The impact of place and legacy framing on climate action: A lifespan approach

ages across the sample, participants were then divided chronologically into three age groups for analysis: young adults 18–35 (n = 327), middle-aged adults 36–52 (n = 324), and older adults 53–87 (n = 337). An estimate of target sample size was determined using results from similar studies run using mTurk workers to provide initial estimates of effect sizes for studies of this type, which suggested a small to medium effect size [29]. Thus, our final sample size of 988 participants (with about 330 for each age group) was deemed appropriate for our planned experimental design. An analysis of demographic information provided by participants revealed that 68% of our sample identified as female (n = 670), 32% identified as male (n = 316), and .2% identified as ‘other’ (n = 2). When asked about the highest level of education completed, .5% reported some high school (n = 5), 10% reported graduating high school (n = 100), 34% reported some college or technical school (n = 338), 37% reported graduating college or technical school (n = 369), and 18% reported receiving post-graduate education (n = 176). Politically, our sample was well divided, with 28% identifying as Republicans (n = 277), 39% as Democrats (n = 384), 28% as Independents (n = 272), and 5% identifying as ‘Other’ (n = 55). Participants were able to select as many ethnicities/races they identified with, and thus a total of 86% identified as Caucasian (n = 849), 9% as African American (n = 93), 5% as Hispanic (n = 45), 5% as Asian (n = 45), 3%

as Native American (n = 29), 1% as Other (n = 10), and .1% as Pacific Islander (n = 1). Finally, when asked whether participants identified as environmentalists, 55% said no (n = 546) and 45% said yes (n = 441). Measures Two measures were used as dependent variables, one was a scale measuring pro-environmen- tal behavioral intentions, and one was a proxy for pro-environmental behavior. The Pro-Envi- ronmental Behavioral Intentions Scale [26] is a 6-level likert scale with six items measuring the participant’s intent to make pro-environmental choices in the next month (Range = 0–6, M = 2.73, SD = 0.85, α = .76; e.g., “over the next month, how often do you intend to use public transportation or carpool?”). Following Zaval et al. (2015) [26], the second measure served as a proxy for real pro-environmental behavior, an open numerical item informing participants that they had been entered into a $10 dollar lottery, which asked how much of that money they would donate to the charity Trees for the Future if they won. Two measures were used as manipulation checks to ensure that each induction was evoking the responses intended. The Legacy Motive Scale [26] was used as a manipulation check for the Legacy induction. This 8-item scale measures how much participants care about their leg- acy (Range = 0–6, M = 4.19, SD = 1.05, α = .90; e.g., “I care about what future generations think about me”). The 5-item Abbreviated Biophilia Scale was used to measure participants’ love and appreciation for nature, and was used as a manipulation check for the Place induction

(Range = 0–7, M = 5.36, SD = 1.10, α = .97; e.g., “I feel joy just being in nature”; adapted from Perkins, 2010 [30]). Participants also provided extensive psychographic data that could potentially mediate the relationship between the inductions and environmental action. Two hazard belief questions were adapted from Miceli, Sotgiu, & Settanni (2008) [31], these 4-level likert questions (Range = 0–4), asked about the participant’s belief in the connection between natural hazards, climate change, and salient danger. A single-item measure of future orientation developed by Steinberg et al. (2015) [32] was included and asked participants to describe on a scale of 1–10 how easily life in the year 2050 comes to mind. An assessment of Future Self-Continuity was included asking participants to choose one picture (of seven available pictures) of two overlap- ping circles (each picture overlapping to increasingly higher degrees) best representing how Impact of framing on climate action PLOS ONE | https://doi.org/10.1371/journal.pone.0228963 February 25, 2020 4 / 17