32 chapter 2 Table 2.3. Results of the Haberman Procedure across all Studies Study 1 Study 2 Study 3 Study 4 PRMSS PRMST PRMSS PRMST PRMSS PRMST PRMSS PRMST Persistence 0.740 0.781 0.877 0.878 0.805 0.842 0.817 0.842 Adaptability 0.791 0.859 0.851 0.855 0.832 0.872 0.861 0.881 Transformability 0.835 0.850 0.785 0.793 0.823 0.841 0.800 0.825 Note. PRMST = proportional reduction in mean squared error based on total score PRMSS = proportional reduction in mean squared error based on subscale score Concurrent and Discriminant Validity We examined concurrent and discriminant validity of the transilience scale by computing bivariate correlations with relevant variables, using the custom function corstars in R (Bertolt, 2008). Higher transilience was associated with higher perceptions of climate change risks and higher negative affect about climate change, with a medium effect (i.e., above .24; Lovakov & Agadullina, 2021), indicating that, as expected, higher transilience does not imply perceiving climate change as less problematic. Interestingly, higher transilience is associated with higher perceived climate change risks. Furthermore, as expected, transilience was positively and significantly related to both self-efficacy and outcome efficacy (see Table 2.4), with a medium-to-large effect size (i.e., above .41; Lovakov & Agadullina, 2021). Yet, as expected, these correlations were far below the cut-off for construct overlap (i.e., around .85; Kenny, 2016). Predictive and Incremental Validity As expected, higher transilience was associated with more climate change adaptation behaviours and higher support for adaptation policies, with a medium effect (see Table 2.4). Yet, unexpectedly, transilience was not significantly related to adaptation intentions. Next, we found that transilience was still significantly correlated with adaptation behaviours and policy support when controlling for self- or outcome efficacy (see Table 2.5).
RkJQdWJsaXNoZXIy MTk4NDMw