,949. We excluded 207 respondents who refused to indi- cate their political affiliation in the survey. These respondents are distinct from those with no party affiliation, who were retained in the sample. The former did not provide any response to the party affiliation question, while the latter explicitly said they were not affiliated with any party. Therefore, in the main analysis, there were 16,742 respondents consisting of 6,255 Republicans, 6,823 Democrats, 1,821 independent/other, and 1,843 categorized as having no party affiliation or interest in politics. The average number of respondents for each wave was 1,116 (SD = 164.99). Survey measurements We selected fifteen variables from the survey datasets to construct network maps of the belief systems. These variables were chosen based on four criteria. First, we considered variables of significance in climate change communication research. Second, we included diverse psycho- logical and behavioral elements, which could conceivably capture an adequate representation of a climate change belief system. Third, we endeavored to provide balance by including differ-

ent types of psychological and behavioral elements within the belief system, thereby reducing any bias (e.g., an overestimated strength centrality due to strong relationships between the same type of elements). Finally, we selected variables that were included in a sufficient number of survey waves. Table 1 offers a summary of the items. Global warming (GW, hereafter) happening. To measure GW happening, we provided a short definition of global warming and then asked “Do you believe that global warming is happening?” with response options on a three-point scale: (1) “No,” (2) “Don’t know,” and (3) “Yes” (M = 2.49, SD = .77). GW human cause. GW human cause was measured with a question asking “Assuming global warming is happening, do you think it is. . .” with response options on a four-point scale: (1) “Neither because global warming isn’t happening,” (2) “Caused mostly by natural changes in the environment,” (3) “Caused by human activities and natural changes,” and (4) “Caused mostly by human activities” (M = 3.03, SD = 1.07). GW consensus. GW consensus was measured by asking “Which comes closest to your own view?” with response options on a three-point scale: (1) “Most scientists think global warming is not happening,” (2) “There is a lot of disagreement among scientists about whether or not global warming is happening,” and (3) “Most scientists think global warming is happen- ing” (M = 2.49, SD = .58). GW worry. GW worry was measured with a question asking “How worried are you about

global warming?” with response options on a four-point scale: (1) “Not at all worried,” (2) “Not very worried,” (3) “Somewhat worried,” and (4) “Very worried” (M = 2.52, SD = .96). Collective efficacy. Collective efficacy was measured with a question asking “Which of the following statements comes closest to your view?” with response options on a four-point scale: (1) “Humans can’t reduce global warming, even if it is happening,” (2) “Humans could reduce global warming, but people aren’t willing to change their behavior, so we’re not going to,” (3) “Humans could reduce global warming, but it’s unclear at this point whether we will do what’s needed,” and (4) “Humans can reduce global warming, and we are going to do so successfully” (M = 2.45, SD = .85). PLOS ONE Climate change belief system PLOS ONE | https://doi.org/10.1371/journal.pone.0300048 March 20, 2024 4 / 16