Effects of Local vs. Global Social Media News-Posts on Pro-Environmental Behavioural Intentions, Environmental Self-Efficacy and Eco-Anxiety.

(1)

Effects of Local vs. Global Social Media News-Posts on Pro-Environmental Behavioural Intentions, Environmental Self-Efficacy and Eco-Anxiety.

Master Thesis

Graduate School of Communication

Master’s Programme of Communication Science - Entertainment Communication Track

February 4^th, 2022 Student: Chiara Genta Student Number: 12657522

Supervisor: Dr. Ewa Międzobrodzka Wordcount: 7462

(2)

Abstract

Climate change issues are increasingly discussed and shared on social media, where they are most often framed as global issues. The following study investigated whether different framings of climate change-related social media news, portrayed either as a local or as a global issue, may impact pro-environmental behavioural intentions, environmental self-efficacy, and eco-anxiety.

An exploratory analysis also investigated eco-anxiety as a possible mediator. The online experiment randomly assigned 270 participants to one of the two conditions (local or global), and then measured the three outcome variables. The results suggested no effects of the manipulation, neither on pro-environmental behavioural intentions, nor on environmental self-efficacy. Notably, although eco-anxiety was not found to mediate the relationship with environmental self-efficacy, a significant positive effect of eco-anxiety on environmental self-efficacy was found, suggesting that increases in eco-anxiety may go hand in hand with increases in environmental self-efficacy.

Given the novelty of eco-anxiety as a research area, this result paves the way for future research, suggesting that investigating the underlying mechanism between the two variables could prove illuminating.

(5)

1

Introduction

When thinking of current public affairs, climate change is considered one of the most salient, urgent topics: thus, its prevalence has taken the spotlight in various areas of public discussion and mass information (Kellstedt et al., 2008), which, nowadays, expectedly include social media. In fact, the climate change issue has gained traction and has increasingly been shared and discussed online, often trending on social media (Berglez & Al-Saqaf, 2021; Tuitjer & Dirksmeier, 2021).

Further support comes from a recent study by Aquadro Maran and Begotti (2021), which found that for the tested Italian sample, Facebook appeared as the top source for environmental news – followed by TV and newspapers.

However, social media are not only used for information seeking: while originally hubs for facilitating communications, in the past 15 years they have also become a major source of entertainment among people (Cunningham & Craig, 2019; Reinecke et al, 2014). Nevertheless, notwithstanding the prominence of entertainment as a driver for Facebook use, studies increasingly find that it is considered a primary source of news by a majority of its users, particularly for political or public affairs (Gil de Zuniga et al, 2012; Lu & Lee, 2021; Skoric et al, 2016). Worth noting in the context of this study, research has conceptualised as “news” even simple posts briefly relaying news-stories or concerning current affairs – in and by itself, without necessarily linking to external articles (Müller et al., 2016).

In general, incidental exposure to news on social media has proven to be an unexpected source of learning, knowledge, and even online political participation (Anderson et al., 2021; Lee

& Kim, 2017; Valeriani & Vaccari, 2016). Thus, such trend of learning through entertainment media is far from new, and may well be accosted to the concept of ‘edutainment’: in this context,

(6)

2

information is imparted in a less didactic, more engaging and informal way – and in fact, the affordances of Facebook as an edutainment medium have been proven successful in creating a solid knowledge base in an international sample of students (Isacsson & Gretzel, 2011). Hence, the potential of such widespread entertainment platforms as possible sources of news and learning for audiences should not be overlooked.

Building on this knowledge, three aspects are worth considering in the scope of research:

firstly, the way in which climate change-related news, as found on social media, may impact engagement with the issue. Secondly, whether different framings of climate change-related news (i.e., as a local or as a global issue) may impact users’ differently. Finally, equally important to consider, are the negative effects climate change-related news engender in the public (recently defined as eco-anxiety), and the possible ways in which such emotional state may impact action and engagement with the topic of climate change.

In fact, in this context, it is important to note how much of the coverage on climate change focuses on global aspects and portrays large-scale problems and solutions – which divert public attention from the possible local implications (Liu et al., 2008). As a consequence, the climate crisis is often perceived as an abstract and distant threat. An essential implication of this is the psychological distance which is consequently perceived between the people and the issue itself.

Furthermore, extending from Roe (Horta et al., 2017), a “global hegemony of news” risks excusing inaction at national and local levels, and risks increasing the distance between citizens and political processes – in this particular case, rather than political processes, it could decrease engagement with climate change issues. Likewise, it is argued that a shift towards the global poses the risk of a retreat from social engagement (Chandler, 2009), which again in the case of climate-change, may pose a significant issue in directing people to appropriate actions. This is relevant because, in the

(7)

3

context of the climate change, engagement (particularly, engagement with pro-environmental behaviours) translates in actions which can reduce negative impacts and thus, reduce the severity of climate change. Then, the question arises as to whether similar effects may also be observed on social media. Would exposure to globally-focused climate change related news in social media posts have different impacts on people than exposure to locally-focused ones?

On the other hand, exposure to environmental news on entertainment media also has the potential of creating exceedingly negative affective states. Information on, and awareness of climate change and related disasters are increasingly found to create a distressing emotional state, recently defined as “eco-anxiety” (Cunsolo et al., 2020). Although still lacking an official diagnosis, “eco-anxiety” is a specific form of anxiety created by the stress deriving from environmental changes and, importantly, from our knowledge of them (Usher et al., 2019).

Furthermore, another aspect worth investigating is how eco-anxiety may also be intertwined with self-efficacy states: in fact, in recent research characterizing eco-anxiety, Pikhala (2020) denotes a dimension of powerlessness and helplessness in dealing with the climate change, which suggests a perceived inability to tackle the issue. Thus, based on this information, it could be expected that for individuals suffering from eco-anxiety, some aspect of state self-efficacy may also play a part – in that anxiety may partly stem from not feeling capable to be effective in limiting climate change. It is in fact still unclear in what ways eco-anxiety may be connected to the state self-efficacy, and whether exposure to closer (locally-focused) and more tangible social media news about the climate change could affect the level of state eco-anxiety in individuals. Finally, in what ways do these mechanisms play out in the context of social media?

Building on the previous knowledge, this research aims to explore the way in which locally- or globally-focused social media posts about climate change may impact pro-

(8)

4

environmental behavioural intentions and state self-efficacy. Further, it explores the possibly mediating role of eco-anxiety in the relationship between social media posts exposure and self- efficacy. Thus, the following research questions were formulated:

RQ: To what extent do locally-focused versus globally-focused climate change related posts on social media impact pro-environmental behavioural intentions and environmental state self-efficacy? Does eco-anxiety mediate the effect of exposure to locally- vs globally-focused climate change-related social media posts on environmental state self-efficacy?

Academically speaking, although some studies exist which employ the Construal Level Theory to investigate perceptions of Climate Change as a distant or proximal threat, this has not yet been done with the specific, briefer and ostensibly more easily digestible format of social media posts. The high ecological validity of the current approach may help predict, based on posts from one of the most frequently used social media platforms – Facebook – how people may react to the climate change issue portrayed in different ways (either locally- or globally-focused) in their daily social media use. Moreover, this study adds practical suggestions to the information on climate change, suggesting how to counteract the issues relayed – which adds to the research field, as the previous studies on perceptions of the climate change portrayed causes and consequences of the climate change only. Thus, this study hopes to unveil possible mechanisms underlying effects of social media on the climate change perceptions. Furthermore, academic research on eco-anxiety is still at its outset: thus, this study aims to also contribute to the research gap on eco-anxiety, particularly in the context of social media.

In terms of societal relevance, the study aims to contribute to a more proactive society, more confident in its ability to counteract the climate change with individual and collective actions.

Furthermore, considering the issues that eco-anxiety poses to the quality of life of those

(9)

5

experiencing it, this study was compelling in offering the possibility of contributing to the understanding of perceptions of audiences on such a pressing matter, and hopefully offering direction for future research.

Furthermore, many Non-Governmental Organisations and Governmental Organizations like NASA are widely followed on social media, and often share relevant and novel information on climate change for learning and entertainment alike (Greenpeace, 2017; Greenpeace International, 2019; NASA, 2021). If any new knowledge will arise from this study on how to better and more effectively inform the public of their possibilities in counteracting environmental degradation, this could prove useful in creating more effective strategies to reach the public without creating excessively negative emotions, while maximising positive outcomes such as more sustainable, pro-environmental behaviours.

Theoretical Framework

Locally- vs globally-focused climate-change related news-posts (IV)

Climate change related news-stories do not necessarily need to explicitly refer to the climate change to be climate change-related: in fact, stories about droughts, floods or causing factors of climate change are all the same related to it, even when lacking explicit mention (Hart

& Feldman, 2014). By extension, it can be argued that any news-story or media product related to these practices and events can be considered as climate-change related. In fact, as conceptualised in a recent study by Olteanu et al. (2015), any information dealing with causes such as CO2 emissions, consequences, current practices, and actions to limit environmental degradation is considered climate-change related. As mentioned earlier, in the specific context of social media,

(10)

6

even simple posts briefly relaying news-stories (including climate change-related ones) are considered as news (Müller et al., 2016).

In general, climate change-related news relay events of disproportionate magnitude and distance – like the melting of ice caps or migrations of populations impacted by rising temperatures (Hart & Feldman, 2014). It is important to note the focus that most of these climate change-related news employ: they could be considered as globally-focused, as they portray large-scale events, often displaced from local levels, which emphasize a far-away nature of the problem (Liu et al., 2008; O’ Neill & Nicholson-Cole, 2009). An example of globally-focused news could be news related to general rising of CO2 levels in the atmosphere.

On the other hand, locally-focused news refer to any news related to climate change causing factors or impacts taking place in specific countries and regions. The coverage of locally- focused climate change-related news may be much less frequent than globally-focused coverage:

in the case of Portugal, for instance, it was found that climate-change related news were disproportionally covered at the global level, while a local focus on the issue was nearly absent – with three quarters of the coverage focusing on the international/global scale of the issue. In this case, as with any coverage neglecting the local focus to employ a global one, attention is diverted from local and national realities, hindering the addressing of the issue (Horta et al., 2017).

Thus, when referring to locally focused climate-change related news, these relate to any news concerning the national (or even regional) levels. In recent studies as well, local is defined as the national context (Liu, 2018): the same approach will be used in this study, where local refers to the country on which one of the manipulations focused – the Netherlands.

(11)

7

A necessary mention in the scope of locally- vs globally-focused climate news is the consequent psychological distance perceived between individuals and the events portrayed:

composed of four subdimensions (spatial, temporal, social and hypothetical), psychological distance plays a part in how any information is received, and the abstractness or concreteness of its mental representation (Trope & Liberman, 2010). When distance with an object decreases in either of these intercorrelated dimensions, the object is perceived as psychologically closer and more concrete, albeit arguably to different extents (Trope & Liberman, 2010). Construal level theory helps understand this by hypothesizing more abstract representations of objects when they are perceived as psychologically distant, whereas for spatially close objects it hypothesizes a mental representation as more tangible and relatable, and composed of more practical steps (Brügger et al., 2016). In the context of news, this is relevant because a psychologically closer (local) event is expected to lead to more concrete, practical mental representation for the audience;

conversely, a psychologically distant (global) event will be perceived as more abstract. The important implication of psychological distance and the way it shapes construal of objects is that it can influence the way people act, their voting intentions, self-control and even the pro-social behaviours they carry out (Soderberg et al., 2015). In fact, the mental representation (construal) of any object plays a central part in individual’s evaluations, predictions, and behaviours (Trope &

Liberman, 2010). Thus, news perceived as (psychologically) distant are likely to induce abstract and vague conceptualizations of the issues relayed: a possible consequence of this in the context of climate change, for instance, is that it might hinder the ideation of strategies to counteract it (Spence et al., 2012).

(12)

8

Thus, in the context of climate change-related news as well, different levels of perceived psychological distance may yield different outcomes and perceptions, or encourage different types of behaviour.

Pro-environmental Behavioural Intentions (DV1)

Pro-environmental behavioural intentions indicate the extent to which individuals are motivated to take personal actions to reduce negative impacts towards the environment, or produce positive ones – such as recycling to reduce one’s own carbon footprint (Steg & Vlek, 2019; Yuriev et al., 2020). Research suggests that, particularly in the milieu of environment, information may be a successful strategy in encouraging pro-environmental behaviours, as it is assumed that new knowledge results in attitudinal changes, which consequently affect behaviour (Steg & Vlek, 2019). For instance, informational prompts specifically tailored to environmental sustainability and energy savings have been shown to successfully improve pro-environmental behaviour in households – in one instance, reducing energy expenses by 5.1% (Abrahamse et al., 2007).

Whereas it has been argued that information alone may not always be enough to change behaviours (Geller, 1994), several interventions which used informational strategies on water and energy conservation were found to be effective, even immediately after the intervention itself (Lehman & Geller, 2004). Moreover, there is evidence suggesting that information tailored to fit specific situations may be more effective than general information; furthermore, useful in the context of this study, it was found that prompts to change behaviour (thus, tailored information) worked even better when the target behaviour was easy to perform (Lehman & Geller, 2004).

In terms of locality and proximity, most studies found that framing climate change as proximal yields higher intentions of mitigation practices and pro-environmental behaviours (Maiella et al., 2020). A study by Passafaro et al., (2019) leveraging on the Theory of Planned

(13)

9

Behaviour had also found that social norms were more likely to be internalized, and consequent prosocial behaviours carried out, when a component of spatial proximity was added. In the study it was argued, in fact, that simply sharing the same social-physical space with others could be a potential source of normative influence (Passafaro et al., 2019). Some suggestions of the importance of spatial proximity on individual behaviour has even been found in the field of health (Carroll et al., 2018). Therefore, it will be interesting to investigate whether similar findings can be obtained regarding the climate change. Building on that, the following hypothesis is proposed:

H1) Locally focused climate change-related news in social media posts will lead to higher pro-environmental behavioural intentions as compared to exposure to the globally- focused social media posts.

Environmental Self-efficacy (DV2)

Bandura (1986) defined self-efficacy as the belief in one’s own capabilities to act upon issues and reach a designated goal. Rather than a trait, in this case, self-efficacy is seen as a context- dependent state, which can thus be increased or decreased by external factors (Nissen, 2019).

Research has consistently proven that people who feel more self-efficacious are more likely to engage in issues, and more likely to try to solve them: in political science literature, for instance, it has been found that feeling internally efficacious makes individuals more likely to try and influence political processes by taking personal action and actively participating (Hart & Feldman, 2014).

In the context of environment, recent research has specifically characterised self-efficacy as “environmental-self efficacy” when it refers to individuals’ beliefs in their perceived ability to take actions that have a tangible, positive impact in mitigating climate change. In this context, the study found that higher environmental self-efficacy had a positive direct effect on pro-

(14)

10

environmental behaviours (Huang, 2016). Nonetheless, mixed results about state self-efficacy have also emerged, particularly in the context of environment and climate change: for instance, while overwhelming coverage and imagery of climate change heightens concern but drastically reduces state self-efficacy (Hart & Feldman, 2014), other results suggest that localised, context- specific information about how individuals can personally mitigate climate change might actually increase perceived personal agency – which is partly determined by state self-efficacy (Bandura, 2000; Sheppard et al, 2011). Further support is given by Brugger et al. (2016), whose results suggested that proximal messages may have a positive, albeit weak effect on increasing environmental state self-efficacy. Moreover Chu & Yang (2020), who also used CLT as founding theory for their study, found in their experiment that when participants in the closer spatial distance condition were exposed to information on feasible solutions to climate change, their self-efficacy beliefs increased. Based on this literature, it seems reasonable to expect similar effects in the context of social media posts, leading to the second hypothesis:

H2: Locally focused climate-change related news in social media posts will lead to higher environmental state self-efficacy as compared with globally focused climate change-related social media posts.

Eco-anxiety

Eco-anxiety is a recently coined term indicating a common emotional state of audiences towards the state of the planet and the approach of climate-change (Hickman, 2020). Beside the distress and negative emotions caused by witnessing first-hand natural disasters attributed to climate-change, the awareness itself of the state of the world is causing depression, anxiety and even suicidal thoughts among people all around the world (Cunsolo et al, 2020). In fact, although not yet diagnosable by medical practitioners, the largest survey yet on the topic showed evidence

(15)

11

(particularly in young adults) of people feeling doomed, hopeless, and helpless, and pointing to the damages this is causing to their mental health (Harrabin, 2021).

Worth considering in the context of emotions and psychological distance of information, a study by van Lent et al. (2017) found that the Ebola outbreak only elicited fear precisely when it became a closer issue both spatially and psychologically. Tang et al. (2019), on the other hand, found that the more distant an event portrayed online was, (thus, the higher the spatial and psychological distance) the more negative the emotions it elicited. Brugger et al. (2016) on the other hand, included fear in their study but did not elaborate on the extent to which it differed across conditions. Results are thus mixed. However, as mentioned earlier, much of the coverage on climate change concerns large-scale events: this study explores the possibility that, if faced with more relatable and spatially close events, individuals’ eco-anxiety might be lower because, by relating to more concrete and relatable events, it might feel easier to act upon them and the object of anxiety.

As mentioned earlier, an essential component of eco-anxiety is the feeling that an individual is helpless in combatting this phenomenon (Harrabin, 2021), and although research shows mixed results so far, the individual aspect of state environmental self-efficacy might play a part in this feeling, as suggested by two recent studies (Aquadro Maran & Begotti, 2021; Pikhala, 2020). In fact, while Pikhala’s review (2020) simply denoted a feeling of powerlessness in the context of eco-anxiety, Aquadro Maran & Begotti’s results (2021) precisely confirmed that climate-anxiety was positively related with self-efficacy. Moreover, another possible relationship between eco-anxiety and state (environmental) self-efficacy has also been suggested by the results of a recent study: Bostrom et al. (2019) found that the level of concern respondents expressed towards climate change was a significant mediator of the effect of self-efficacy on support of

(16)

12

sustainable policies. Moreover, Verplanken et al. (2020) also found that different levels of anxiety yielded different types of behaviour towards climate change: eco-anxiety was unconstructive only for those scoring high on pathological worry, while constructive and possibly conducive of positive action for all other respondents. As not many other studies have yet delved in this possible relationship (particularly in terms of climate change information on social media), one of the purposes of this study is exploring the mediating role of eco-anxiety in the relationship between exposure to social media post and state (environmental) self-efficacy:

RQ1: To what extent does eco-anxiety mediate the effect of exposure to locally vs globally-focused climate change-related social media posts on environmental self-efficacy?

Figure 1. Conceptual Model

The amount of time spent on social media, particularly in the milieu of climate change, has mostly been found to increase individuals’ awareness of the issue, efficacy, and likelihood to carry out pro-environmental actions (Huang, 2016; Tuitjer & Dirksmeier, 2021). For this reason, time spent on social media will be used as a control variable to unveil possible effects on either

(17)

13

environmental state self-efficacy, state eco-anxiety or pro-environmental behavioural intentions.

Likewise, since gender has often been found to predict different levels of beliefs in climate change (Milfont et al., 2015) and engagement with it (Scannel & Gifford, 2013) – with females scoring higher than males in both – this variable will also be controlled for in the analyses. These hypotheses, as well as the method and analyses plan were pre-registered at AsPredicted as a part of a bigger project with international collaborators: https://aspredicted.org/QJX_DQK

Methods

Participants

Participants were recruited in December 2021 employing a mix of convenience and snowball sampling, both through social media networks (~41%) and the UvA participants’ pool (~59%). A total of N = 315 participants (76.6% female; 21.5% male; rest – non-binary/not answered) were recruited with the Qualtrics link. Participants’ ages ranged from 18 up to 57 years old (M = 22.34; SD = 4.79). All respondents were currently living in the Netherlands (60.3%

temporarily; 36.5% permanently; 3.2% not answered). Most of the participants achieved High School education level (60.1%); 32.3% completed a Bachelor’s degree; 6% achieved a Master’s degree; two participants (0.6%) achieved a PhD, and three (0.9%) people in the sample had achieved an education level below that of a high school.

Participants could participate in an online lottery for one out of six Bol.com vouchers of 25 EUR each (total prize pool: 150 EUR, in line with ethical guidelines); six winners were selected using a random number generator. Moreover, UvA student participants recruited via the participants’ pool were rewarded with 0.5 research credits. The study was accepted by the Ethical Committee of the UvA on the 9^th of December 2021 (2021-YME-14256).

(18)

14 Procedure

Upon entering the Qualtrics online experiment, participants were presented an information page including a consent form. After agreeing to participate, participants were asked about their demographics and general social media use (frequency and purpose of social media use). Next, participants were randomly and evenly assigned to one of the two conditions: two posts for global social media posts condition (n = 130), or two posts for local (Netherlands) social media posts condition (n = 140). After exposure to the social media posts, participants were asked a question pertaining the content of the posts as an attention check. Finally, after being exposed to two posts in each condition, all participants were asked about their current perceived level of eco-anxiety, their perceived level of state self-efficacy, and their current pro-environmental behavioural intentions. After completing the survey, participants were thanked, debriefed with an informational statement about the precise scope of the research, and given a chance to enter their email address for participating the e-vouchers lottery and collecting credits for research if they were UvA students. On average, the online experiment took around 7 minutes to complete.

Global vs local posts manipulation

Before the manipulation, participants were informed to carefully read two social media posts since they would be asked questions related to their content. Participants were randomly assigned to one of the two conditions: global social media posts condition, or local (in the Netherland) social media posts condition. In both conditions, participants viewed two tailored Facebook posts, featuring widely known facts about climate change causes and consequences.

Each post was displayed for a minimum of 10 seconds to ensure that participants fully read it, correctly experiencing each stimulus. The stimuli differed between the two conditions only in the fictional setting of the posts: global, in the world vs local, in the Netherlands (see Figure 2 for

(19)

15

stimulus example). Two posts were used in each condition. The first post concerned microplastics and their contribution to pollution and global warming; the second concerned CO2 emission, rising temperatures and possible mitigation practices. The manipulation (reading two posts per condition) took around 2 minutes. To ensure that participants paid attention to the stimuli content, an attentional check was applied in both conditions, asking about a topic of each post. For the scope of the analyses, the independent Variable of Local/Global climate-change related news-post was coded as a dichotomous variable, where 1 indicated the Local climate-change related news-post condition, and 0 indicated Global climate-change related news-post condition.

Stimuli example:

Condition 1 (Global): Globally, sea and clearwater are increasingly being polluted by microplastics – microscopic fragments of incorrectly discarded plastic products. These microplastics, beside damaging the fauna’s health, hinder CO2 sequestration on part of the sea, which contributes to increase of temperatures. Cigarette butts and plastic bags are one of the most common culprits to microplastics in water. Correctly disposing of plastics and cigarette butts in recycling bins, and using plastic-free products, effectively contributes to reducing these effects.

Condition 2 (Local): In the Netherlands, the North Sea, rivers, lakes and canals are increasingly being polluted by microplastics – microscopic fragments of incorrectly discarded plastic products. These microplastics, beside damaging the fauna’s health, hinder CO2 sequestration on part of the sea, which contributes to global increase of temperatures. Cigarette butts and plastic bags are one of the most common culprits to microplastics in water. Correctly disposing of plastics and cigarette butts in recycling bins, and using plastic-free products, effectively contributes to reducing the microplastic pollution.

(20)

16

Figure 2. Stimulus example; tailored climate change-related news-post in Global Condition

Measures

Dependent Variables

Pro-environmental Behavioural Intentions. Participants’ pro-environmental behavioural

intentions were measured with a set of 12 questions derived from Fox et al., (2020). Participants were asked to indicate the level of likeliness to carry out certain actions to support the environment and to reduce climate change. Example item: ‘(I am likely to) Correctly dispose of cigarette butts and other plastic products’ (NB: this item was adapted to relate to the topic of stimuli A1 and B1, see Appendix). The likeliness to carry out actions was measured on a continuous level, where 1 = extreme unlikeliness to carry out an action, and 5 = extreme likeliness to do it. Reliability analyses were carried out for the items, and the scale yielded a Cronbach’s alpha of .82. For the purpose of analyses, the 12 items were computed in single variable, a mean index resulting in a continuous variable where the lowest scores indicated the lowest levels of pro-environmental behavioural intentions.

(21)

17

State Environmental Self-Efficacy. To measure perceived state self-efficacy, 4 items were

derived from Huang (2016). They measured individuals’ perceptions of their current ability and capability to take actions to effectively mitigate climate change. Example item: ‘If I now start to take actions to mitigate global warming, I can help effectively reducing climate change.’. The variable was continuous, and answers ranged from 1 (complete disagreement) to 7 (complete agreement). The scale yielded a Cronbach’s alpha of .83. A mean index of the questions was computed to indicate the level of overall state-self efficacy where the lowest score (1) indicated the lowest levels of state environmental self-efficacy, and conversely, (7) indicated the highest levels.

State Eco-anxiety. Eco-anxiety state was measured as a continuous variable, by asking 10 items based on Hogg et al. (2021). The questions were answered on a Likert scale, where 1 indicated complete disagreement with currently feeling a certain emotion, and 5 indicated complete agreement with currently feeling a certain emotion. Example item: ‘(Thinking about the posts, now) I feel nervous, anxious or on the edge’. The scale yielded a Cronbach’s Alpha of .90.

A mean index of the 10 items was computed resulting in a continuous variable where the lowest score (1) indicated the lowest levels of state eco-anxiety.

Control variables

Gender: Gender was measured as a categorical variable with the following answer options:

‘Male’, ‘Female’, ‘Nonbinary’; ‘Prefer not to say’.

Social Media Use: Participants selected which of the most popular social media platforms

they used (Instagram, Tiktok, Facebook and/or Twitter). Next, they were asked for how many hours daily they used general social media (a continuous variable). Finally, depending on the selected social media platforms, they were asked about the purposes that drove them to using such

(22)

18

platforms (Skoric et al., 2016). The following purpose options were measured with 5 different items, asking to indicate the frequency of use of each platform for specific purposes (communication, entertainment, et cetera): answers ranged from 1 (Never) to 7 (All the time). An extra item regarding environmental news was added. Example Item: ‘How often do you use Facebook to communicate with other people’.

Attention check

To make sure that participants all correctly perceived the stimulus, two questions (one for stimulus) were asked regarding the topic of the post they had just read. Participants could only indicate one of the options.

Analyses plan

To test Hypotheses 1 and 2, which presumed different levels of pro-environmental behavioural intentions and state environmental self-efficacy depending on exposure to global or local social media posts, independent samples t-tests were performed.

To explore the research question about the mediating role of state eco-anxiety, a mediation analysis using the Process MACRO (model 4; Hayes, 2012) was performed. The first mediation analysis tested the mediating role of state eco-anxiety in the relationship between the IV and state environmental self-efficacy; the second mediation analysis tested whether environmental self- efficacy mediates the relationship between eco-anxiety and the IV.

Results

Preliminary Analyses

Out of the 331 participants, n = 34 were excluded due to incomplete data, resulting in N = 297. Upon inspection of the attention check, n = 27 of the participants had failed either one or both

(23)

19

the attention checks and following the pre-registration, were excluded from analyses. This resulted in a final sample of N = 270 participants to be included in the analyses (79% female; 20.4% male;

rest – non-binary/not answered). This resulted in the following groups in the two conditions: nglobal

= 130 participants (48.1%) (coded as 0) ; nlocal =140 (51.9%) (coded as 1).

Table 1. Descriptive statistics of the key variables (N 270)

Variables Overall

M (SD)

Global condition M (SD)

Local condition M (SD)

Eco-anxiety 2.785 (.82) 2.82 (0.81) 2.76 (0.84)

State Environmental Self-Efficacy 4.84 (1.23) 4.87 (1.22) 4.80 (1.24) Pro-environmental Behavioural

Intentions

3.58 (0.65) 3.58 (0.61) 3.57 (0.70)

Social Media Use Time 3.20 (2.32) 3.34 (2.64) 3.08 (1.97)

(24)

20

Table 2. Descriptives of Social Media Purposes of Use, including Means and SDs, (N=270)

Social Media Platform Purposes of Use

Instagram (90.4%)

TikTok (37.8%)

Facebook (36.3%) Communication 4.92 (1.62) 2.61 (1.79) 3.85 (2.13) Keeping up to date 5.01 (1.45) 3.09 (1.83) 4.00 (1.76) Entertainment 5.95 (1.07) 6.69 (0.70) 4.22 (1.74) Self-expression 3.79 (1.63) 2.35 (1.71) 1.89 (1.17) Learning about important

issues (e.g., environment)

4.01 (1.59) 2.85 (1.72) 3.01 (1.86)

Note. SDs are presented in the parentheses.

Hypotheses testing

To test Hypotheses 1 and 2, which respectively presumed that exposure to local social media posts will result in higher pro-environmental behavioural intentions (H1) and environmental self-efficacy (H2), as compared with the exposure to the global condition, independent samples t- tests were performed. The analysis for hypothesis 1 found that the effect of the manipulation of locally- versus globally-focused climate change-related news-posts on pro-environmental behavioural intentions was not significant, t (268) = 0.155, p = .877, 95% CI [–.144, .169]. Thus, hypothesis 1 is rejected.

The testing of hypothesis 2, which predicted higher state environmental self-efficacy in the condition of Locally focused climate change-related news-posts as opposed to the Global condition, showed that the effect of the manipulation on state environmental self-efficacy was not

(25)

21

significant, t (268) = 0.644, p = .463, 95% CI [–.225, .364]. Thus, hypothesis 2 was likewise rejected.

Exploratory analyses

For exploratory reasons, the levels of eco-anxiety were inspected as well: much alike the previous tests, the effect of the manipulation on eco-anxiety proved statistically non-significant, t (268) = 0.610, p = .542, 95% CI [–.136, .259].

To answer the research question, a mediation analysis tested the mediating role of the eco- anxiety in the relationship between the IV (Local vs Global social media posts manipulation) and the environmental self-efficacy levels as an outcome variable, using Process MACRO model 4 (Hayes, 2012; 5000 bootstrap samples, 95% CI), with Local/Global climate-change related posts (IV) as a predictor, environmental self-efficacy (DV) as outcome variable, and eco-anxiety (M) as mediator.

For the direct effect of the manipulation (IV) on environmental self-efficacy, (DV), results showed that the manipulation did contribute to the regression model and was significant: F(2, 267)

= 10.65, p = < .001, explaining 7.39% of the variance in total environmental self-efficacy (DV).

Here, the effect of the manipulation (Locally Focused climate-change related news-posts, IV) on environmental self-efficacy (DV) was negative but non-significant, b = -.045, SE = .145, t (267) = -0.31; p = .758; CI [-.330, .240], repeating the results for the H2.

For the effect of the manipulation (IV) on the mediator eco-anxiety (M), the model was non-significant, F(1, 268) = 0.37, p = .542, explaining only 0,14% of the variance. Moreover, the effect of the manipulation (IV) on eco-anxiety (M) was negative but non-significant, b = - 0.07,

(26)

22

SE = 0.10, t (268) = -0.61; p = .542; CI [-.259, .136], meaning that the manipulation did not affect levels of eco-anxiety in participants.

Lastly, only the effect of eco-anxiety (M) on environmental self-efficacy (DV) was positive and significant, b = 0.04, SE = 0.09, t (267) = 4.59; p < .001; CI [.231, .577], suggesting that higher eco-anxiety was predicted higher environmental self-efficacy.

However, based on the non-significant effect of the IV on eco-anxiety (M) and the significant effect of the IV on environmental self-efficacy (DV), the mediation of eco-anxiety was not found.

A further analysis including total time of social media use and gender as covariates was carried out for exploratory purposes. The results unveiled a significant negative effect of social media use time on eco-anxiety (M): b = - 0.05, SE = 0.02, t (260) = -2.32; p = .021; CI [-0.092, - 0.008], indicating that the more time was spent using social media, the lower the eco-anxiety levels. Gender was also found to have a significant positive effect on eco-anxiety, b = 0.37, SE = 0.12, t (260) = 2.99; p = .003; CI [0.125, 0.611], suggesting that being female predicted higher levels of eco-anxiety. Adding these two covariates did not change the general pattern of the mediation, with the manipulation effect still non-significant, and the relationship between eco- anxiety and self-efficacy still significant.

Discussion

This study aimed to understand the possible ways in which portraying climate change- related news through social media posts, from either a local or global focus, could impact social media users’ environmental state self-efficacy, pro-environmental behavioural intentions and eco- anxiety.

(27)

23

Firstly, as for the 1^st hypothesis, the results suggest that locally-focused climate change related social media posts did not affect pro-environmental behavioural intentions when compared to exposure to the globally focused climate-change related posts. It must be mentioned that this result seems to confirm findings previous studies on the same topic (Brügger et al., 2016; Schuldt et al., 2018). This study’s results loosely mirror those from a study by Brügger et al. (2016), although the latter focused on individuals’ support for environmentally friendly policies rather than personally enacted pro-environmental behavioural intentions. Here, they had found similar results in respects of the supposed effectiveness of proximising climate change: according to their results, psychological distance did not impact levels of engagement with the climate change issue in a straightforward manner. In fact, proximisation of climate change by itself was suggested to be ineffective in increasing individual action (Brügger et al., 2016).

Nonetheless, the results of the present study also appear to clash with earlier research on the same topic (Scannel and Gifford, 2013) – however, the possible reason for this and its implications are essential to interpret these results. Scannel and Gifford (2013) had found that framing climate change from a closer perspective did have a positive effect on the engagement with the issue: however, it must be pointed out that the authors had included a variable of place attachment in the study, and found that it was the latter specifically to be directly correlated with increases in engagement. Taken together, these results seem to suggest that proximisation of climate change may be effective, but only so when the audience is particularly attached to the location where the climate change issue is framed. Important when considering this aspect, the sample for this study was mostly temporarily living in the Netherlands: then, theoretically speaking, it would be logical to assume that the levels of place attachment for that major portion

(28)

24

of the sample would be lower, thus possibly explaining the non-significant effects of proximisation in this study.

Similarly, as for the second hypothesis, locally-focused climate change-related posts did not result in higher state environmental self-efficacy than the globally-focused ones. The study by Sheppard et al. (2011) seemed to support the idea that closer information would bolster a sense of agency (and thus self-efficacy) in people. However, although according to their study there was an increase in engagement and awareness of possible mitigation practices, they did not extensively cover the increase in agency or state self-efficacy – and it might be that proximising issues does not automatically make people feel more self-efficacious in tackling them.

Thus, the results did not confirm the expectations for the two main hypotheses. Inconsistent with the construal level theory (Trope & Liberman, 2010), the diminishing of spatial distance with the aim of reducing psychological distance did not have the expected effect. Nevertheless, the results should not be taken as a definitive confirmation that proximising climate change on social media posts is of no use: a potential shortcoming of this study is that only one out of four dimensions of psychological distance has been manipulated (spatial distance), and albeit the interrelation of these dimensions, it might be that proximising only one of them might have a weaker effect in reducing psychological distance, than the manipulation of several dimensions.

Similar results had in fact been found by Brugger et al. (2016), who also manipulated spatial distance only: it might be that the results of both these studies have been partly determined by the manipulation of one dimension only.

This offers some direction for future studies on social media related to the topic of climate change, as it offers grounds to believe that manipulation of only one dimension of psychological distance may not be enough to change attitudes and perceptions, giving further confirmation of

(29)

25

Maiella et al.’s suggestions (2020). Future research should thus aim to manipulate more than one of these dimensions, and investigate whether that approach might be more successful, providing more thorough proof on the effectiveness (or lack thereof) of proximisation.

Lastly, the exploratory question on the mediation between eco-anxiety and environmental state self-efficacy showed that although no mediation was proven, there was a significant effect of eco-anxiety on environmental self-efficacy – confirming results by Bostrom et al. (2019) and Aquadro-Maran & Begotti (2021) who also found associations between self-efficacy and concern about climate change. This in itself can pave the way for future studies: eco-anxiety research is still at its outset, and exploring the possible relationships between state eco-anxiety and environmental state self-efficacy may be useful in understanding the various ways in which they operate, and devise possible strategies in entertainment and entertainment communication in order to yield the full potentials of these two factors. The current results suggested that increases in eco- anxiety counterintuitively contributed to increases in environmental state self-efficacy. Future research could evaluate studying this relationship, investigating in what ways feeling eco-anxious may contribute to state self-efficacy, or on the other hand, how feeling self-efficacious may impact one’s eco-anxiety. Moreover, the effect of the two variables in predicting pro-environmental behavioural intentions could be investigated.

Nonetheless, other interesting findings emerged from the other exploratory analyses:

firstly, social media use time appeared to be negatively related to eco-anxiety. On one hand, this might be because social media themselves reduce eco-anxiety; on the other, it could mean that the less eco-anxious one is, the more likely they are to spend time on social media. It is particularly relevant to understand which of these two scenarios actually takes place online, in order to extract the full potential of this finding. If social media prove to be a source where users go to manage

(30)

26

their emotions and moods in relation to climate change, then the possibilities of this could be harnessed by the creation of content tailored to that end. Conversely, if eco-anxious individuals avoid social media because climate change information exacerbates their anxiety, ways could be devised to convey the issues more appropriately.

Limitations and future directions

This study was not without limitations, which should be taken into account when considering its results. Firstly, there were only two conditions in the experiment: in case both posts were equally effective, only the presence of a 3^rd condition would have been able to confirm that.

Thus, a possibly fruitful suggestion for future research on the topic would be to add a 3^rd, neutral control condition in the manipulation, so to discern more specifically between framing of messages, and inspecting whether the manipulation conditions (local vs global) can create different results when compared to a neutral, control condition.

Secondly, this study did not measure variable levels before the exposure to the stimuli: any possible increase or decrease in pro-environmental behavioural intentions, state environmental self-efficacy or state eco-anxiety has not been detected, making it difficult to prove the actual effectiveness of the posts. Thus, to better investigate the extent to which locally- or globally- focused climate change-related news-posts affect the dependent variables, future studies could employ a mixed design including pre- and post-measures of the dependent variables, to provide a baseline measurement. Engagement with the climate change issue could also be used as an individual level covariate, to understand whether people with different levels of engagement react differently to information.

Moreover, this study was among the first of its kind, which lead to a few shortcomings: to ensure a dropout rate as low as possible, the manipulations were few, and as concise as possible,

(31)

27

leading to a total manipulation time of approximately 2 minutes. This might have affected the outcomes, as the manipulation might have been too short to affect participants. Furthermore, with climate change as an overarching topic, the posts only featured two specific subtopics, namely temperature increases and microplastics pollution; by featuring more varied subtopics, generalisability could be increased, also allowing for more informed results. Thus, future lab experiments are advised to include longer, and more manipulations.

Another point of interest, the inspection of social media use purposes demonstrated that most respondents in the sample used Instagram much more than Facebook to keep up with current events and environmental news: thus, it might be that the manipulation was not as effective also because taking place on a less popular platform. This suggests that, when studying users’

perceptions of climate change on social media, it might be fruitful to conduct the study on platforms with different affordances and audiences: furthermore, the inclusion of pictures or interactive material in the posts may also prove useful in decreasing psychological distance.

Moreover, this study was conducted in the Netherlands, commonly known to have high levels of awareness and engagement in the climate change issues. On the other hand, most of the participants were only temporarily living in the country, making it impossible to discern the reasons for their levels of pro-environmental behavioural intentions, self-efficacy or eco-anxiety.

It may be that possible consequences of climate change in the Netherlands have been underestimated by non-locals because their home country would not be directly affected, or that, as a foreigner, their environmental state self-efficacy level would not translate directly to pro- environmental actions because of language barriers, different bureaucracy, et cetera. Furthermore, the sample included mainly students, a very specific subpopulation. Thus, future studies should not only strive to include representative samples of the country, but also replicate the study in

(32)

28

different countries, and aim to investigate possible differences between permanent and temporary residents.

Lastly, 79% of the participants were women: when considering the significant effect of gender on eco-anxiety, this gender imbalance should be kept in mind, as the effect might be due precisely to the preponderance of women in the sample. Any future study should thus strive to include an equal number of males and females, in order to better inspect possible gender differences.

Conclusion

In conclusion, the manipulation with local and global social media posts on climate change did not affect pro-environmental behavioural intentions, environmental self-efficacy or eco- anxiety. However, the results of this study should not be taken as a further confirmation that proximisation of the climate change cannot change attitudes. Instead, they may suggest that presenting the climate change as a local issue may be equally effective as presenting it as a global concern. Nonetheless, illuminating results could emerge in future research, by studying whether manipulation of several subdimensions of psychological distance could succeed in proximising the issue of climate change. Furthermore, although no mediation effect of eco-anxiety was found, a relationship between eco-anxiety and environmental self-efficacy was found. With the increasing urgency of climate change in our society, this finding may be useful in directing future research to investigate audience’s reactions to climate change coverage, and the way in which eco-anxiety and environmental self-efficacy might interact in shaping attitudes towards climate change and related pro-environmental behaviours.

(33)

29

References

Abrahamse, W., Steg, L., Vlek, C., & Rothengatter, T. (2007). The effect of tailored information, goal setting, and tailored feedback on household energy use, energy-related behaviors, and behavioral antecedents. Journal of Environmental Psychology, 27(4), 265–276.

https://doi.org/10.1016/j.jenvp.2007.08.002

Anderson, J.T.L., Howell, E. L., Xenos, M. A., Scheufele, D. A., & Brossard, D. (2021).

Learning without seeking?: Incidental exposure to science news on social media &

knowledge of gene editing. Journal of Science Communication, 20(4), A01.

https://doi.org/10.22323/2.20040201

Aquadro Maran, D. & Begotti, T. (2021). Media Exposure to Climate Change, Anxiety, and Efficacy Beliefs in a Sample of Italian University Students. International Journal of Environmental Research and Public Health, 18(17), 9358.

https://doi.org/10.3390/ijerph18179358

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.

Englewood Cliffs, N.J : Prentice-Hall

Bandura, A. (2000). Self-efficacy: The foundation of agency. Control of human behavior, mental processes, and consciousness: Essays in honor of the 60th birthday of August Flammer, 16-30.

https://books.google.nl/books?hl=en&lr=&id=GjQik3MhmnMC&oi=fnd&pg=PA16&dq

=Self-

efficacy:+The+foundation+of+agency.&ots=0JOtYEx_ka&sig=G83qwr37tpPuTdpXxnk

(34)

30 aJrR3BH8#v=onepage&q=Self-

efficacy%3A%20The%20foundation%20of%20agency.&f=false

Berglez, P., & Al-Saqaf, W. (2021). Extreme weather and climate change: social media results, 2008-2017. Environmental Hazards, 20(4), 382–399.

https://doi.org/10.1080/17477891.2020.1829532

Bostrom, A., Hayes, A. L., & Crosman, K. M. (2019). Efficacy, Action, and Support for Reducing Climate Change Risks. Risk Analysis, 39(4), 805–828.

https://doi.org/10.1111/risa.13210

Brügger, A., Morton, T. A., & Dessai, S. (2016). “Proximising” climate change reconsidered: A construal level theory perspective. Journal of Environmental Psychology, 46, 125–142.

Carroll, S. Z., Niyonsenga, T., Coffee, N. T., Taylor, A. W., & Daniel, M. (2018). Local descriptive body weight and dietary norms, food availability and 10- year change in glycosylated haemoglobin in an Australian population based biomedical cohort. BMC Public Health 17:149. doi: 10.1186/s12889-017-4068-3

Chandler, D. (2009). The Global Ideology: Rethinking the Politics of the “Global Turn” in IR.

International Relations (London), 23(4), 530–547.

https://doi.org/10.1177/0047117809350989

Chu, H., & Yang, J. Z. (2020). Risk or Efficacy? How Psychological Distance Influences Climate Change Engagement. Risk Analysis, 40(4), 758–770.

https://doi.org/10.1111/risa.13446

(35)

31

Cunningham, S., & Craig, D. (2019). Social media entertainment. New York University Press.

Cunsolo, A., Harper, S. L., Minor, K., Hayes, K., Williams, K. G., & Howard, C. (2020).

Ecological grief and anxiety: the start of a healthy response to climate change? The Lancet Planetary Health, 4. https://doi.org/10.1016/S2542-5196(20)3014

Fox, J., McKnight, J., Sun, Y., Maung, D., & Crawfis, R. (2020). Using a serious game to communicate risk and minimize psychological distance regarding environmental pollution. Telematics and Informatics, 46, 101320.

https://doi.org/10.1016/j.tele.2019.101320

Geller, E. S. (1992). It takes more than information to save energy. American Psychologist, 47, 814-815. http://dx.doi.org/10.1037/0003-066X.47.6.814

Gil de Zúñiga, H., Jung, N., & Valenzuela, S. (2012). Social Media Use for News and Individuals’ Social Capital, Civic Engagement and Political Participation. Journal of Computer-Mediated Communication, 17(3), 319–336. https://doi.org/10.1111/j.1083- 6101.2012.01574.x

Greenpeace [@Greenpeace] (2017, February 20). If #climatechange deniers say “the climate has changed before”, show them this timeline. [Tweet]. Twitter.

https://twitter.com/greenpeace/status/833564344193421313

Greenpeace International (2019, December 11). Climate change is intensifying the bushfire smoke in Australia and continuously harming people's health. This is #climatecrisis. Act like it [Status Update]. Facebook.

https://www.facebook.com/greenpeace.international/posts/climate-change-is-

intensifying-the-bushfire-smoke-in-australia-and-continuously-/10157358537773300/

(36)

32

Harrabin, R. (2021, September 14). Climate change: Young people very worried - survey. BBC News. https://www.bbc.com/news/world-58549373

Hart, P. S., & Feldman, L. (2014). Threat Without Efficacy? Climate Change on U.S. Network News. Science Communication, 36, 325–351. https://doi.org/10.1177/1075547013520239 Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation,

moderation, and conditional process modeling.

Hickman, C. (2020). We need to (find a way to) talk about … Eco-anxiety. Journal of Social Work Practice, 34:4, 411-424, DOI: 10.1080/02650533.2020.1844166

Hogg, Stanley, S. K., O’Brien, L. V., Wilson, M. S., & Watsford, C. R. (2021). The Hogg Eco- Anxiety Scale: Development and validation of a multidimensional scale. Global

Environmental Change, 71, 102391–. https://doi.org/10.1016/j.gloenvcha.2021.102391

Horta, A., Carvalho, A., & Schmidt, L. (2017). The Hegemony of Global Politics: News Coverage of Climate Change in a Small Country. Society & Natural Resources, 30(10), 1246–1260. https://doi.org/10.1080/08941920.2017.1295497

Huang. (2016). Media use, environmental beliefs, self-efficacy, and pro-environmental behavior.

Journal of Business Research, 69(6), 2206–2212.

https://doi.org/10.1016/j.jbusres.2015.12.031

Isacsson, A., & Gretzel, U. (2011). Facebook as an edutainment medium to engage students in sustainability and tourism. Journal of Hospitality and Tourism Technology.DOI 10.1108/17579881111112430

(37)

33

Kellstedt, P. M., Zahran, S., & Vedlitz, A. (2008). Personal Efficacy, the Information

Environment, and Attitudes Toward Global Warming and Climate Change in the United States. Risk Analysis, 28(1), 113–126. https://doi.org/10.1111/j.1539-6924.2008.01010.x Lee, J.K., & Kim, E. (2017). Incidental exposure to news: Predictors in the social media setting

and effects on information gain online. Computers in Human Behavior, 75, 1008–1015.

https://doi.org/10.1016/j.chb.2017.02.018

Lehman, & Geller, E. S. (2004). Behavior Analysis and Environmental Protection:

Accomplishments and Potential for More. Behavior and Social Issues, 13(1), 13–33.

https://doi.org/10.5210/bsi.v13i1.33

Liu, S. C. (2018). Environmental education through documentaries: assessing learning outcomes of a general environmental studies course. Eurasia Journal of Mathematics, Science and Technology Education, 14(4), 1371-1381. https://doi.org/10.29333/EJMSTE%2F8365

Liu, X., Vedlitz, A., & Alston, L. (2008). Regional news portrayals of global warming and climate change. Environmental Science & Policy, 11(5), 379–393.

https://doi.org/10.1016/j.envsci.2008.01.002

Lu, Y., & Lee, J. K. (2021). Determinants of cross-cutting discussion on Facebook: Political interest, news consumption, and strong-tie heterogeneity. New Media & Society, 23(1), 175–192. https://doi.org/10.1177/1461444819899879

Maiella, R., La Malva, P., Marchetti, D., Pomarico, E., Di Crosta, A., Palumbo, R., Cetara, L., Di Domenico, A., & Verrocchio, M. C. (2020). The Psychological Distance and Climate Change: A Systematic Review on the Mitigation and Adaptation Behaviors. Frontiers in Psychology, 11, 568899–. https://doi.org/10.3389/fpsyg.2020.568899

(38)

34

Milfont, T. L., Milojev, P., Greaves, L. M., & Sibley, C. G. (2015). Socio-structural and psychological foundations of climate change beliefs. New Zealand Journal of Psychology, 44(1), 17–30.

https://link.gale.com/apps/doc/A419533833/AONE?u=amst&sid=bookmark- AONE&xid=6a940c51

Müller, P., Schneiders, P., & Schäfer, S. (2016). Appetizer or main dish? Explaining the use of Facebook news posts as a substitute for other news sources. Computers in Human Behavior, 65, 431–441. https://doi.org/10.1016/j.chb.2016.09.003

NASA [@NASA] (2021, July 13) Recognizing that climate change is an urgent global chal- lenge, we've formed a strategic partnership with @ESA to observe Earth and [Tweet].

Twitter. https://twitter.com/nasa/status/1415037877055066112

Nissen, J.M. (2019). Gender differences in self-efficacy states in high school physics. Physical Review. Physics Education Research, 15(1), 013102.

https://doi.org/10.1103/PhysRevPhysEducRes.15.013102

Olteanu, A., Castillo, C., Diakopoulos, N., & Aberer, K. (2015). Comparing Events Coverage in Online News and Social Media: The Case of Climate Change. Proceedings of the Ninth International AAAI Conference on Web and Social Media.

https://infoscience.epfl.ch/record/211214

O’Neill, S. & Nicholson-Cole, S. (2009). Fear Won’t Do It: Promoting Positive Engagement With Climate Change Through Visual and Iconic Representations. Science

Communication, 30(3), 355–379. https://doi.org/10.1177/1075547008329201

(39)

35

Passafaro, P., Livi, S., & Kosic, A. (2019). Local norms and the theory of planned behavior:

Understanding the effects of spatial proximity on recycling intentions and self-reported behavior. Frontiers in Psychology, 10, 744.doi: 10.3389/fpsyg.2019.00744

Pihkala, P. (2020). Eco-Anxiety and Environmental Education. Sustainability (Basel, Switzerland), 12(23), 10149–. https://doi.org/10.3390/su122310149

Reinecke, L., Vorderer, P., & Knop, K. (2014). Entertainment 2.0? The Role of Intrinsic and Extrinsic Need Satisfaction for the Enjoyment of Facebook Use. Journal of

Communication, 64(3), 417–438. https://doi.org/10.1111/jcom.12099

Scannell, L., & Gifford, R. (2013). Personally Relevant Climate Change: The Role of Place Attachment and Local Versus Global Message Framing in Engagement. Environment and Behavior, 45, 60–85. https://doi.org/10.1177/0013916511421196

Schuldt, J. P., Rickard, L. N., & Yang, Z. J. (2018). Does reduced psychological distance increase climate engagement? On the limits of localizing climate change. Journal of Environmental Psychology, 55, 147–153. https://doi.org/10.1016/j.jenvp.2018.02.001

Sheppard, S. R. J., Shaw, A., Flanders, D., Burch, S., Wiek, A., Carmichael, J., Robinson, J., &

Cohen, S. (2011). Future visioning of local climate change: A framework for community engagement and planning with scenarios and visualisation. Futures : the Journal of Policy, Planning and Futures Studies, 43(4), 400–412.

https://doi.org/10.1016/j.futures.2011.01.009

Skoric, M. M., Zhu, Q., Goh, D., & Pang, N. (2016). Social media and citizen engagement: A metanalytic review. New Media & Society, 18, 1817–1839.

https://doi.org/10.1177/1461444815616221

(40)

36

Soderberg, C. K., Callahan, S. P., Kochersberger, A. O., Amit, E., & Ledgerwood, A. (2015).

The Effects of Psychological Distance on Abstraction: Two Meta-Analyses.

Psychological Bulletin, 141(3), 525–548. https://doi.org/10.1037/bul0000005

Spence, A., Poortinga, W., & Pidgeon, N. (2012). The Psychological Distance of Climate Change. Risk Analysis, 32(6), 957–972. https://doi.org/10.1111/j.1539-

6924.2011.01695.x

Steg, & Vlek, C. (2009). Encouraging pro-environmental behaviour: An integrative review and research agenda. Journal of Environmental Psychology, 29(3), 309–317.

Tang, Z., Wang, P., Sui, X., & Fan, Y. (2019). Effects of psychological distance on the negative emotions of immoral events—A study based on Weibo data. Human Behavior and Emerging Technologies, 1(3), 208–215. https://doi.org/10.1002/hbe2.168

Trope, Y., & Liberman, N. (2010). Construal-Level Theory of Psychological Distance.

Psychological Review, 117(2), 440–463. https://doi.org/10.1037/a0018963

Tuitjer, L., & Dirksmeier, P. (2021). Social media and perceived climate change efficacy: A European comparison. Digital Geography and Society, 2, 100018–.

https://doi.org/10.1016/j.diggeo.2021.100018

Usher, K., Durkin, J., & Bhullar, N. (2019). Eco‐anxiety: How thinking about climate change‐

related environmental decline is affecting our mental health. International Journal of Mental Health Nursing, 28(6), 1233–1234. https://doi.org/10.1111/inm.12673

(41)

37

Valeriani, A., & Vaccari, C. (2016). Accidental exposure to politics on social media as online participation equalizer in Germany, Italy, and the United Kingdom. New Media &

Society, 18(9), 1857–1874. https://doi.org/10.1177/1461444815616223

van der Linden, S., Maibach, E., & Leiserowitz, A. (2015). Improving Public Engagement With Climate Change: Five “Best Practice” Insights From Psychological Science. Perspectives on Psychological Science, 10, 758–763. https://doi.org/10.1177/17456916155985

Van Lent, L. G., Sungur, H., Kunneman, F. A., Van De Velde, B., & Das, E. (2017). Too far to care? Measuring public attention and fear for Ebola using Twitter. Journal of Medical Internet Research, 19(6), e7219.

Verplanken, B., Marks, E., & Dobromir, A. I. (2020). On the nature of eco-anxiety: How constructive or unconstructive is habitual worry about global warming? Journal of Environmental Psychology, 72. https://doi.org/10.1016/j.jenvp.2020.10152

Yuriev, A., Boiral, O., & Guillaumie, L. (2020). Evaluating determinants of employees’ pro- environmental behavioral intentions. International Journal of Manpower, 41(7), 1005–

1019. https://doi.org/10.1108/IJM-08-2019-0387

(42)

38

Appendix

Stimuli

Figure A1. Global Condition, Stimulus 1

Figure A2. Global Condition, Stimulus 2

(43)

39

Figure B1. Local Condition, Stimulus 1

Figure B2. Local Condition, Stimulus 2

Effects of Local vs. Global Social Media News-Posts on Pro-Environmental Behavioural Intentions, Environmental Self-Efficacy and Eco-Anxiety.