Environmental knowledge is inversely associated with climate change anxiety

This study tests the hypotheses that overall environmental knowledge and climate-specific knowledge are inversely related to climate change anxiety, such that people who know more (less) about the environment in general, and about climate in particular, are less (more) anxious about climate change. Time lagged data were collected from N = 2,066 individuals in Germany. Results showed that, even after controlling for demographic characteristics, personality characteristics, and environmental attitudes, overall environmental knowledge and climate-specific knowledge were negatively related to climate change anxiety (both B = -.09, p < .001).


Introduction
The British mathematician and philosopher Bertrand Russell (1872Russell ( -1970 is quoted as saying, "The degree of one's emotions varies inversely with one's knowledge of the facts, the less you know the hotter you get." Consistent with this idea, we report the results of a study that tested the hypotheses that overall environmental knowledge and climate-specific knowledge are negatively related to climate change anxiety. The rationale for investigating these relationships is that climate change anxiety may be reduced through interventions that enhance environmental knowledge. Environmental knowledge refers to the body of acquired facts and learned expertise in the environmental domain (including climate; Geiger et al. 2019). Climate change anxiety involves people's self-reported negative emotional responses associated with their awareness of climate change (Clayton 2020). Accordingly, the measure used in this study assesses responses such as worrying and distress, reduced well-being, as well as concentration and sleep difficulties due to people's thoughts about climate change (Clayton and Karazsia 2020). Compared to more constructive forms of worrying about global warming that result in adaptive responses (e.g., Verplanken et al. 2020), it seems desirable to reduce climate change anxiety because it has been shown to have detrimental consequences for people's general health and well-being, including higher levels of depression, anxiety, and distress (Schwartz et al. 2022;Searle and Gow 2010).
Psychological theorizing suggests that a lack of knowledge about a topic leads to greater feelings of uncertainty which, in turn, may result in anxiety associated with the topic, whereas greater available knowledge about a topic is likely to reduce experiences of topic-associated uncertainty and anxiety (Hirsh et al. 2012;Jungermann and Thüring 1993). Indeed, empirical studies conducted on various topics (e.g., aging, interpersonal and intergroup relations, finances, entrepreneurship) suggest that knowledge about a topic is negatively related to uncertainty and anxiety (Nuevo et al. 2009;Stephan et al. 1999;Townsend et al. 2018;Wang et al. 2011). For example, a meta-analysis showed that knowledge about one's own and others' emotions is inversely related to social anxiety (O'Toole et al. 2013).
Several motivational predictors of pro-environmental behavior, such as environmental awareness, attitudes, values, and identity, are positively related to climate change anxiety (e.g., Verplanken et al. 2020). So far, however, no research on the association between environmental knowledge and climate change anxiety exists (Coffey et al. 2021;Whitmarsh et al. 2022). An exception is a qualitative case study on environmental knowledge and various emotions experienced by activists (e.g., fear, hope; Martiskainen et al. 2020). The descriptive evidence presented in this study suggests that both a positive and a negative relationships may be possible: "Frontline protesters were really concerned about climate change… They were often the most knowledgeable about climate change …" (p. 14). In contrast, the researchers also observed that "In terms of cognitive knowledge, a surprising number of our strikers were not very aware about climate change; with some even stated that they had no knowledge and were merely striking to impress others (such as a romantic partner) or to give a friend company" (p. 16). Another study showed that knowledge about the causes of climate change and its potential health consequences positively predicted cognitive and affective risk judgments (Sundblad et al. 2007). However, the risk judgment measures used in this study differ from the more personal and emotional construct of climate change anxiety, in that participants were asked to estimate the probability of, and their worries about, serious negative consequences of climate change in three specific countries (i.e., Sweden, the Netherlands, Bangladesh) for three different time horizons (i.e., 5, 50, 100 years).
In summary, drawing on psychological theorizing and research on associations among knowledge, uncertainty, and anxiety, we hypothesize that people with greater overall environmental knowledge and climate-specific knowledge experience generally less climate change anxiety. In contrast, people with less overall environmental knowledge and climate-specific knowledge should be more likely to experience higher climate change anxiety.
Hypothesis 1: Overall environmental knowledge is inversely associated with climate change anxiety.
Hypothesis 2: Climate-specific knowledge is inversely related to climate change anxiety.

Participants and procedure
Time lagged data were collected from a large sample in Germany at four time points across three months (i.e., Time [T] 1 = August, T2 = September, T3 = October, and T4 = November of 2022). The time lags help reduce the potential problem of common method bias and allow drawing stronger inferences regarding the assumed temporal ordering of variables. Collecting constructs at different time points was also necessary to reduce the length of each survey. Participants were recruited by an ISO 26362 certified professional online panel company. At T1, participants provided information on demographic characteristics and environmental attitudes as control variables. At T3, participants took an environmental knowledge test (n.b., no data collected at Time 2 is used in the current study). Finally, at T4, participants reported their climate change anxiety, as well as the Big Five personality characteristics as control variables.
A sample of 2,066 employed adults in Germany participated in this study. The sample was 50.39% female and on average 47.07 years of age (SD = 12.18, range 18-85 years). Most participants (43.47%) held a college/university or technical college degree. The sample was not representative of the general population, as it did not include children, retirees, and unemployed people. The sample was also older and more highly educated than the working population in Germany (i.e., average age of 44 years, 24% college/university degree). Further information on the sample (e.g., income, industry) is available in Table S1 in the online supplemental materials (OSM; https:// osf. io/ 5mqjh).
The first survey was initiated by 3,566 people. Sample sizes for observed variables varied by time point (T1: 3,256-3,562; T3: 2,350-2,353; T4: 2,336-2,337), suggesting some degree of attrition over time and some degree of missingness within each time point. Our analyses are based on a sample of 2,066 individuals who provided complete data on our substantive and control variables across all time points. Table S1 compares these complete responders to incomplete responders (n = 1,500). We also ran a logistic regression model showing that T1 demographics predicted less than 2% of the variance in attrition, suggesting that bias due to attrition was not a significant concern. Complete data and R code to reproduce the analyses reported in this paper can be found in the OSM.

Measures
Reliability information for all measures is reported in Table S2 in the OSM.

Environmental knowledge A 35-item multiple-choice environmental knowledge test, in
which answers to the questions can be scored as "correct" or "incorrect," was collected from participants at T3 (Geiger et al. 2019). An example item is: "Which of the following phenomena has been the main cause of global warming over the last 20 years? (a) Reduction of the ozone layer (the so-called ozone hole), (b) Increased emissions of greenhouse gases (the so-called greenhouse effect), (c) Changes in ocean currents, (e.g., "el Niño"), (d) Changes in the skewness of earth's axis." Each participant's overall environmental knowledge score was represented as the number of correct answers out of 35. Consistent with Geiger et al. (2019), we further computed seven dimension scores based on subsets of the items, including basic ecology (5 items), economy (3 items), consumption behavior (9 items), environmental contamination (5 items), climate (5 items), resources (4 items), and society/politics (4 items).
Climate change anxiety A 13-item measure of climate change anxiety was collected at T4 (Clayton & Karazsia 2020;Wullenkord et al. 2021). Example items are "Thinking about climate change makes it difficult for me to concentrate" and "I find myself crying because of climate change." Responses were provided on a 7-point scale ranging from 1 = strongly disagree to 7 = strongly agree. Clayton and Karazsia (2020) suggested that their scale consists of two dimensions, cognitive-emotional and functional impairment, which were strongly correlated in their studies (i.e., r = 0.78, r = 0.84). Consistent with Wullenkord et al. (2021), who were unable to replicate this two-dimensional structure using the German version of the scale, we used a single score of climate change anxiety in our focal analyses. As shown in our correlation table (Table S3 in the OSM), the two dimensions were also strongly correlated in our study (r = 0.88). However, we report results of supplemental analyses for each climate change anxiety dimension. As shown in Table S4 in the OSM, these results were very similar to our focal results. Personality characteristics We additionally controlled for core personality characteristics classified by the "Big Five" model (i.e., extraversion, agreeableness, conscientiousness, emotional stability, openness). Emotional stability (the opposite of neuroticism) is associated with clinically-relevant indices of anxiety (Kotov et al. 2010) and climate change anxiety (Clayton 2020). Moreover, people with higher conscientiousness and openness may be more motivated to obtain environmental knowledge. We collected the 21-item Big Five Inventory at T4 (Rammstedt and John 2005). Participants provided their responses on a 7-point rating scale ranging from 1 = strongly disagree to 7 = strongly agree.

Demographic characteristics
Environmental attitudes Finally, we controlled for environmental attitudes using a wellestablished eight-item scale collected at T1 (Bamberg 2003). Environmental attitudes (and closely related motivational constructs, such as environmental awareness) are positively related to climate change anxiety (e.g., Verplanken et al. 2020). An example item is "For the benefit of the environment we should be ready to restrict our momentary style of living." Responses were provided on a 5-point scale ranging from 1 = strongly disagree to 7 = strongly agree.

Statistical analyses
We used ordinary least squares (OLS) regression for our focal analyses. In a first model, T4 climate change anxiety was regressed onto T1 demographics and environmental attitudes, T4 personality characteristics, and T3 overall environmental knowledge. In a second model, we regressed climate change anxiety on the control variables and the seven environmental knowledge dimensions (instead of overall environmental knowledge).

Results
Descriptive statistics and correlations are provided in Tables S1 and S3 in the OSM. The mean of climate change anxiety was 2.10 (on a scale ranging from 1 to 7), and on average there were 21.50 correct responses to the 35 environmental knowledge questions (range: 5-35). The bivariate correlations of overall environmental knowledge and climate-specific knowledge with climate change anxiety were r = -0.34 and r = -0.24, respectively.
Results of regression analyses (see Table 1) showed that overall environmental knowledge negatively predicted climate change anxiety (B = -0.09, p < 0.001), above and beyond the effects of demographics, environmental attitudes, and personality characteristics (see Fig. 1). Thus, Hypothesis 1 was supported: people who possess more (less) overall environmental knowledge experience less (more) climate change anxiety. Moreover, scores of six out of the seven content domains were negatively associated with climate change anxiety in a separate analysis, including knowledge about consumption behavior (B = -0.10, p < 0.001), basic ecology (B = -0.06, p = 0.022), economy (B = -0.12, p < 0.001), society/politics (B = -0.06, p = 0.015), contamination (B = -0.09, p < 0.001), and climate (B = -0.09, p < 0.001, see Table 1). The latter finding supports Hypothesis 2: people who possess more (less) climate-specific knowledge experience less (more) climate change anxiety.
The effects of the control variables were largely consistent with expectations and previous research (e.g., Whitmarsh et al. 2022). Specifically, in both models, education, environmental attitudes, and openness positively predicted climate change anxiety, whereas age, conscientiousness, and emotional stability negatively predicted climate change anxiety. In contrast to previous research, sex was not significantly correlated with climate change anxiety at the bivariate level (Table S3), but negatively predicted it in the regression analyses, suggesting that women experience less climate change anxiety when men. A supplemental specification curve analysis suggested that environmental attitudes, emotional stability, and environmental knowledge may be masking this effect in the correlations (see Figure S1 in the OSM).

Supplemental analyses
Due to a skewed distribution of climate change anxiety, we additionally considered generalized linear models (inverse Gaussian with an identity link) as a robustness check. The results of these models were equivalent to the OLS models in terms of the statistical conclusions drawn from our primary analysis (see Table S5 in the OSM).
In addition, although previous research suggests a linear relationship between knowledge and anxiety, we explored nonlinear associations between environmental knowledge and climate change anxiety. As shown in Table S6 in the OSM, there was evidence for nonlinear effects of overall environmental knowledge and its consumption and climate dimensions on climate change anxiety. However, as shown in Figure S2, these relations had a very similar form as the linear relations (i.e., "L-shaped," suggesting "diminishing returns" rather than "U-shaped" effects).

Discussion
The results of this study add to the growing body of evidence that higher domain-specific knowledge is associated with lower domain-associated anxiety. They also contribute to the emerging literature on the construct of climate change anxiety, which so far has focused on demographics, attitudes, and personality characteristics as predictors and neglected the role of environmental knowledge (Clayton 2020). In terms of practical implications, the main finding that environmental knowledge is negatively related to climate change anxiety suggests that efforts to improve environmental knowledge, for instance through educational and training interventions, may help reduce such anxiety. This seems important given demonstrated links between climate change anxiety and more general forms of mental illhealth, including generalized anxiety, depression, and distress (Schwartz et al. 2022 and Gow 2010). These interventions could be targeted especially at younger people and people with higher environmental awareness and attitudes (e.g., climate scientists), who are at risk of experiencing higher climate change anxiety (Clayton 2020;Crandon et al. 2022;Verplanken et al. 2020). However, consistent with previous research (Wullenkord et al. 2021), the average level of climate change anxiety in our sample was already quite low and, thus, it may not be possible to decrease it much further through interventions. It could also be questioned whether climate change anxiety has only (and always) negative consequences or, under certain circumstances, may even lead to positive outcomes. For example, recent research has suggested that engaging in collective action to address climate change can buffer the effect of climate change anxiety on depressive symptoms (Schwartz et al. 2022). Moreover, the measure used in our study is based on a definition of climate change anxiety as "a more clinically significant 'anxious' response" (Clayton and Karazsia 2020, p. 9). Future research should also focus on the potentially beneficial outcomes of more adaptive forms of climate change anxiety, such as practical anxiety (Pihkala 2020) or rational worrying (Verplanken and Roy 2013). Compared to clinical forms of climate change anxiety, these more adaptive forms may be constructively channeled into productive responses, such as increased information seeking and pro-environmental behavior (Verplanken et al. 2020).
Author contributions Both authors contributed to the study conception and design. Material preparation and data collection were performed by Hannes Zacher, and analyses were performed by Cort W. Rudolph. The first draft of the manuscript was written by Hannes Zacher and Cort W. Rudolph and both authors commented on previous versions of the manuscript. Both authors read and approved the final manuscript.
Funding Open Access funding enabled and organized by Projekt DEAL. The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability
The dataset generated and analyzed during the current study are available in the Open Science Framework, https:// osf. io/ 5mqjh

Competing interests
The authors have no relevant financial or non-financial interests to disclose.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.