Climate change knowledge influences attitude to mitigation via efficacy beliefs

Theories proposing climate change apathy is explained by inadequate knowledge do not account for why many informed and concerned Americans fail to act. While correlations between knowledge, efficacy for climate change, and attitude to mitigation have been observed, few studies have examined efficacy for climate change as a mediator. This study aimed to investigate the influence of specific climate change knowledge on attitude to mitigation via efficacy beliefs. A cross‐sectional survey of 205 US adults recruited from Amazon's Mechanical Turk assessed participants’ climate change knowledge, efficacy for climate change, and attitude to mitigation. Indirect effects of self‐efficacy for climate change were observed in three mediation models, suggesting efficacy for climate change explains some of the relationship between specific climate change knowledge and attitude to mitigation. The findings suggest risk communication can motivate pro‐environmental attitudes with interventions that deliver information about climate change and develop efficacy for mitigation behavior.


INTRODUCTION
Human consumption behavior since preindustrial times has resulted in an unsustainable increase in greenhouse gas emissions and global temperature rise (Allen et al., 2018). Without mitigation, global warming will result in severe weather events like drought, cyclones, and rising sea levels. The impact on land biodiversity and ecosystems will result in species loss and food insecurity, and threaten water supply, economic growth, and human health (Allen et al., 2018). A large proportion of the US adult population express concern for the effects of global warming, and belief that human behavior can mitigate them (Leiserowitz et al., 2015). Despite this concern, the United States is one of the largest emitters of greenhouse gas (Boden & Andres, 2014;Muntean et al., 2018), and its emissions continue trending upward (Muntean et al., 2018).
Explanations for the discrepancy between concern and apathy range across several variables, including knowledge deficits and social-psychological factors (Bamberg & Möser, estimate of the successful impact of their actions (Bandura, 1977). Although an understanding of effective action is important to behavior change, self and response efficacy motivates and sustains the effort. SCT suggests that self and response efficacy mediates the relationship between knowledge and behavior change (Bandura, 1982), a relationship seen in health and education research (Amin et al., 2018;Rimal, 2000). An SCT framework of pro-environmental behavior suggests successful mitigators possess knowledge of the rationale for pro-environmental behaviors and effective actions, perceived capacity to act, and conviction in their success.

Climate change knowledge
Inconsistent support for the influence of climate change knowledge over attitudes to climate change has fueled debate largely in favor of social-psychological variables. Early information-deficit theories have been challenged by studies of social-psychological predictors including political ideology and personal values (Ajzen et al., 2011;Hornsey et al., 2016;Kahan et al., 2012). Despite the considerable attention given to social−psychological models, researchers continue to identify climate change knowledge as an important predictor of concern and attitude to mitigation (Bamberg & Möser, 2007;Hines et al., 1987;Kollmuss & Agyeman, 2002;Siegel et al., 2018). Additional support for informing the public come from evidence, suggesting significant numbers of well-intentioned U.S. adults may have misperceptions about effective mitigation strategies Read et al., 1994;Reynolds et al., 2010). The inconsistent results for climate change knowledge as a predictor to attitude may be related to conceptualization and measurement of the knowledge variable. Two studies claiming no relationship between climate change knowledge and policy support (Dietz et al., 2007;Rosentrater et al., 2013) measured personal beliefs about the effects of climate change. When measured objectively, climate change knowledge appears to demonstrate some predictive value. A comparison of factual versus false climate change knowledge among 1218 US adults found factual knowledge was a stronger predictor of proenvironmental intentions (Bord, 2000). Similarly, a meta-analysis of 25 polls and 171 studies found that factual climate change knowledge was a stronger predictor of attitude to mitigation than self-rated knowledge (Hornsey et al., 2016). A climate change knowledge intervention also found knowledge had significant and lasting effects on acceptance and attitude to mitigation (Ranney & Clark, 2016). In particular, climate change causes knowledge has been identified as a predictor of acceptance (Guy et al., 2014) and concern for climate change (Shi et al., 2015;Tobler et al., 2012). It also predicts attitude to mitigation and proenvironmental behavioral intentions (Bord, 2000;O' Connor et al., 1999;Shi et al., 2015).
Several explanations of the mechanisms of influence of knowledge over climate change attitude to mitigation have been proposed. It has been suggested detailed climate change knowledge challenges decision-making heuristics informed by cultural worldview and political ideology (Ranney & Clark, 2016). This view was supported in a series of experiments on political decision making that tested if challenging participants' policy knowledge would influence their position (Fernbach et al., 2013). Asking participants to provide a mechanistic explanation of a policy resulted in a shift toward a more moderate position, whereas asking for support reasons had little effect. The findings support the position that people deem their climate change knowledge to be accurate until detailed information challenges the perception (Ranney & Clark, 2016).
Evidence is emerging that efficacy for climate change beliefs explain some of the relationship between climate change knowledge and attitude to mitigation. A climate change cause-and-effect knowledge intervention among 1975 US university students and science-learning-center guests indicated climate change knowledge can increase self-and response efficacy for climate change (Geiger et al., 2017). The intervention group participants who reported greater selfefficacy also reported increased pro-environmental attitude, specifically, willingness to discuss climate change with peers. A second experiment indicated that response efficacy for climate change mediated the relationship between the intervention and pro-environmental behavior. Additional evidence for the mediating role of efficacy for climate change beliefs has been observed in a test of social-psychological predictors of climate change behavior. A survey of a social model of climate change engagement among 1000 US adults showed self-efficacy for climate change uniquely mediated the relationship between climate change knowledge and mitigation behavior (Estrada et al., 2017). Although these researchers were studying different theoretical models (information vs social-psychological), they both provide support for efficacy for climate change as a mediator between climate change knowledge and attitude to mitigation.

The current study
While specific climate change knowledge and efficacy for climate change are correlated and have also been individually identified as predictors of attitude to mitigation, there has been little exploration of how knowledge and efficacy influence attitude to mitigation. An investigation of potential interrelationships can further our understanding of how information influences attitudes to mitigation for the purposes of developing effective risk communication. The aim of the current study is to investigate the influence of efficacy for climate change beliefs in the relationship between climate change knowledge and attitude to mitigation among US adults. It is hypothesized that efficacy for climate change beliefs will mediate the relationship between climate change knowledge and attitude to mitigation. Further analysis will investigate specific effects of self and response efficacy for climate change in the relationships between individual climate change knowledge facets and attitude to mitigation behavior and policies.

Sample
A total of 205 participants were recruited through Amazon's Mechanical Turk (MTurk) online open marketplace. MTurk provides access to a large, diverse pool of US adults who can supply quality data with acceptable power and reliability (Buhrmester et al., 2011;Buhrmester et al., 2018). The sample consisted of 43% men (n = 90), 54% women (n = 111), and 2% (n = 4) who reported their gender as nonbinary or not listed. The age range was 18-75 years, mean age was 37 years (SD = 14). See further demographic information provided in Table 1 including a comparison between the current MTurk sample and the US adult population.

Climate change knowledge
Climate change knowledge was measured using items from scales developed and tested by Tobler et al. (2012) and used in additional studies of climate change knowledge and attitudes (Shi et al., 2015(Shi et al., , 2016. The scale measures were climate change causes knowledge, climate change physical knowledge, and climate change consequences knowledge. The 12 selected items were previously tested on a sample that included a group of US adults (Shi et al., 2016). The items ranged in difficulty, representing information reported by media outlets, to expert knowledge (Tobler et al., 2012). All items presented statements and each facet included incorrect items representing common misperceptions of climate change. The climate change causes knowledge facet assessed participants knowledge of the causes of climate change, including the role of carbon dioxide (CO 2 ), temperature increases, and human involvement. The facet scale included four questions, such as "Climate change is mainly caused by human activities." The climate change physical knowledge facet assessed participants knowledge of the physical characteristics of CO 2 , including CO 2 production and greenhouse gas effects. The facet scale included four questions, for example: "Burning oil produces CO 2 ." The climate change consequences knowledge facet assessed participants TA B L E 2 Climate change knowledge items, response frequencies, scale and item coefficients, and reliabilities knowledge of climate change consequences, including its impact on weather and sea levels. The facet scale included four questions prefaced with "for the next decades, most climate scientists expect…," followed by a statement, for example, "… an increase in extreme events such as droughts, floods, and storms." For all knowledge items, participants were asked to indicate on a 3-point scale if they thought the statement was true (1), wrong (2), or I do not know (3). Prior to analysis, responses to incorrect statements were reverse-coded and all item responses were converted to binary scores (0 = incorrect and I do not know responses, 1 = correct responses). Facet scale reliability was estimated with the Molenaar Sijtsma reliability statistic (Molenaar & Sijtsma, 1984). It should be noted that reliability estimates for dichotomous item sets are typically lower than multicategory items (Molenaar & Sijtsma, 1988), and is influenced by the number of scale items and sample size (Sijtsma & Molenaar, 1987). Knowledge for climate change items, scale reliability, and response frequencies are included in Table 2.

2.2.2
Efficacy for climate change Efficacy for climate change was measured using items developed and tested by Bostrom et al. (2019). The scale measures two distinct facets of perceived efficacy for climate change, including self-efficacy for performing a mitigation behavior and the response efficacy (expected outcome) for performing a mitigation behavior. The self-efficacy for climate change facet asked participants to rate the degree of difficulty to perform a climate change behavior. The facet scale lists five highly effective and common mitigation behaviors, such as "reducing household energy use by 20%" and "talking about global warming with others." Participants were asked to indicate how easy or hard it would be for them to perform the behavior on a 7-point scale (1 = extremely hard, 7 = extremely easy mitigation policies. The attitude to mitigation behaviors facet asked participants to rate their willingness to adopt certain climate change mitigation behaviors. The facet scale listed 11 mitigation behaviors, such as "walking, cycling, or taking public transportation instead of driving cars." Participants were asked to indicate how willing they were to adopt the behaviors using a 7-point scale (1 = I am not willing at all, 7 = I am already doing this). The attitude to mitigation policies facet asked participants to rate their support for certain climate change mitigation policies. The facet scale listed nine policy measures, such as a "CO 2 tax on gasoline and diesel fuel." Participants were asked to indicate how they felt about the policy measures on a 6-point scale (1 = not acceptable, 6 = totally acceptable). Attitude to mitigation items, descriptive statistics, and response distributions is included in Table 4.

Procedure
Ethical approval was obtained from the Monash University Human Research Ethics Committee. Participants were invited to complete the survey online and they were provided with an explanatory statement describing the study purpose. The survey began with demographic questions, followed by the efficacy for climate change scales, attitude to climate change mitigation scales, and then climate change knowledge scales. To discourage cheating, participants were asked to avoid using the internet to search for correct answers prior to accessing the climate change knowledge scale (Goodman et al., 2013). An incentive of $1.75 USD was paid to qualifying respondents.

Correlations
Significant relationships were observed between several study variables. All variable facets were interrelated within their domain. Small positive effects between climate change causes knowledge and physical knowledge (r = 0.17, p = 0.02), and medium positive effects between causes knowledge and consequences knowledge (r = 0.43, p = < 0.001) were observed. There was a large positive correlation between attitude to mitigation behaviors and mitigation policies (r = 0.64, p < 0.001). There was also a large positive correlation between self-efficacy and response efficacy for climate change (r = 0.5, p < 0.001). Climate change causes knowledge predicted all other study variables, with varying effect sizes. Correlations between all variables are reported in Table 5.

Mediation analyses
Mediation models were used to test the indirect effects of efficacy for climate change on the relationship between climate change knowledge and attitude to mitigation. The analysis was conducted using the SPSS PROCESS macro tool V3.5 (Hayes, 2012), with 5000 bootstrap samples for percentile bootstrap confidence intervals at a 95% confidence level. To test the hypothesis, we first examined mediation at the domain level with total climate change knowledge as the independent variable, attitudes to mitigation policies and behavior as the dependent variable, and self and response efficacy beliefs as the mediator. A mediation analysis was conducted to examine whether the climate change knowledge domain predicted attitude to mitigation through efficacy beliefs. There was a total effect where climate change knowledge predicted attitude to mitigation when efficacy beliefs were not in the model, b = 2.00, SE = 0.57, p < 0.001. There was a direct effect where climate change knowledge predicted attitude to mitigation when efficacy beliefs were in the model, b = 1.59, SE = 0.56, p = 0.005. Most importantly, there was an indirect effect where climate change knowledge predicted attitude to mitigation through efficacy beliefs, b = 0.41, BootSE = 0.19, 95% BootCI [0.09, 0.82]. See Figure 1 for an illustration of the model.
Further analyses to investigate specific effects at the facet level were performed. Twelve additional mediation models were analyzed. Climate change causes knowledge, climate change physical knowledge, and climate change consequences knowledge facets were independent variables. Attitudes to mitigation behaviors and attitude to mitigation policies were dependent variables. Self-efficacy for climate change and response efficacy for climate change were mediators. For the sake of brevity, we only report full details where an indirect effect was found. A summary of mediation coefficients can be found in Table 6.
A mediation analysis was conducted to examine whether climate change causes knowledge predicted attitude to mitigation behaviors through self-efficacy for climate change beliefs. There was a total effect where climate change causes knowledge predicted attitude to mitigation behaviors when self-efficacy for climate change beliefs were not in the model, b = 4.00, SE = 0.65, p < 0.001. There was a direct effect where climate change causes knowledge predicted attitude to mitigation behaviors when self-efficacy for climate change beliefs was in the model, b = 3.42, SE = 0.64, p = < 0.001. Most importantly, there was an indirect effect where climate change causes knowledge predicted attitude to mitigation TA B L E 5 Standardized correlation coefficients and descriptive statistics V a r i a b l e s 1 2 3 4 5 6 7 8 9 1 0  Figure 2 for an illustration of the model. There was no significant indirect effect of climate change physical knowledge on attitude to mitigation behaviors through self-efficacy for climate change.
A mediation analysis was conducted to examine whether climate change consequences knowledge predicted attitude to mitigation behaviors through self-efficacy for climate change beliefs. There was a total effect where climate change consequences knowledge predicted attitude to mitigation behaviors when self-efficacy for climate change beliefs were not in the model, b = 1.47, SE = 0.81, p = 0.07. There was a direct effect where climate change consequences knowledge predicted attitude to mitigation behaviors when self-efficacy for climate change beliefs were in the model, b = 0.71, SE = 0.79, p = 0.37. Most importantly, there was a significant indirect effect where climate change consequences knowledge predicted attitude to mitigation behaviors through self-efficacy for climate change beliefs, b = 0.77, BootSE = 0.32, 95% BootCI [0.2, 1.42]. See Figure 3 for an illustration of the model.
Mediation analyses suggest none of the three climate change knowledge scales had a significant indirect effect on attitude to mitigation behaviors or attitude to mitigation policies through response efficacy. Climate change causes and physical knowledge scales similarly had no significant indirect effect on attitude to mitigation policies through self-efficacy for climate change.
A mediation analysis was conducted to examine whether climate change consequences knowledge predicted attitude to mitigation policies through self-efficacy for climate change beliefs. There was a total effect where climate change consequences knowledge predicted attitude to mitigation policies when self-efficacy for climate change beliefs were not in the model, b = 1.48, SE = 0.65, p = 0.025. There was a direct effect where climate change knowledge predicted attitude to mitigation when self-efficacy for climate change beliefs were in the model, b = 1.15, SE = 0.66, p = 0.08. Most importantly, there was an indirect effect where climate change knowledge predicted attitude to mitigation through selfefficacy for climate change beliefs, b = 0.33, BootSE = 0.19, 95% BootCI [0.3, 0.77]. See Figure 4 for an illustration of the model.

DISCUSSION
The aim of the study was to investigate the influence of efficacy for climate change in the relationship between climate change knowledge and attitudes to mitigation among US adults. The hypothesis that efficacy for climate change beliefs would mediate the relationship between climate change knowledge and attitude to mitigation was supported. The results overall reveal that efficacy for climate change constructs mediates some relationships between specific climate change knowledge and mitigation attitudes. To learn more about the relationships, we examined them at the facet level F I G U R E 4 Moderation model of consequences knowledge as predictor of attitudes to mitigation policies via self-efficacy and found some mediation effects. Self-efficacy for climate change mediated the relationship between climate change causes knowledge and attitudes to mitigation behaviors. Indirect effects of self-efficacy for climate change were also observed between climate change consequences knowledge and attitudes to mitigation behaviors, as well as attitudes to mitigation policies. The previous literature on efficacy for climate change and climate change knowledge offers some potential explanations for the variation across the current results. Previous studies have established that climate change causal knowledge is associated with climate change concern (Bostrom et al., 2019;Shi et al., 2015), and concern predicts climate change attitudes and behavior (Bostrom et al., 2019;Shi et al., 2015). This may explain the moderate relationship between causal knowledge and attitude to mitigation policies and behaviors observed in the current study. Additionally, Bostrom et al. (2019) found concern mediated the relationship between selfefficacy and policy support, but self-efficacy was also directly related to policy support. This suggests that the influence of self-efficacy for climate change has an important role in predicting attitudes. This influence is most apparent in the models that included climate change consequences knowledge as predictors. The current study found no association between consequences knowledge and attitudes to climate change behaviors or policies, but significant indirect effects were found when self-efficacy was added to the model.
The study extends upon the literature supporting both climate change knowledge and self-efficacy for climate change as predictors of pro-environmental attitudes (Bamberg & Möser, 2007;Bostrom et al., 2019;Geiger et al., 2017;Hornsey et al., 2016;Ranney & Clark, 2016). Specifically, the results contribute to the growing body of research on the value of climate change causes and consequences knowledge as a motivator of pro-environmental attitudes and behaviors. The study identified climate change causes knowledge was a reliable predictor of all other knowledge, efficacy, and attitude variables. Climate change consequences knowledge was correlated with self-efficacy for climate change and attitude to mitigation behavior. Similar effects of climate change causes knowledge on attitude to mitigation behaviors were observed by Shi et al. (2015), although direct comparisons cannot be made with their results. While both studies used similar knowledge and attitude measures, Shi and colleagues included additional variables, including demographic, value orientations, and concerns, in their model.
The specific contribution of the current study is the observation of the mediating effects of self-efficacy for climate change. Specifically, self-efficacy for climate change mediated the relationship between climate change causal knowledge and attitudes to mitigation behaviors, as well as climate change consequences knowledge and attitudes to both mitigation polies and behaviors. These findings suggest that self-efficacy for climate change explains some of the observed relationship between climate change knowledge and attitudes to mitigation. This result is comparable to results by Geiger et al. (2017), though, they assessed efficacy and willingness to engage in climate change discussions post cause-and-effect knowledge intervention. The current study measured six self-efficacies for climate change behaviors (including talking about global warming). The results suggest the effect of climate change causes and consequence knowledge may extend beyond confidence in engaging in climate change discussions to a range of mitigation behaviors and policy support. A similar pathway model was observed in a survey by Estrada et al. (2017), with some distinctions. Estrada and colleagues predicted mitigation behaviors, where the current study predicted attitude toward mitigation behaviors. We acknowledge direct comparisons between pro-environmental attitudes and behaviors are difficult, but the correlations between them (Ajzen, 1985) allow for allow some inferences to inform further investigation. When considering these results alongside successful knowledge and self-efficacy interventions in the education and health literature, we can infer a pathway from specific climate change causes and consequence knowledge to pro-environmental attitudes to mitigation via self -efficacy for climate change.

Implications
The results provide further support for the value of specific knowledge about climate change, particularly if the information is be delivered in a manner that develops selfefficacy for climate change mitigation behavior. Messages that develop ones' perception of capability to perform proenvironmental behaviors may increase efficacy for climate change, as can perceptions of engagement being socially normative (Bandura, 1997). The inclusion of a socialpsychological construct to support communication is also consistent with public health research supporting social influences on attitudes (Chess & Johnson, 2007). Messages that deliver explanations of the causes and consequences of climate change with practical information on mitigation may be an effective motivator of pro-environmental attitudes and behavioral change.
The results also suggest brief and targeted climate change messages may be sufficient to influence attitude to mitigation on a large scale. The current study assessed climate change causes and consequences knowledge with eight questions. Past climate change knowledge interventions include a 400-word written explanation (Ranney & Clark, 2016), and a series of information videos totaling 10 min of viewing time (Geiger et al., 2017). We can infer the current samples' sources of climate change knowledge were varied, so there is scope for successful information delivery across different formats. The current study also extends the effects of climate change knowledge and self-efficacy for climate change to a larger population. Past research cited here included samples of university students and park and science-learningcenter guests. The current sample included participants of varying ages, education, and regions. This diversity suggests extensive science literacy is not required to understand climate change and to motivate pro-environmental attitude to mitigation.
The results support a SCT framework for climate change behavior as a useful predictor of attitude to mitigation. Although the outcomes predicted for the current study were attitudes to mitigation, attitude is an antecedent to behavior intentions and behavior (Ajzen, 1991). The mediating effect was proposed by Bandura (1986b) and has support in the health and education literature. The principles of SCT extend beyond self-efficacy and can provide comprehensive guidelines for effective communication to motivate and empower the public toward pro-environmental behavioral change (Bandura, 2006). There are similar constructs explained within SCT that support current social-psychological frameworks of climate change engagement. In other words, constructs like shared beliefs and values have a place in a social cognitive framework for climate change engagement, alongside information and efficacy beliefs. Extension of this framework may support with understanding the unique effects of climate change knowledge and efficacy for climate change among diverse populations and to tailor culturally appropriate messages for engagement.

Study limitations and future directions
We acknowledge the cross-sectional design of this study does not allow assumptions of a causal pathway from knowledge to self-efficacy to attitude. Consistent with SCT and other studies in the field, we have conceptualized self-efficacy as the mediator between knowledge and behavior (Bandura, 2006). However, we recognize an alternative understanding is also possible where, instead of knowledge proceeding efficacy, efficacy precedes knowledge. If people have higher confidence, they may be more likely to explore and seek information. If they have lower efficacy, then they may be less likely to search for new details. As expected, mediation analysis also found an indirect effect when efficacy was the independent variable and climate change knowledge was the mediator. This understanding is also consistent with SCT, which asserts the relationships between the personal influences, behavior, and environment are dynamic and reciprocal (Bandura, 1986b). However, we recommend future studies incorporate a longitudinal element where there is separation in time between the independent variable and the mediator, as this will help clarify the nature of the relationships.
A limitation of the current study may be in the generalizability of results to the US adult population. The sample included a slightly higher proportion of women than observed in the population (US Census Bureau, 2020, April 29). There were also moderate differences in age and relationship status. Because direct comparisons were difficult due to disparities between the data collected and reported census figures, generalization should be made with caution. Cohort effects are particularly relevant in the study of climate change engagement, as attitude can be related to demographic variables, such as age and gender (Pew Research Centre, 2017). The participants were recruited from Amazon's Mechanical Turk (MTurk) online open marketplace. The platform is efficient in that it provides access to a large and diverse pool of study participants to gather data rapidly, but this accessibility must be weighed against the representativeness of the sample. Additionally, while the size of the sample was over the minimum threshold for detection of small effects (Cohen, 1992;Faul et al., 2009), future studies should look to recruit a larger sample to improve generalizability.
Another limitation of the study design is that it cannot account for relationships between attitudes and personal experience of a climate change-related event. Experiencing a severe weather event can influence perception of risk and mitigation behavior and intentions (Bergquist et al., 2019;Siegrist & Gutscher, 2006). It is not known if participants in the sample had such an experience. Longitudinal research can assess emergence of climate change attitudes, as well as their persistence over time. It can also provide opportunities to compare the influence of a major weather event with current risk communication strategies.
The literature overall has also insufficiently explained how knowledge influences self-efficacy. Although self-efficacy theory (Bandura, 1997) identifies four sources of selfefficacy, their relationship to efficacy for climate change are not fully understood. Geiger et al. (2017) suggested their intervention modelled discussions, thereby increasing confidence for discussion through topic mastery and provision of specific language. Their study outcome is too specific to apply to the current findings. Other researchers have avoided suggesting a mechanism of action and often refer to the value of mechanistic knowledge for subject mastery (Ranney & Clark, 2016;Shi et al., 2015;Tobler et al., 2012). Risk communication studies suggest climate change consequences knowledge can induce concern, which in turn can increase self-efficacy, particularly when mitigation options are proposed (Hornsey et al., 2015;Rimal, 2000). While the literature provides support for the relationship, the explanations of the mechanisms of influence are sparse and inconsistent. Testing of SCT-informed knowledge interventions may support understanding in this regard.

CONCLUSIONS
To the best of our knowledge, this is the first study to examine the mediating effect of efficacy for climate change beliefs on specific climate change knowledge and attitudes to mitigation among a community sample of US adult participants. The results support both information-deficit and socialpsychological models of climate change engagement. There are theoretical implications for development of a comprehensive framework for engagement, and practical implications for climate change communication. Specific climate change causes and consequence information may motivate mitigation behavior, particularly with the inclusion of practical guidance. Longitudinal studies can enable further inferences about mechanisms of action between climate change knowledge and efficacy for climate change, and if their influence persists over time. Successful climate change mitigation will require collective and sustained action; identifying RELI-ABLE predictors of engagement is critical to facilitating the essential behavioral change.