Expected climate change consequences and their role in explaining individual risk judgments

on humans (Level 5) and more likely to expect hardly any impacts. Those aged 65+, as com- pared to those aged 15–24, were less likely to mention environmental changes. Respondents aged 65+ or aged 55–64 were more likely to answer that climate change will have hardly any impact on their country compared to participants aged 15–24. The socio-demographic vari- ables had no significant effect on mentioning any of the other levels. Country was included as a control variable, and its effects in the regression complement those reported in the frequency table above (see Table 1). The socio-demographic- and country variables explained far more of the variance in category 1 and 6 as compared to the others, although only country had a signif- icant effect on mentioning attitudes, motives and goals. Predicting individual risk judgments from expected climate change consequences A closer look at the data indicated that respondents in the four countries differed regarding their level of worry about climate change. While about 42% of the respondents in France reported being ‘very worried’ or ‘extremely worried’ about climate change, the same was true for 30% in Germany, 29% in Norway, and 19% in the UK. There were also differences in expec- tations of how positive or negative the effects of climate change will be on one’s own country. The percentage of respondents reporting that they believe the effects will be ‘more negative than positive’ or ‘entirely negative’ was 75% in France, 79% in Germany, 57% in Norway, and 57% in

the UK. A complete overview of survey responses can be found elsewhere [26]. Means and stan- dard deviations of worry and risk evaluation for the full sample in each of the countries are depicted in Table 3; the two measures correlate moderately, r(3830) = .30, p < .001. The relationship between the level at which respondents identified climate change conse- quences for their country and individual risk judgments differed between the outcome mea- sures (see Table 4). Higher levels of worry were positively related to mentioning consequences regarding attitudes, motives and goals (Level 1), emissions and pollution (Level 3), environ- mental changes (Level 4), and impacts on humans (Level 5), and negatively related to hardly any impacts. Furthermore, worry was higher among women and those who placed themselves further left on the political spectrum. Risk evaluation was positively related to attitudes, motives and goals (Level 1), environmental changes (Level 4), and impacts on humans (Level 5), as well as positively related to gender (women). It was also negatively associated with men- tioning hardly any impacts, belonging to any of the three oldest age groups, and placing oneself further right on the political spectrum. These variables explained about 14% of the variance in worry about climate change and about 9% of the variance in risk evaluation. The number of mentioned categories, and the first sub-level depth index, were positively related to worry about climate change (see Table 5). None of the structural indices was related

to risk evaluation. In sum, the tested regression models explained about 9% of the variance in worry about climate change and about 8% of the variance in risk evaluation, with each model controlling for gender, age, political orientation, and country. Table 3. Means and standard deviations for worry about climate change and risk evaluation by country (weighted). Worry about climate change Risk evaluation M SD M SD UK 2.73 1.10 3.58 0.94 Norway 3.01 0.96 3.55 0.91 Germany 2.96 1.09 4.02 0.81 France 3.29 1.03 3.90 0.86 https://doi.org/10.1371/journal.pone.0281258.t003 PLOS ONE Expected climate change consequences and individual risk judgments PLOS ONE | https://doi.org/10.1371/journal.pone.0281258 February 15, 2023 7 / 15