Climate change belief systems across political groups in the United States

purchase energy-efficient vehicles or solar panels” with response options on a four-point scale: (1) “Strongly oppose,” (2) “Somewhat oppose,” (3) “Somewhat support,” (4) “Strongly sup- port” (M = 3.06, SD = .89). Policy support funding renewable energy. Policy support funding renewable energy (hereafter, policy support fund) was measured with a question asking “How much do you sup- port or oppose the following policies?: Fund more research into renewable energy sources, such as solar and wind power” with response options on a four-point scale: (1) “Strongly oppose,” (2) “Somewhat oppose,” (3) “Somewhat support,” (4) “Strongly support” (M = 3.13, SD = .88). Policy support CO2. Policy support CO2 was measured with a question asking “How much do you support or oppose the following policies?: Regulate carbon dioxide (the primary greenhouse gas) as a pollutant” with response options on a four-point scale: (1) “Strongly oppose,” (2) “Somewhat oppose,” (3) “Somewhat support,” (4) “Strongly support” (M = 2.91, SD = .93). Political behavior. Political behavior was measured with a question asking “Over the past 12 months, how many times have you done each of the following?: Written letters, emailed, or phoned government officials about global warming” with response options on a five-point scale: (1) “Never,” (2) “Once,” (3) “A few times (2–3),” (4) “Several times (4–5),” (5) “Many times (6+)” (M = 1.23, SD = .71). Consumer behavior reward. Consumer behavior reward was measured with a question asking “Over the past 12 months, how many times have you done these things?: Rewarded

companies that are taking steps to reduce global warming by buying their products” with response options on a five-point scale: (1) “Never,” (2) “Once,” (3) “A few times (2–3),” (4) “Several times (4–5),” (5) “Many times (6+)” (M = 1.98, SD = 1.38). Consumer behavior punish. Consumer behavior punish was measured with a question asking “Over the past 12 months, how many times have you done these things?: Punished companies that are opposing steps to reduce global warming by NOT buying their products” with response options on a five-point scale: (1) “Never,” (2) “Once,” (3) “A few times (2–3),” (4) “Several times (4–5),” (5) “Many times (6+)” (M = 1.78, SD = 1.30). Political affiliation. Political affiliation was used to segment respondents into four groups. Political affiliation was measured using a two-step approach. Initially, participants were asked to indicate their identification as “Republican,” “Democrat,” “Independent,” “other,” or “no party/not interested in politics.” Subsequently, individuals who selected “Inde- pendent” or “other” were presented with a follow-up question inquiring whether they consid- ered themselves closer to the “Republican party,” “Democratic party,” or “Neither.” Respondents were classified as Republicans or Democrats if they initially identified as either party or, alternatively, did not initially identify with either party but expressed a closer affinity for one of the parties (i.e., “leaners”) in the follow-up question. The “Independent” category

excluded all such leaners. Participants who responded with “no party/not interested in poli- tics” were categorized in the “no party” group. Network measurements The R package bootnet [18] was used to estimate a climate change belief system consisting of fifteen nodes and several edges. The strength of each edge is based on a partial correlation between two nodes. Specifically, the method uses regularized partial correlation coefficients. Partial correlation coefficients offer advantages when assessing belief systems as they elucidate the relationship between two variables while controlling for influences from all other variables. Lasso regularization was used to eliminate potential spurious associations between nodes [19]. PLOS ONE Climate change belief system PLOS ONE | https://doi.org/10.1371/journal.pone.0300048 March 20, 2024 6 / 16