40 chapter 2 Method Participants and Procedure We used amazon MTurk to invite a random sample of 198 members of the United States population to participate in our study, with a compensation of $1.11 Participants who disagreed with the reality of climate change were directly sent directly to the end of the survey. After data inspection and cleaning,12 192 responses were retained (61% male; Mage = 36; SDage = 10.8; see more demographics in Supplementary Material). Again, this sample meets the minimum size appropriate for scale validation (Boateng et al., 2018). Measures All measures were assessed on a seven-point Likert-scale, from 1 = strongly disagree to 7 = strongly agree, unless otherwise specified.13 The survey started with the transilience scale, and ended with measures of well-being, psychological resilience, and demographics. All other measures were presented in randomised order. Most measures were like in Study 1, and we report below when and how they differed (see all items in Appendix B). For all scales, we computed mean scores and recoded reverse-coded items. Descriptives and reliability coefficients for the measures are provided in Table 2.10. Climate Change Transilience. Participants responded to the same 15 transilience items as in Study 2, presented in randomised order. The same three reverse-coded items were again not working well, so we report on the final 12 climate change transilience items (4 items for each component) used in Study 2. Climate Change Affect. Participants rated to what extent they experience certain emotions when they think about dealing with climate change risks, on a 7-point Likert scale from 1 = not at all to 7 = very strongly. Negative affect was assessed with five items (distressed, concerned, paralyzed, pessimistic, angry). Positive affect was assessed with two items (optimistic, hopeful). Climate Change Risk Perception. We included the same items as in Study 1, with the additional item ‘climate change poses a risk to my community’. 11 As in Study 1, we conducted another study at the same time to assess whether people perceive they can adapt to climate change collectively, that is, as a community. As the second study did not include any item on individual transilience, we do not discuss it here. 12 From the initial sample we removed 6 participants (3%) based on the following criteria. First, one duplicated IP address was removed. Participants who filled the survey within 2 minutes were removed (n = 2), as we did not consider this a realistic time to provide accurate responses (median completion time = 6.2 minutes). Finally, participants who failed an attention check asking them to select the response ‘agree’ were removed (n = 3). Analyses with the total sample led to similar results as those reported. 13 We also included measures to assess political and party preference, perceived consequences of climate change, common fate, and social identification. As these measures are not relevant for the purpose of the present study, they are not further discussed.
RkJQdWJsaXNoZXIy MTk4NDMw