Inspiration or Distraction?
Assessing the Role of Positive Exemplars in
Climate Change News Coverage
Master’s Thesis
Authored by Valerie Krall (student number: 12846953) Supervised by Dr. Andreas R. T. Schuck
Date of Completion 29/05/2020 Word Count: 8,412
University of Amsterdam, Graduate School of Communication Erasmus Mundus Master’s Joint Degree
Journalism, Media and Globalisation
Author Note
Abstract
Journalism on climate change often features famous people associated with the cause. In recent years the spotlight has been on climate activists like Greta Thunberg. However, there is a lack of academic literature on what effects this kind of news reporting has on news
consumers. Thus, this study investigates the role of positive exemplars in climate change news coverage. It was hypothesised that exposure to positive exemplars in a news article would have a more motivating effect on people than a solely informative article. A between-subjects experiment (N = 221) was conducted to compare the effects of five different versions of a news article: one solely informative one, three featuring a celebrity exemplar (Greta Thunberg, Bernie Sanders, Leonardo DiCaprio), and one featuring a non-famous positive exemplar (US American climate activist Jamie Margolin). Findings show that exposure to positive exemplars can indeed have a positive effect on people’s behavioural intentions and actual behaviour in regard to climate change through positive emotions. However, this effect was only found for exposure to two celebrity exemplars, Bernie Sanders and Leonardo DiCaprio. The results suggest that it matters what type of positive exemplar is used in climate change news coverage. Findings are discussed in light of the academic discussion on the role of positive emotions as mobilisers and research on climate change journalism in general. Implications for professionals in the field of climate change journalism are also discussed.
Key words: positive exemplars, climate change, news coverage, positive emotions,
Introduction
In the past decade climate change has become an indispensable part of the news cycle. Its media coverage, however, has been subject to various criticisms: for taking sides and giving too much space to climate change deniers, and for dramatizing as well as underplaying the issue. Over the past two years, yet another point of critique has been on news surrounding climate change activist Greta Thunberg. On the one hand, it is criticized that the media’s attention on Thunberg distracts from the real issue. On the other hand, it has been seen as a way to promote the importance of climate change. Today, climate change is arguably the most serious threat to humankind – a challenge to which there is no definite solution yet. Moreover, journalism plays a powerful role in public discourse about climate change as many people have to rely on the news for expert knowledge and due to a lack of personal experience with the issue (Höijer, 2010). Thus, studying the impact of climate change news coverage, and more specifically its focus on (famous) exemplars, is of high societal relevance.
Literature on (positive) exemplars as well as human interest framing suggests that individuals when included in a news story can have a powerful impact on people’s perception of the issue (e.g., Hong, 2013; Mastro & Tukachinsky, 2011). On the other hand, climate change is a topic of grave polarisation, especially in the USA (Dunlap, Mccright, & Yarosh, 2016) which might make it difficult to find ways to influence most people’s opinions, let alone behaviour. Moreover, climate change is often still perceived as a distant threat which means that people lack a sense of urgency (Roeser, 2012). Thus, journalism has to find new and appealing ways to report on it and keep consumers interested. Research about celebrity endorsement as well as news values suggests that celebrities can be a way to capture and keep consumers’ attention (Anderson, 2011; Harcup & O’Neill, 2017). At best celebrities can influence people’s opinions and behavioural intentions (e.g., Brown, 2010; Veer, Becirovic, & Martin, 2010). In the past decades there have been numerous celebrities associated with climate change advocacy: from Al Gore and Bono in the 2000s to Leonardo DiCaprio and
Greta Thunberg today. Given their prominence and the importance of the issue they stand for, relatively little academic research has been done on celebrities and climate change journalism. The topic has mainly been studied in the context of who are climate change celebrities, how reporting on them has increased or decreased over time and what theoretical considerations surround the “celebritisation” of climate change communication (Anderson, 2011; Boykoff & Goodman, 2009). Empirical research on the effects of climate change news about positive exemplars is still lacking attention.
This study aims to fill this void by conducting an experiment to answer the following research question: To what extent does journalism that is focused on positive exemplars
motivate people to act against climate change? The experiment compares the effects of
exposure to a solely informative article with four articles that include an exemplar. Three of them are celebrities, namely Greta Thunberg, Bernie Sanders and Leonardo DiCaprio. One article includes a non-famous exemplar, a young climate activist from the US. The study is of academic relevance as it expands research on positive exemplars onto the field of climate change communication and adds to literature on what influences people’s pro-environmental behaviour. It is also hoped that the results will help professionals in the field by giving
pointers as to what journalism that is focused on positive exemplars can and cannot achieve in motivating people to act against climate change.
The following section will lay out the theoretical background for this study by reviewing literature on positive exemplars and human interest frames, the role of celebrities in public advocacy along with other concepts that might influence the effects, namely positive emotions and narrative engagement.
Theoretical Framework Positive Exemplars in News Coverage
Exemplars are “specific, concrete instances of a category” (Ramasubramanian, 2011, p. 489). While exemplars can refer to events, in psychology literature they are most often used
to describe people which represent a certain (out-)group (e.g., Andersen & Cole, 1990, Linville, Fischer, & Salovey, 1989). In this study, positive exemplars refer to specific people publicly fighting against climate change. What makes them “positive” exemplars is that their commitment to climate change mitigation is presented in a positive light (Gonzalez, Steele & Baron, 2017).
News consumers are regularly exposed to exemplars as they are one of the most common components of media stories (Busselle & Shrum, 2003). This exposure is expected to have an influence on readers’ opinions and behavioural intentions as it can be seen as “a form of vicarious contact” (Mares, 2019, p. 1) with the exemplar. Positive or counter-stereotypical exemplars have been found to improve White people’s opinion of African Americans as well as lead to support for affirmative action policies (Power, Murphy, & Coover, 1996; Ramasubramanian, 2011; 2015). Similar effects have been found in regard to ethnic, religious or other minorities (Joyce, Harwood, & Springer, 2019; Mares, 2019; Mastro & Tukachinsky, 2011; Wojcieszak & Azrout, 2016). These effects can be explained with theories of information processing which posit (1) that humans perceive entire categories based on one or more exemplars of each category (e.g., Hintzman, 1986; Turner & Turner, 1992), and (2) that exposure to counter-stereotypical exemplars will gradually alter these perceptions (Hewstone, Johnston, & Aird, 1992; Mastro & Tukachinsky, 2011). In addition, exemplification theory states that exposure to exemplars makes people project their
judgments onto a broader context (Zillmann & Brosius, 2000). Here, media play a very
influential role as they can increase or decrease the amount of exposure to negative or positive exemplars (Mastro & Tukachinsky, 2011).
Most media effects studies about positive exemplars test the effect of a positive exemplar representing an out-group to members of the in-group. What has been investigated less is the effect positive exemplars might have on people’s opinion not towards the group but the cause they stand for, in this case climate change. The current study aims to fill this void.
As positive exemplars have been found to increase policy support concerning causes that affect their group (e.g., Ramasubramanian, 2011; 2015) it can be expected that they will also have a positive effect on people’s motivation to act themselves.
This idea is further substantiated by literature about effects of human focused news stories. Putting a human face on a news story is common practice in journalism (Bas & Grabe, 2015). While sometimes referred to as personalization or episodic1 framing, this is most
commonly known as human interest framing. The terms can be defined as distinct concepts but are often confused or used interchangeably in communication research. Research results are thus not always perfectly comparable.
Framing is the practice of selecting and highlighting certain elements in a story with the goal to promote a specific interpretation of the narrative (Entman, 2007). A news story with a human interest frame, therefore, highlights a human exemplar to show how individuals or groups are affected by a given issue (Hong, 2013; Semetko & Valkenburg, 2000). In this sense news stories about climate change that feature a positive exemplar can be seen as human interest frames. However, in order to qualify as a human interest frame, the exemplar needs to be central to the article. Merely reporting a single quote of Greta Thunberg, for example, would not constitute a human interest frame.
Human interest stories appeal to readers in multiple ways. Firstly, they are assumed to be more emotional and dramatic than other frames (Semetko & Valkenburg, 2000). Stories with a focus on personalisation have been found to evoke stronger emotional responses, for example in the context of medical advancements (Hong, 2013) and political issues (Gross, 2008). Secondly, human interest stories have stronger effects on people’s beliefs and attitudes
1 Episodic frames offer “a specific example, case study or event oriented report” (Gross, 2008, p. 171). Opposed
to this are thematic frames that consider the broader context of an issue, for example by presenting statistics or expert comments (Gross, 2008). Episodic frames differ from human interest frames as human interest frames must be focused on humans while episodic frames can also be about events or other non-human exemplars. For example, Hart (2011) studied the effects of a news article about a polar bear – an episodic frame – on people’s support for climate change mitigation policies.
than solely informative news (e.g., Hong, 2013; Shen, Ahern, & Baker, 2014). Oliver, Dillard, Bae, and Tamul (2012) showed that a narrative story focused on an individual not only leads to more compassion towards the individual but also to more beneficial behavioural intentions toward the group of which the individual is an example. The focus of the study lay on the narrative format of the article. However, the premise and results connect well to the exemplar literature as one story of a positive exemplar led to an improvement in attitudes and
behavioural intentions towards the whole group. Conversely, human interest frames have also been shown to have potential negative effects on recipients as they can distract from the real issue and make people blame individuals instead of society for a given problem (Iyengar, 1991).
Based on this discussion it can be argued that positive exemplars can positively change people’s attitudes and behavioural intentions. However, these effects need to be investigated further in the context of climate change. Thus, the following main hypothesis is proposed:
H1a: People who read an article with a positive exemplar are more motivated to act against climate change than people who read a solely informative article.
While all positive exemplars can be expected to have this effect, a stronger influence can be assumed for celebrity exemplars.
Celebrities in Climate Change Coverage
Over the past decades, celebrities have increasingly become more prominent in voicing their opinions about climate change and oftentimes appear as legitimate political actors in news coverage about it (Anderson, 2011; Boykoff & Goodman, 2009). Celebrities are a common news factor and thus an appealing news item for journalists (Harcup & O’Neill, 2017). In this study, the effect of three celebrities will be tested who have been in the public eye for their climate change advocacy: Leonardo DiCaprio, Bernie Sanders and Greta Thunberg. They were chosen because they represent different types of climate change
celebrities. According to Boykoff & Goodman (2009), celebrities can be defined by the “political or social determinants that shape their actions” (p. 400). Leonardo DiCaprio is a celebrity actor who has narrated and produced several movies about global warming and is one of the most famous faces of climate change (Boykoff & Goodman, 2009; Leas et al., 2016). Bernie Sanders belongs to the category of celebrity politicians. Having run for the US presidency in 2020, he supported the Green New Deal and was the Democratic candidate with the most extensive and expensive plan to mitigate climate change (Friedman, 2019). Greta Thunberg is possibly the most famous climate activist at this time (Alter, Haynes, & Worland, 2019). In 2018, she started a school strike in Sweden which successively inspired millions of people to attend global climate demonstrations and made her the face of the movement “Fridays for Future” (Alter et al., 2019). However, she does not belong to any of the
categories proposed by Boykoff and Goodman (2009). Their categorisation lacks a mention of celebrity activists, namely persons who become famous because of their climate activism. Considering the recent increase in prominence of young climate activists, this category seems to be an important addition and justifies including Greta Thunberg in this study.
Due to their fame and the human interest angle they provide, celebrities can bring the issue of climate change closer to news consumers for whom the topic feels far away from their reality (Anderson, 2011). A climate change celebrity can be portrayed as a “heroic individual” (Boykoff & Goodman, 2009, p. 396) which taps into the idea of positive exemplars. However, whether news coverage about climate change celebrities is effective remains a controversial issue. While some argue that celebrities can raise awareness about climate change and inspire action (e.g., Street, 2004), others see the “celebritisation” of climate change as distracting. They fear that it diverts attention away from the actual problem (Boykoff & Goodman, 2009). This argument shows similarities to the one made by scholars who argue that human interest stories can distract people from the broader issue (Iyengar, 1991).
Celebrities are known to support charitable causes, endorse politicians and advertise products. Research in the field of advertising shows that celebrities significantly increase brand awareness, brand attitudes as well as purchase intentions in comparison to non-famous testimonials (for a literature review see Bergkvist & Zhou, 2016). They have also been found to influence people’s attitudes and behaviour in relations to other issues such as public health, humanitarianism, political engagement, and environmental behaviour (Austin, Vord,
Pinkleton, & Epstein, 2008; Brown, 2010; Brown & Basil, 1995; Jackson & Darrow, 2005; Veer et al., 2010; Wheeler, 2009). It can, therefore, be hypothesised that:
H1b: News articles featuring celebrity exemplars have a more motivating effect than articles featuring non-famous exemplars.
These main effects are expected to be influenced by other variables that either
moderate or mediate the relationship (see Figure 1). These will be explained in the following.
Figure 1
Celebrity Involvement
A celebrity’s influence might depend on how well he or she is known and liked by the consumer or recipient. Different concepts have been considered to explain the effects of celebrity influence, amongst them identification with the celebrity (e.g., Austin et al., 2008; Brown & Basil, 1995), parasocial interaction (e.g., Brown, 2010) and likeability (e.g., Veer et al., 2010; Wheeler, 2009). These factors have been aggregated into the concept of celebrity involvement which describes “a type of emotional and psychological attachment that occurs through prolonged media or event exposure” (Brown & de Matviuk, 2010, p. 361). It is a multi-dimensional construct that consists of three components: affinity, parasocial relationship (PSR) and identification (Wen & Cui, 2014).
Affinity quite simply describes the condition of an individual liking a celebrity (Giles, 2002). The concept of parasocial interaction was first introduced by Horton and Wohl (1956). It describes a strong and intimate connection an individual feels toward a celebrity. The individual will “desire a closer relationship with the celebrity, such as friendship or even a romantic connection. Moreover, the fan would perceive some similarities between his or herself and the celebrity” (Wen & Cui, 2014, p. 413). The concept of the third component, identification, is based on Kelman’s work (1958). It describes an internal process in which an individual adopts attitudes, beliefs, and behaviours from the celebrity (Wen & Cui, 2014).
Extant literature that investigates celebrity involvement finds overwhelming support for its effect on people’s opinions, attitudes and behavioural intentions (Brown, 2010; Brown & Basil, 1995; Brown, Basil, & Bocarnea, 2003; Wen & Cui, 2014)2. However, most of the
studies are based on surveys which are not apt to establish causal relationships. The study at hand will extend the research on celebrity involvement by testing it as a moderating factor in an experiment. It is expected that:
2 Brown and colleagues’ research mainly focuses on the concepts of identification and parasocial interaction.
They use the term “audience involvement” (Brown, 2010; Brown & Basil, 1995; Brown, Basil, & Bocarnea, 2003; Brown & de Matviuk, 2010).
H1c: Stronger celebrity involvement will increase the effect of exposure to celebrity exemplars on recipients’ motivation to act against climate change.
Emotions
One key reason for why positive exemplars in climate change coverage might have a stronger impact on people’s motivation to act than neutral articles are emotions. Theories on human interest frames as well as celebrity involvement assess emotionality to be an important factor. However, often it is not specified whether these emotions should be positive or
negative. For example, Gross’ (2008) measures pity and sympathy as indicators of empathy in his study about framing effects on policy opinions. However, while empathy is clearly a positive emotion, pity cannot easily be categorised as either positive or negative. Ambiguities like this provide the opportunity to take a closer look at what kind of emotions positive exemplars might evoke and what the effects of this will be.
Discrete emotions are “internal, mental states representing evaluative, valenced reactions to events, agents, or objects that vary in intensity” (Nabi, 2003, p. 226). They are regarded as being “short-lived, intense, and directed at some external stimuli” (Nabi, 2003, p. 226). While traditionally, emotions were seen as distinct from and opposite of rationality, it is now agreed that emotions are “natural and necessary parts of all meaning-making” (Höijer, 2010, p. 720). Emotions are a decisive factor in risk perception which makes them a highly relevant concept to study in climate change communication (Roeser, 2012). In addition, it has been argued that too little emotional involvement with climate change might explain low concern and inaction in regard to climate change mitigation (Feldman & Hart, 2018).
Different discrete emotions have been found to influence people’s attitudes and opinions about climate change as well as their motivation to act (Wang, Leviston, Hurlstone, Lawrence, & Walker, 2018). While negative emotions, first and foremost fear, have been investigated extensively, more research on positive emotions is required as they might be more encouraging and mobilising (Smith & Leiserowitz, 2014). Positive emotions such as
enthusiasm, pride and hope can have mobilising effects, for example in the context of political participation (Brader, 2005; Valentino, Brader, Groenendyk, Gregorowicz, & Hutchings, 2011). They have also been found to play an important role as mediators of framing effects (Lecheler, Bos, & Vliegenthart, 2015). Furthermore, Lecheler, Schuck and De Vreese (2013) showed that positive emotions can be evoked with positive news frames. Thus, positive exemplars in climate change journalism can be assumed to lead to more positive emotions amongst recipients.
Resulting from these considerations, the study at hand will focus on assessing positive emotions as mediators to motivate people to take action against climate change. Amongst all positive emotions, hope has been investigated as the most promising emotion in encouraging climate change behaviour. People who feel hopeful in regard to climate change have been found to be more likely to engage in proenvironmental behaviour (Ojala, 2012) and support national policies to mitigate climate change (Smith & Leiserowitz, 2014). However,
Chadwick (2015) found that feeling hopeful did not have a positive effect on climate protection behavioural intentions.3 The author explained this result with the lack of
recommended action clues in the stimuli messages that would have been needed to motivate people. Apart from hope, other positive emotions might also play an important role in
increasing people’s motivation to act against climate change, for example pride and gratitude (Markowitz & Shariff, 2012), or enthusiasm (Feldman & Hart, 2016).
Several emotion studies in the field of politics treat multiple positive emotions, such as hope, pride, joy, happiness, confidence and enthusiasm, as one category of “enthusiasm” as they have been found difficult to be differentiated in self-reports, and show very similar behavioural responses (Brader & Marcus, 2013; Vasilopoulou & Wagner, 2017). Being exposed to a positive exemplar engaging in climate change mitigation can be expected to evoke several of these emotions in people. The optimistic message appeals to solidarity,
common goals and the prospect of a cleaner and more peaceful world. Also, positive
exemplars can be perceived as encouraging to everyone who wants to or is already engaging in climate-friendly behaviour. This “positive feedback” makes people pursue the behaviour, spurred on by the feeling that the goal is within their reach (Brader & Marcus, 2013). Derived from the above, two connected hypotheses will be tested:
H2a: Articles featuring a positive exemplar will lead to more positive emotions than the neutral article.
H2b: Stronger positive emotions will lead to a stronger motivation to act against climate change.
Further, emotions are essential for celebrity involvement as it is a “type of emotional and psychological attachment” (Brown & de Matviuk, 2010, p. 361). Especially, feeling empathy and sympathy is inherent to the concept of identification (Cohen, 2001). Thus, it can be expected that involvement with a celebrity will moderate recipients’ emotional response:
H2c: Stronger involvement with a celebrity will lead to stronger feelings of empathy and sympathy which in turn will increase recipients’ motivation to act against climate change.
Narrative Engagement
Not only can readers be involved with a celebrity they read about, the stories themselves can evoke feelings and make people feel involved in the narrative. This is what the concept of narrative engagement entails. Introduced by Busselle and Bilandzic (2008; 2009) it describes how people process stories by becoming “engaged, transported, or
immersed in a narrative” (Busselle & Bilandzic, 2009, p. 321). It is derived from the concepts of transportation and identification and encompasses four dimensions: narrative
understanding, attentional focus, emotional engagement and narrative presence (Busselle & Bilandzic, 2009). Narrative understanding describes how easily one can comprehend a story. Attentional focus means that the reader concentrates on the narrative and is not distracted.
The process of emotional engagement occurs when the narrative evokes emotions that the reader feels with or for one or more characters in the story, for example fear or sympathy.4
Narrative presence describes the sensation of someone losing their sense of awareness and entering the time and place of the story they are reading (Busselle & Bilandzic, 2009).
The stimulus in this study is not explicitly narrative as it is supposed to represent a common news article. However, it can be assumed that articles about a positive exemplar will be perceived as more narrative than a neutral article. Characters constitute an essential
element of narratives and are assumed to be the main reason for why stories evoke emotions and enable identification amongst recipients (Busselle & Bilandzic, 2009). Furthermore, news stories with a human focus have been found to facilitate a better understanding for readers (Bas & Grabe, 2015). Thus, it is hypothesised that:
H3a: Articles about a positive exemplar will lead to stronger narrative engagement than the neutral article.
When a story successfully engages readers, it can influence people’s beliefs and attitudes about the issue(s) discussed (e.g., Bilandzic & Sukalla, 2019; Busselle & Bilandzic, 2009). In the context of climate change Bilandzic and Sukalla (2019) showed that narrative engagement with a dystopian fiction film about the future of a warmer earth, affects people’s intentions to engage in climate change mitigation behaviour. Most research on narrative engagement has focused on fictional narratives and there are only a few studies in the field of journalism (van Krieken, Hoeken, & Sanders, 2015). These studies usually test the difference between narrative and non-narrative forms of journalism (e.g., Oliver, Dillard, Bae, & Tamul, 2012; Shen et al., 2014). Narrative engagement with news stories has been found to influence attitudes on political issues (Shen et al., 2014) as well as behavioural intentions (Oliver et al.,
4 This component is similar to the notion of identification (a main concept of celebrity involvement) where
individuals identify with the characters of the narrative and interpret the story “as if the events were happening to them” (Cohen, 2001, p. 245).
2012). This makes it plausible to assume that narrative engagement can have an effect on people’s motivation to act against climate change.
H3b: Stronger narrative engagement will positively influence people’s motivation to act against climate change.
Methodology Study Design
To investigate the effects positive exemplars have on people’s motivation to act against climate change, an online experiment was conducted in April 2020. In a single-factor, post-test only, between-subjects design, participants were randomly assigned to one of five conditions. Each condition consisted of an article about a current climate change topic which discussed either the comments of a celebrity exemplar or a non-famous exemplar or was solely informative (control condition). The questionnaire was set up on the platform Qualtrics. Sample
The sample was comprised of 221 people. Convenience sampling was used as this method was best suited to the researcher’s time and resource constraints. As the survey poses quite a universal research question and the design is not specifically directed at a certain population, anyone above the age of 18 could participate.5 Over the course of two weeks,
participants from the social network of the researcher were contacted using social media and email. Out of 365 participants 135 did not finish the survey (37%).6 The aimed sample size
was set at 200 so that every experimental group would have around 40 participants. Due to the non-random sampling technique the sample is not representative of the overall population. Procedure
In the beginning of the survey, participants were informed about the purpose of the study (perception of online news stories about current political issues), the conditions, and
5 The only other requirements were internet access and sufficient English skills.
6 Nine participants who took an extremely short or long time to finish the survey were excluded from the
their rights. After they confirmed their consent, participants answered questions about their knowledge of, opinion about, and involvement with the three celebrities (Sanders, Thunberg, DiCaprio) as well as their political ideology. Participants were then exposed to one of five conditions: an article either with a celebrity exemplar (three different versions), a non-famous exemplar, or without an exemplar (control condition). Respondents were required to stay on the article page for a minimum of 20 seconds to ensure thorough reading of the article. After the stimulus, an open-response question asked participants about their initial reactions. Hereafter the manipulation check (see below) and a question about respondents’ emotional response followed. The post-test further entailed questions about narrative engagement as well as behavioural intentions and actual behaviour. Finally, participants’ demographic data were collected, and respondents were debriefed (see Appendix A for questionnaire). Before the data analysis a randomization check was conducted with regards to demographics. No between-group differences were found.7
Stimulus Material
Five experimental conditions were created, four of them featuring a positive exemplar. Greta Thunberg, Leonardo DiCaprio and Bernie Sanders represent the celebrity exemplars. As a fourth exemplar but non-celebrity Jamie Margolin was chosen, a young climate activist from the USA.
To ensure high external validity the article was put together with material from real news stories. It was designed in the layout of the news agency Reuters to enhance credibility (see Appendix B). The article focuses on a UN climate report about the consequences of climate change. All four exemplar conditions consist of the same structure and the same quotes. Changes have been made only to describe the respective person. The exemplar’s age and occupation are mentioned as well as their role in the fight against climate change.
7 The randomisation test showed that there were no differences between conditional groups in regard to gender
(χ2 (4) = 4.66, p > .05), native and non-native speakers (χ2 (4) = 1.52, p > .05), age F (4, 216) = 1.67, p > .05,
Considerable effort was taken to ensure that the descriptions did not make any exemplar seem better, more likeable, more successful or impactful than the others. As the aim of the study was to find out whether there is an effect on people’s motivation to act, the exemplar is quoted twice referring to the need for individual action. The control article consists of the same information and quotes, only that the quotes are attributed to the UN report and not to a specific person. All articles were about the same length (between 276 and 293 words). Manipulation Check
To check successful manipulation respondents were asked to identify any person that was explicitly mentioned in the article. All four exemplars as well as a “none of the above” option was given. Multiple selections were possible. 202 participants (91.4%) selected the correct option.8
Mediators
Emotions. Participants indicated how strongly the article made them feel different positive and negative emotions on a scale from 1 (not at all) to 7 (very much). All positive emotions (enthusiasm, hope, happiness, confidence, empathy, sympathy) were averaged into a ‘positive emotions’ index (α = .845, M = 3.48, SD = 1.3). In comparison, negative emotions consisted of “disgusted”, “angry”, “afraid”, “worried” and “irritated” (α = .803, M = 3.13, SD = 1.28).
Narrative engagement. The measurement of narrative engagement was based on the validated scale by Busselle and Bilandzic (2009). However, the original scale was created for fictional narratives rather than news articles. Therefore, adapted items from van Krieken, Hoeken and Sanders (2015) were used. Three out of the four sub-scales were measured: attentional focus (e.g., “I found my mind wandering while reading the news article”), narrative presence (e.g., “During reading, I visualised what was described in the news
8 As there were no differences between the results of participants with correct and false manipulation checks, all
article”9), and emotional engagement (e.g., “The article affected me emotionally”).
Participants rated their agreement with the statements from 1 (Completely disagree) to 7 (Completely agree). Following van Krieken et al. (2015) the sub-scale of narrative
understanding was not measured as it was not deemed relevant for a brief and fact-oriented news article and for the dependent variables included in this study. Responses were averaged to create a narrative engagement scale and three subscales (Narrative engagement: α = .608,
M = 3.49, SD = 0.92; Attentional focus: α = .854, M = 3.4, SD = 1.72; Narrative presence: α =
.77, M = 3.35, SD = 1.5; Emotional engagement: α = .718, M = 3.72, SD = 1.47). Moderators
Celebrity involvement. Celebrity involvement was measured with 9 items from two validated scales, the Celebrity-Persona Parasocial Interaction scale (Bocarnea & Brown, 2007) and the Celebrity-Persona Identification Scale (Brown & Bocarnea, 2007). The
selection of items was based on a shorter scale by Wen & Cui (2014) who had introduced the three-dimensional construct of celebrity involvement (affinity, parasocial relationship, identification). Items included “Every time when Sanders/Thunberg/DiCaprio appears in the media, he/she catches my attention” (affinity), “Sanders/Thunberg/DiCaprio understands the kinds of things I want to know” (PSR), and “I advocate the same things that
Sanders/Thunberg/DiCaprio advocates” (identification). Participants answered the celebrity involvement questions for all three celebrities separately. Answers were measured on a seven-point Likert scale (1 = Strongly disagree, 7 = Strongly agree). As reliability tests did not yield acceptable reliability for all sub-scales, it was decided to create one scale for each celebrity (Thunberg: α = .641, M =3.75. SD = 0.93; Sanders: α = .746, M = 3.6. SD = 1.15; DiCaprio:α = .754, M =3.66. SD = 1.28)10.
9 The phrasing of this item was slightly changed after the pre-test to clarify its meaning. Instead of “I visualised”
it says “I saw before me” in the original scale (van Krieken et al., 2015).
10 The reliability for celebrity involvement with Greta Thunberg is still reasonably reliable but much lower than
Dependent Variables
Behavioural intentions. Participants’ motivation to act against climate change was measured with 9 items. Respondents were asked about how likely it was for them to engage in different behaviours related to climate change in the next six months. Items ranged from asking about low cost behaviours such as “talk about the issue of climate change with family, friends or acquaintances” and “pay more attention to whether your own behaviour in your everyday life contributes to climate change” to more costly behaviours such as “take climate-friendly transportation even if it takes longer” or “take part in a protest/rally/demonstration regarding the issue of climate change”. All answers were given on a seven-point Likert scale ranging from 1 (Very unlikely) to 7 (Very likely). They were averaged and collapsed into one behavioural intentions scale (α = .842, M = 4.89, SD = 1.14).
Actual behaviour. To measure actual behaviour, respondents were asked to indicate if they wanted to read more about climate change, sign up for a newsletter with official
information about climate change or donate to an NGO working to mitigate climate change once the questionnaire was over. Participants were debriefed later that their responses did not have any consequences which means that they did not actually engage in real behaviour. However, at the time of answering the questions, respondents did commit to the behaviour which is why this operationalisation is considered to simulate actual behaviour. The number of ‘Yes’ answers was summed up to create an individual score ranging from 0 to 3 (M = 0.86,
SD = 0.9).
Control Variables
Political ideology. Political ideology was measured on a 11-point scale from 0 (left) to 10 (right) (M =3.88, SD = 2.07).
Age. The average age was 29.08 which shows that it was a comparatively young sample. The youngest participant was 18 years old, the oldest 62.
Gender. More than half of the participants (57.4%) were women, 40.7 percent were men. Four people indicated that they did not identify with either category, one chose to self-describe (0.5%) and three preferred not to say (1.4%).
Education. The majority of the sample (102 participants, 46.1%) had a bachelor’s degree as their highest level of education. 29.9 percent held a master’s degree or similar and 4.1 percent a doctorate title. 35 participants’ highest education was a high school degree or equivalent (15.8%). 9 participants indicated a different type of education, for example a German university diploma or associate degree (4.1%).
Native speakers. Out of 221 participants 38 were English native speakers (17.2%). Data Analysis
The data were analysed with the statistics software SPSS, version 26. Moderation and mediation analyses were conducted with the PROCESS macro by Andrew Hayes, version 3.4.
Results
The main hypothesis of this study states that people who read an article about a positive exemplar are more motivated to act against climate change than people who read a solely informative article. To test this hypothesis, a dichotomous variable for the two groups exemplar and non-exemplar was created. The analysis showed that exposure to positive exemplars led to a higher score for behavioural intentions (M = 4.9, SD = 1.14.) than the control group (M = 4.83, SD = 1.18), t(219) = -0.346, p > .0511. Similarly, participants
exposed to positive exemplars had a higher score for actual behaviour (M = 0.87, SD = 0.89) compared to those who read the neutral article (M = 0.77, SD = 0.93), t(219) = -0.658, p > .05. However, both effects were not significant which is why hypothesis 1a cannot be confirmed.
Hypothesis 1b poses that news articles about celebrity exemplars have a stronger effect on behavioural intentions and actual behaviour compared to non-famous exemplars. Another dichotomous variable was created that distinguishes between the three celebrities and
the non-famous exemplar. The analysis cannot confirm the hypothesis: Exposure to celebrities resulted in stronger behavioural intentions (M = 4.93, SD = 1.19) compared to the
non-famous one (M = 4.81, SD = 0.95), t(219) = -.0346, p > .05, but the effect was not significant. In terms of actual behaviour exposure to celebrities (M = 0.87, SD = 0.9) did not differ from exposure to the non-famous exemplar (M = 0.88, SD = 0.88), t(219) = -0.658, p > .05.12
To test if there is a moderating effect of celebrity involvement on motivation to act against climate change (H1c) the three celebrity involvement indices were tested separately in three moderation analyses per DV (behavioural intentions and actual behaviour). For neither
behavioural intentions (b = -.27, SE = .27, p > .05), 95% CI: [-0.8, 0.25] nor actual behaviour
(b = -.31, SE = .22, p > .05), 95% CI: [-0.12, -0.73] was there a moderating effect of celebrity
involvement with Greta Thunberg for participants exposed to her article. Involvement with
Bernie Sanders did not have a moderating effect on the article about him either (behavioural
intentions: b = .08, SE = .23, p > .05), 95% CI: [-0.38, 0.54] (actual behaviour: b = .11, SE =
.18, p > .05), 95% CI: [-0.24, 0.47] and the same is true for Leonardo DiCaprio (behavioural
intentions: b = -.32, SE = .19, p > .05), 95% CI: [-0.69, 0.04] (actual behaviour: b = -.03, SE
= .15, p > .05), 95% CI: [-0.33, 0.42]. Thus, celebrity involvement did not function as a moderator and the hypothesis has to be rejected.
Positive Emotions
The second set of hypotheses is concerned with f positive emotions. In order to test if exposure to positive exemplars leads to more positive emotions (H2a) which in turn affects
behavioural intentions and actual behaviour (H2b), a mediation analysis was conducted.
When compared to the control group (informative article), results showed significant indirect effects for two celebrities, Bernie Sanders and Leonardo DiCaprio, on behavioural intentions via the mediator positive emotions (Sanders: b = .11, SE = .06), 95% CI: [0.01, 0.25]
12 The results for the main effects did not differ when each celebrity group was compared to the other
(DiCaprio: b = .1, SE = .06), 95% CI: [0.01, 0.25]. Reading about Sanders (b = .77, SE = .28, p = .006) and DiCaprio (b = .72, SE = .28, p = .01) led to more positive emotions. Positive
emotions increased behavioural intentions (b = .14, SE = .06, p = .02) to fight climate change.
However, the mediation effects are quite weak. Moreover, the mediation effect is only significant in comparison to the control article. When compared to Thunberg and Margolin (non-famous exemplar) there were no significant mediation effects (see Appendix C Table 1).
In regard to actual behaviour the results are similar. For exposure to Sanders and DiCaprio there is an indirect effect on actual behaviour via positive emotions (Sanders: b = .11, SE = .06), 95% CI: [0.02, 0.25] (DiCaprio: b = .1, SE = .06), 95% CI: [0.01, 0.23]. As shown above, exposure to Sanders and DiCaprio led to more positive emotions. Positive
emotions in turn led to more actual behaviour (b = .14, SE = .05, p = .002). Like in the case of behavioural intentions, however, there were no significant mediation effects in comparison to
Thunberg and Margolin (see Appendix C Table 2). Thus, hypothesis 2a can be partly
confirmed for exposure to Sanders and DiCaprio, but not for Thunberg and Jamie Margolin. Hypothesis 2b was confirmed even though the effects of positive emotions on behavioural
intentions and actual behaviour are quite weak.13
To test the hypothesis that exposure to celebrities leads to stronger feelings of empathy and sympathy when celebrity involvement is higher, and that this in turn increases behavioural
intentions and actual behaviour (H2c), moderated mediation analyses were performed.
Neither of the three celebrity conditions (when compared to the control group) was part of a significant moderated mediation where celebrity involvement moderates the relationship between exposure to the celebrity and behavioural intentions or actual behaviour via
sympathy or empathy (see Appendix C Table 3, 4). The hypothesis has to be rejected.
13 It was also tested whether exposure to positive exemplars reduced negative emotions which in turn would
affect behaviour. No significant indirect effects were found in regard to actual behaviour. In the model with
behavioural intentions, exposure to Bernie Sanders was significantly mediated by negative emotions (b = -.12,
SE = .07), 95% CI: [-0.28, -0.01] so that it reduced negative emotions which then led to stronger behavioural
Narrative Engagement
Hypothesis 3a and b state that news articles about positive exemplars will lead to stronger narrative engagement compared to the neutral article which will in turn increase people’s motivation to act against climate change. In order to test this, simple mediation analyses were performed. As the subscales of narrative engagement were more reliable than the combined scale itself, three different analyses were conducted for both DVs to test
attentional focus, narrative presence and emotional engagement as possible mediators.14
Results showed no indirect effect for either of the three (see Appendix C Table 5, 6) which means that narrative engagement does not mediate the relationship between positive exemplars and behavioural intentions and actual behaviour. Moreover, neither of the exemplar conditions significantly influenced either of the components (attentional focus,
narrative presence, emotional engagement) (see Appendix C Table 7)
However, two components of narrative engagement, narrative presence and
emotional engagement, have an effect on the dependent variables. Narrative presence (b =
.12, SE = .05, p = .02) and emotional engagement (b = .22, SE = .05, p < .001) positively influence behavioural intentions, but attentional focus has no effect (b = .03, SE = .05, p > .05). Mirroring these results, narrative presence (b = .09, SE = .04, p = .033) and emotional
engagement (b = .15, SE = .04, p < .001) also increase actual behaviour, whereas attentional focus does not (b = -.04, SE = .04, p > .05). While hypothesis 3a has to be rejected, hypothesis
3b can be partly confirmed. Narrative presence and emotional engagement positively influence behavioural intentions and actual behaviour (i.e., b path of mediation model).
Narrative engagement does not function as a mediator, however, because it was not triggered
by any of the stimuli (i.e., a path of mediation model).
14 Simple analyses were chosen over a multiple mediation analysis as the different mediators were highly correlated
Discussion and Conclusion
This study set out to investigate to what extent positive exemplars in climate change news coverage increase people’s motivation to act against climate change. An experiment was conducted in which people’s behavioural intentions and actual behaviour in regard to climate change mitigation were measured after exposing them to a news article containing a positive exemplar or no exemplar. Furthermore, potential differences in effects between exposure to celebrity and non-famous exemplars were tested. Contrary to expectations, exposure to a positive exemplar – celebrity or non-famous – had no direct impact on behavioural intentions and actual behaviour. However, an indirect effect was found: Exposure to two of the celebrity exemplars increased positive emotions which in turn led to more behavioural intentions and actual behaviour.
These main findings shed light on the literature that built the conceptual framework of this study. Research about positive exemplars was employed to argue for a direct effect of positive exemplars. However, while it has been shown that positive exemplars can improve people’s opinion and behavioural intentions towards a group of people (e.g., Mastro & Tukachinsky, 2011; Power, Murphy, & Coover, 1996), this might not work in case of a cause (i.e., climate change). In this respect, research on human interest framing was more predictive of the results as it acknowledges emotions to be an important mechanism in influencing behavioural intentions (Gross, 2008; Hong, 2013; Semetko & Valkenburg, 2000). This study suggests that positive exemplars – like human interest frames – can lead to positive emotions which then increase people’s motivation to act against climate change.
Interestingly, this was only true for two exemplars: Bernie Sanders and Leonardo DiCaprio. Exposure to Greta Thunberg and the non-famous exemplar, Jamie Margolin, did not lead to significantly more positive emotions compared to the control (neutral) article. This shows that a study design testing the effects of exposure to different types of celebrities is indeed justified as it seems to matter what type of exemplar people are exposed to.
The discrepancy in effects between exposure to the two celebrities and the non-famous exemplar is in line with expectations in this study. Advertising literature suggests that
exposure to celebrities has a stronger effect on behavioural intentions than non-famous
persons (e.g., Friedman & Friedman, 1979). What is less apparent is why exposure to the third celebrity, Thunberg, did not have the same effect as exposure to Sanders and DiCaprio. This could be explained with a possible ceiling effect: Thunberg is so prominent in the climate change debate that her appearance in a news article cannot evoke strong positive reactions in readers (anymore).15 Sanders and DiCaprio, on the other hand, are not primarily known for
their climate advocacy. Hence, getting introduced to their activism might have a more positive effect than reading about someone already associated with it. However, more
research testing other celebrity actors and politicians is required to confirm that the effects are not related to DiCaprio and Sanders personally, but the category they belong to. Secondly, the fact that Thunberg’s overall climate communication is connotated rather negatively might play a role. Thunberg is criticised for emphasising the negative consequences of climate change rather than possible solutions. This could mean that even if she is portrayed as spreading an optimistic message in a particular news article, readers might perceive this as less positive due to their preconceived notions of her.16 Connected to this point, it might also
be that Thunberg is too polarising and that people’s opinions on her are unlikely to change: someone who already likes her cannot get a more positive opinion on her, and someone who dislikes her will not change their opinion due to a positive depiction in an article. Thirdly, gender and age differences between the exemplars could also matter. Interestingly, both young women exemplars did not have a significant effect on positive emotions. There is ample research showing that women are seen more negatively in authority positions than men
15 Ceiling effects have been used to explain findings in communication research, for example in regard to media
consumption and (political) knowledge (De Vreese & Boomgarden, 2006; Eveland & Scheufele, 2000).
16 In order to control for this, source and message perception were measured in the experiment. The analysis
showed that perceptions did not significantly differ for either of the three celebrities. It is, of course, possible that an item which was not measured could explain the differences.
(e.g., Jacobson, Antonelli, Winning, & Opeil, 1977; Rudman & Kilianski, 2000). However, this might not necessarily apply to celebrities. For instance, no gender differences were found in a study that examined celebrity endorsements of politicians and their effects on voting intentions (Morin, Ivory, & Tubbs, 2012). Further investigations are needed here as the
current study can only speculate about the influence of a gender bias. Moreover, adults tend to perceive young people and adolescents more negatively than other adults (e.g., Chan et al., 2012; Trzesniewski & Donnellan, 2014). This could explain why exposure to the two younger celebrities (Thunberg and Margolin) did not evoke significantly more positive emotions. It also suggests that there are other relevant factors that may matter for the effects of positive exemplars than just the category the celebrity falls into (i.e., politician, actor, activist). These relate to the characteristics of the person as such and need to be addressed in future research. For that, it is necessary to test different celebrity exemplars and be clear about the theoretical foundations based on which they were considered to differ from each other in relevant ways. Linking these considerations back to theory, it appears that the controversy of whether celebrities distract from the real issue or inspire action (Boykoff & Goodman, 2009) has to be answered differently for different celebrities. While Sanders and DiCaprio inspired action, albeit only to a modest degree, Thunberg might be associated “too much” with climate change to make a difference. However, it has to be stressed that the effects for Sanders and DiCaprio were only significant in comparison to the control article. In fact, they did not differ
significantly from exposure to Thunberg. However, exposure to Thunberg had no significant effect compared to the control group. The conclusions drawn here should, therefore, not be overstated. On the other hand, even if the effects are not very strong in terms of magnitude, they seem to be robust as they were found for behavioural intentions and actual behaviour.
Next to positive emotions, narrative engagement was measured as a possible mediator. In contrast to positive emotions, neither of the components of narrative engagement
current study. This stands in contrast to research on narrative engagement (e.g., Bilandzic & Sukalla, 2019; Busselle & Bilandzic, 2009). A possible reason for this discrepancy is the difference in study designs. Most studies that apply the concept of narrative engagement to journalism compare narrative story forms to non-narrative ones (e.g., Oliver et al., 2012; van Krieken et al., 2015; Shen et al., 2014). The premise of the current study was to extend this research: It was hypothesised that a non-narrative article would be more narrative – and thus more engaging – simply by focusing it on a person (i.e., a positive exemplar). The results clearly refute this hypothesis. Hence, there might a very simple reason for why neither of the experimental articles led to more narrative engagement: they were not written as narratives.17
However, the experiment also showed that in cases where recipients reported a stronger narrative engagement (i.e., narrative presence and emotional engagement) it did increase behavioural intentions and actual behaviour. This suggests that if a news article does indeed trigger narrative engagement it could be expected to have a significant effect on people’s motivation to act. In a follow-up study it would be prudent to expose participants to positive exemplars in both narrative and non-narrative journalistic articles.
Lastly, celebrity involvement was predicted to play a moderating role in the
relationship between exposure to positive exemplars and people’s motivation to act against climate change. However, celebrity involvement did not moderate the main effect, nor did it increase the effect exposure to celebrity exemplars had on positive emotions. These findings counter research results regarding related issues, for example that celebrity involvement influences wildlife conservation behaviour (Brown, 2010). It is possible that celebrity involvement plays less of a role when it comes to climate change mitigation. For example, exposure to the celebrity with the highest involvement (Thunberg) did not lead to the highest
17 As discussed in the theoretical framework, a non-narrative, “generic” news article was tested in order to draw
scores in behavioural intentions and actual behaviour. Thus, familiarity18 with and perceived
closeness to celebrities do not seem to be decisive in motivating climate change action. Instead, other potential moderators might play a role, for example general perceptions of celebrities. For this reason, personality traits (e.g., how competent or inspiring a celebrity is) were controlled for in the experiment. The analysis provided some indication that personality traits could function as a moderator in this type of study design.19 Future research should
probe this by focussing on personality traits and related concepts as moderators. As all research, this study has some limitations. These are mainly related to the composition of the sample. Firstly, while the questionnaire and stimuli were in English, the majority of participants were not English native speakers. This might have impacted how participants understood the article and survey questions. However, the manipulation was quite strong and unambiguous as the only difference between stimuli was the person (i.e., positive exemplar) featured in the article. It is, therefore, unlikely that participants did not understand the stimulus due to language barriers. Additionally, a randomisation check confirmed that stimuli groups did not differ in their composition of native and non-native speakers.Secondly, the study relies on an online opt-in sample which is not representative of the overall
population. Convenience samples are commonly discredited as lacking external validity. However, studies questioning this assumption have shown that convenience samples often do not significantly differ from representative samples in experimental research (Druckman & Kam, 2011; Mullinix, Leeper, Druckman, & Freese, 2015). Thus, it can be assumed that similar effects would occur for different samples, even though it is acknowledged that the use of a convenience sample as in the current study has its obvious limitations. Thirdly, the findings are bound to the context of this study, specifically the choice of positive exemplars
18 Along with celebrity involvement, celebrity knowledge (i.e., knowledge about the respective celebrity) was
measured in the experiment. Thunberg had the highest score here, too: M = 4.39. SD = 1.51 (cf. Sanders: M = 3.68. SD = 1.69, DiCaprio: M = 4.05, SD = 1.71).
19 The results were not included in the main analysis as they were not the core focus of this study. Furthermore,
and the way the news article stimuli were written. The results cannot be generalised to effects of exposure to positive exemplars in general. Fourthly, other variables not included here could have had an influence on the effect of exposure to positive exemplars on people’s motivation to act against climate change. For example, the basic notion of how important the topic is to people might play a role. In fact, personal issue importance was also tested in this study. However, the findings were excluded from the main analysis as personal issue importance in the current sample was very high which led to a skewed distribution of the variable with little variance (see Appendix D). Lastly, the results in this study are limited in their generalisability to online news articles. Different news formats such as TV or radio might work differently. Future research should test similar material across different media types.
These considerations notwithstanding, this study contributes to previous academic research in several ways. First, it shows that positive exemplars, who are under-researched in regard to climate change, can play an important role in influencing individual behaviour. Secondly, it shows how positive emotions can be triggered by specific exemplars but that a positive article itself does not automatically have the same effect. Thirdly, important research avenues were pointed out, for example in regard to narrative engagement which seems to have the capability to function as a mediator for climate change news, if triggered by a news article to begin with. The research also gives pointers to professionals who report on climate change. Using positive exemplars can be an impactful tool for journalists to elicit action amongst readers. However, not every positive exemplar will lead to positive behaviour. The importance of evoking positive emotions should be considered, and saturation (i.e., reporting extensively on the same exemplar) avoided. As climate change can be said to be one of the defining issues of our times, it is necessary to expand this research area to test the conclusions drawn here and improve academic as well as professional knowledge on climate change news coverage and its effects on pro-environmental behaviour.
Acknowledgements
On the way home from a journalism conference, I had a discussion about the role and usefulness of Greta Thunberg in the global climate change movement with Manas Pratap Singh. A few weeks later at university I pitched the idea for my thesis. My first big thanks to him is for inspiring the topic of this thesis. Besides that, I want to thank him for helping me with the acquisition of participants, for taking the time to proofread and proofread again, and for his constant support throughout the process.
I would also like to thank my thesis supervisor Dr. Andreas R. T. Schuck for his assistance and valuable feedback. Not only did he make this study possible. It is also because of his teaching that I became interested in studying emotions which turned out to be central to my thesis.
Even though they might never know that they are mentioned here, I would like to say a big thank you to every single person that participated in my survey. For taking the time to do the (long) questionnaire without getting any credit for it. Thank you!
Lastly, I want to thank my parents for their unwavering support. They made it possible for me to study this programme and encourage me in every step of my life. Special thanks to my dad who also proofread the thesis.
References
Alter, C., Haynes, S., & Worland, J. (2019, December). Time 2019 person of the year. Greta Thunberg. Time. Retrieved from https://time.com/person-of-the-year-2019-greta-thunberg/
Andersen, S., & Cole, S. (1990). “Do I know you?” The role of significant others in general social perception. Journal of Personality and Social Psychology, 59(3), 384–399. https://doi.org/10.1037/0022-3514.59.3.384
Anderson, A. (2011). Sources, media, and modes of climate change communication: The role of celebrities. Wiley Interdisciplinary Reviews: Climate Change, 2(4), 535–546. https://doi.org/10.1002/wcc.119
Austin, E., Vord, R., Pinkleton, B., & Epstein, E. (2008). Celebrity endorsements and their potential to motivate young voters. Mass Communication and Society, 11(4), 420–436. https://doi.org/10.1080/15205430701866600
Bas, O., & Grabe, M. (2015). Emotion-provoking personalization of news: Informing citizens and closing the knowledge gap? Communication Research, 42(2), 159–185.
https://doi.org/10.1177/0093650213514602
Bergkvist, L., & Zhou, K. (2016). Celebrity endorsements: A literature review and research agenda. International Journal of Advertising, 35(4), 642–663.
https://doi.org/10.1080/02650487.2015.1137537
Bilandzic, H., & Sukalla, F. (2019). The role of fictional film exposure and narrative engagement for personal norms, guilt and intentions to protect the climate.
Environmental Communication, 13(8), 1069–1086.
https://doi.org/10.1080/17524032.2019.1575259
Bocarnea, M., & Brown, W. (2007). Celebrity-persona parasocial interaction scale. In R. A. Reynolds, R. Woods, & J. D. Baker (Eds.), Handbook of research on electronic
Boykoff, M., & Goodman, M. (2009). Conspicuous redemption? Reflections on the promises and perils of the “celebritization” of climate change. Geoforum, 40(3), 395–406. https://doi.org/10.1016/j.geoforum.2008.04.006
Brader, T. (2005). Striking a responsive chord: How political ads motivate and persuade voters by appealing to emotions. American Journal of Political Science, 49(2), 388– 405. https://doi.org/10.1111/j.0092-5853.2005.00130.x
Brader, T., & Marcus, G. (2013). Emotion and political psychology. In The Oxford Handbook
of Political Psychology (2nd ed.). Oxford, England: Oxford University Press.
https://doi.org/10.1093/oxfordhb/9780199760107.013.0006
Brown, W. (2010). Steve Irwin’s influence on wildlife conservation. Journal of
Communication, 60(1), 73–93. https://doi.org/10.1111/j.1460-2466.2009.01458.x
Brown, W., & Basil, M. (1995). Media celebrities and public health: Responses to “Magic” Johnson’s HIV disclosure and its impact on AIDS risk and high-risk behaviors. Health
Communication, 7(4), 345–370. https://doi.org/10.1207/s15327027hc0704_4
Brown, W., Basil, M., & Bocarnea, M. (2003). The influence of famous athletes on health beliefs and practices: Mark McGwire, child abuse prevention, and
androstenedione. Journal of Health Communication, 8(1), 41–57. https://doi.org/10.1080/10810730305733
Brown, W., & Bocarnea, M. (2007). Celebrity-persona identification scale. In R. A. Reynolds, R. Woods, & J.D. Baker (Eds.), Handbook of research on electronic surveys and
measurements (pp. 302–305). Hershey, PA: Idea Group Reference
Brown, W., & de Matviuk, M. (2010). Sports celebrities and public health: Diego Maradona’s influence on drug use prevention. Journal of Health Communication, 15(4), 358–373. https://doi.org/10.1080/10810730903460575
Busselle, R., & Bilandzic, H. (2008). Fictionality and perceived realism in experiencing stories: A model of narrative comprehension and engagement. Communication
Theory, 18(2), 255–280. https://doi.org/10.1111/j.1468-2885.2008.00322.x
Busselle, R., & Bilandzic, H. (2009). Measuring narrative engagement. Media
Psychology, 12(4), 321–347. https://doi.org/10.1080/15213260903287259
Busselle, R., & Shrum, L. (2003). Media exposure and exemplar accessibility. Media
Psychology, 5(3), 255–282. https://doi.org/10.1207/S1532785XMEP0503_02
Cacioppo, J., & Petty, R. (1984). The elaboration likelihood model of persuasion. Advances in
Consumer Research, 11, 673–675. Retrieved from
http://search.proquest.com/docview/1293569678/
Chadwick, A. (2015). Toward a theory of persuasive hope: Effects of cognitive appraisals, hope appeals, and hope in the context of climate change. Health Communication,
30(6), 598–611. https://doi.org/10.1080/10410236.2014.916777
Chan, W., Mccrae, R., De Fruyt, F., Jussim, L., Löckenhoff, C., De Bolle, M., … Terracciano, A. (2012). Stereotypes of age differences in personality traits: Universal and accurate?
Journal of Personality and Social Psychology, 103(6), 1050–1066.
https://doi.org/10.1037/a0029712
Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication & Society, 4(3), 245–264. Retrieved from http://search.proquest.com/docview/60421306/
De Vreese, C., & Boomgaarden, H. (2006). News, political knowledge and participation: The differential effects of news media exposure on political knowledge and participation.
Acta Politica, 41(4), 317–341. https://doi.org/10.1057/palgrave.ap.5500164
Druckman, J. N., & Kam, C. D. (2011). Students as Experimental Participants. Cambridge
Handbook of Experimental Political Science, 41–57.
Dunlap, R., Mccright, A., & Yarosh, J. (2016). The political divide on climate change: Partisan polarization widens in the U.S. Environment: Science and Policy for
Sustainable Development, 58(5), 4–23.
https://doi.org/10.1080/00139157.2016.1208995
Entman, R. (2007). Framing bias: Media in the distribution of power. Journal of
Communication, 57(1), 163–173. https://doi.org/10.1111/j.1460-2466.2006.00336.x
Eveland, W., & Scheufele, D. (2000). Connecting news media use with gaps in knowledge and participation. Political Communication, 17(3), 215–237.
https://doi.org/10.1080/105846000414250
Feldman, L., & Hart, P. (2016). Using political efficacy messages to increase climate activism: The mediating role of emotions. Science Communication, 38(1), 99–127. https://doi.org/10.1177/1075547015617941
Feldman, L., & Hart, P. (2018). Is there any hope? How climate change news imagery and text influence audience emotions and support for climate mitigation policies. Risk
Analysis, 38(3), 585–602. https://doi.org/10.1111/risa.12868
Friedman, H., & Friedman, L. (1979). Endorser effectiveness by product type. Journal of
Advertising Research, 19(5), 63–71.
Friedman, L. (2019, August 22). Bernie Sanders’s ‘Green New Deal’: A $16 trillion climate plan. The New York Times. Retrieved from
https://www.nytimes.com/2019/08/22/climate/bernie-sanders-climate-change.html Giles, D. (2002). Parasocial interaction: A review of the literature and a model for future
research. Media Psychology, 4(3), 279–305. https://doi.org/10.1207/S1532785XMEP0403_04
Gonzalez, A., Steele, J., & Baron, A. (2017). Reducing children’s implicit racial bias through exposure to positive out-group exemplars. Child Development, 88(1), 123–130. https://doi.org/10.1111/cdev.12582
Gross, K. (2008). Framing persuasive appeals: Episodic and thematic framing, emotional response, and policy opinion. Political Psychology, 29(2), 169–192.
https://doi.org/10.1111/j.1467-9221.2008.00622.x
Harcup, T. & O’Neill, D. (2017). What is news? News values revisited (again). Journalism
Studies, 18(12), 1470–1488, https://doi.org/10.1080/1461670X.2016.1150193
Hart, P. (2011). One or many? The influence of episodic and thematic climate change frames on policy preferences and individual behavior change. Science Communication, 33(1), 28–51. https://doi.org/10.1177/1075547010366400
Hewstone, M., Johnston, L., & Aird, P. (1992). Cognitive models of stereotype change: (2) Perceptions of homogeneous and heterogeneous groups. European Journal of Social
Psychology, 22(3), 235–249. https://doi.org/10.1002/ejsp.2420220304
Hintzman, D. (1986). "Schema abstraction" in a multiple-trace memory model. Psychological
Review, 93(4), 411–428. https://doi.org/10.1037/0033-295x.93.4.411
Hong, H. (2013). The effects of human interest framing in television news coverage of medical advances. Health Communication, 28(5), 452–460.
https://doi.org/10.1080/10410236.2012.693013
Horton, D., & Wohl, R. (1956). Mass communication and para-social interaction. Psychiatry,
19(3), 215–229. https://doi.org/10.1080/00332747.1956.11023049
Höijer, B. (2010). Emotional anchoring and objectification in the media reporting on climate change. Public Understanding of Science, 19(6), 717–731.
https://doi.org/10.1177/0963662509348863
Iyengar, S. (1991). Is anyone responsible? How television frames political issues. Chicago, IL: University of Chicago Press.
Jackson, D., & Darrow, T. (2005). The influence of celebrity endorsements on young adults’ political opinions. Harvard International Journal of Press/Politics, 10(3), 80–98. https://doi.org/10.1177/1081180X05279278
