Toward a unified understanding of climate anxiety: Examining measures of climate and eco-anxiety

PLOS One | https://doi.org/10.1371/journal.pone.0339562 December 29, 2025 6 / 19 common indices (e.g., Cronbach’s alpha [46]). For climate and eco-anxiety, reliability is reported for the best measurement model among the tested competing models. Risk perception of climate change–related health risks. Risk perception of climate change–related health risks was assessed using 18 items [41,42]. Participants rated nine specific health risks linked to climate change (e.g., extreme weather events and increasing heat and heat waves) on two dimensions: perceived probability (“Please indicate in each case how likely these consequences of climate change are to occur in your lifetime“; 1 = highly unlikely to 7 = highly likely) and perceived severity (“Now please indicate in each case how dangerous you think these consequences of climate change are for your life“; 1 = harmless to 7 = dangerous). Since probability and severity ratings of the same health risks were highly correlated (despite other theoretical concep- tualizations [47]), we modeled nine parcels by calculating the means of the two variables per health risk. Reliability for this model was high, with factor saturation ⍵ = .914 [46]. Climate change–related fatigue. To capture behavioral and informational fatigue, we used the following two items: “I am losing my drive to fight climate change” and “I am tired of hearing about climate change.” In the Pandemic Fatigue Scale (PFS [35]), from which the items originate, the two items loaded highest on the factors of behavioral and

informational fatigue and were adapted to the context of climate change. Answers were rated on a seven-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). In our study, both items, loadings constrained to equality, were modeled as one factor, showing a reliability of ⍵ = .652. Self-efficacy to show individual emission-reducing behavior. Self-efficacy regarding individual climate protection behaviors was measured using four items assessing the difficulty or ease of showing high-impact individual behaviors that reduce emissions (e.g., “Refraining from private air travel”; see [42]; one item per sector of mobility, housing and energy, food, and other consumption behavior). Participants rated the perceived difficulty of showing each behavior on a seven- point Likert-type scale ranging from 1 (extremely difficult) to 7 (extremely easy). The four items were modeled as one factor and showed a reliability of ⍵ = .564 in our study. Readiness for climate action. The readiness for climate action is defined as a latent trait indicated by individual high-emission-related behavior, acceptance of climate change mitigation policies, and political participation [41]. The three indicators are measured as follows. Individual behavior was operationalized using the 21 behaviors most impactful in terms of individual CO2 emissions in Germany and was rated on impact-based response scales. Behaviors were sampled from the domains of mobility, hous- ing and energy, food, and other consumption behavior (e.g., food: “I buy organic food”). Most answers were collected

as behavioral frequency ratings ranging from 1 (never) to 7 (always), but there were some exceptions—for example, for annual distance driven by car (collected as kilometers), room temperature in winter (in degrees Celsius), duration of a shower (in minutes), and type of energy supply. To calculate the total score [42], items are weighted and thus approximate the actual footprint as measured using a CO2 calculator [48]. Policy acceptance was measured using 12 items asking participants about their agreement with policies covering guid- ing principles and policies pertaining to the mobility, housing, energy, and food sectors (e.g., food: “Subsidies in agriculture should only be available for climate-friendly farms”). Germany’s Citizen Climate Council recommendations constitute the basis for the policies (for details on the selection process, see [42]). Answers were rated on a scale from 1 (strongly dis- agree) to 7 (strongly agree). Political participation was operationalized using 11 items [42] measuring conventional participation, activism, and peer group–related participation behavior (e.g., conventional: “I sign petitions for more climate protection”) on a scale ranging from 1 (never) to 7 (always). Items were adapted from several existing scales measuring political participation and contex- tualized to the topic of climate change [42]. In sum, the readiness for climate action is modeled in a bifactor model with a general readiness factor and specific

(orthogonal) factors for each indicator group (individual behavior, policy acceptance, and political participation [41]). In the