Abstract
Prevalent in mass media worldwide, climate change imagery appears to be similar across countries. Replicating a study from the United States, United Kingdom, and Australia, we analyze whether these images are perceived in similar ways cross-nationally by studying Germany, Switzerland, and Austria. A total of 75 respondents sorted images with respect to their perceptions of salience and self-efficacy (Q method). They associated images of climate change impacts most strongly with salience, while they related imagery of renewable energies and mobility to self-efficacy. These findings suggest that perceptions of climate change visuals are largely consistent cross-culturally. They indicate that imagery that is frequently used in media is rarely associated with feelings of salience or self-efficacy.
Introduction: The Significance of Climate Change Imagery
Inside the conference center, the 16th Conference of the Parties (COP) to the United Nations Framework Convention on Climate Change was well under way; outside, a number of Greenpeace activists walked down to Cancun Beach. They had alerted journalists to come along, and a crowd gathered as they started to inflate a giant life ring with the nongovernmental organization’s logo. The activists laid down to form the word “Hope” with their bodies, and cameras started clicking and filming, producing images that quickly appeared in media around the world, for example, in the British Telegraph and Daily Mail and in the NBC and BBC news.
When news media provide information and orientation about the transnational issue of climate change, they use visual images frequently and prominently (O’Neill & Smith, 2014). That some of these pictures, such as Greenpeace’s “Hope” symbol during COP 16, are shared widely in international media is in line with studies showing that visual representations of climate change are similar in many countries. They often emphasize the threat of climate change impacts, focus on nature themes, visualize climate models, cover people (especially politicians), and visualize related themes such as energy issues (Grittmann, 2013; O’Neill, 2013; O’Neill & Smith, 2014; Schneider, 2012; Schneider & Nocke, 2014).
Little is known about the audience’s perception of these images, however. Preliminary results indicate that they may also be similar across countries. O’Neill, Boykoff, Niemeyer, and Day (2013) have shown that perceptions of climate change imagery in the United States, United Kingdom, and Australia are quite consistent. Similar images increased audience members’ perception of the importance (salience) of climate change in all three countries. Likewise, a different set of images—also similar across countries—promoted people’s sense that they could do something about climate change (self-efficacy).
While there has been research on the domestication of media coverage on global issues in different national contexts, it has rarely been investigated whether perceptions of images differ cross-culturally in visual communication research. Therefore, it is unclear whether similar perceptions are limited to Anglophone countries—which are relatively homogenous when it comes to attitudes toward climate change and its media representations (cf. Schmidt, Ivanova, & Schäfer, 2013)—or whether these similarities extend to other regions of the world. The three largest German-speaking countries—Germany, Switzerland, and Austria—provide an excellent case for comparison in this respect, as they differ from the Anglo-American countries in several aspects: They are affected differently by climate change impacts (DARA and the Climate Vulnerable Forum, 2012), have less coverage of global warming in the media (Peters & Heinrichs, 2008; Schmidt et al., 2013), and have lower levels of climate skepticism among their populations (Engels, Hüther, Schäfer, & Held, 2013) and in the media (Kaiser & Rhomberg, 2015; Schäfer, forthcoming). The question thus arises whether or not images of climate change are perceived differently in these countries.
In order to investigate whether there are cultural differences in how climate change imagery is perceived in different contexts, this article replicates O’Neill et al.’s (2013) study investigating perceptions of climate change images in Germany, Switzerland, and Austria. Using Q sorting research, it focuses on the kinds of visual representations of climate change that evoke perceptions of salience (“The image makes me feel that climate change is an important issue”) and self-efficacy (“The image makes me feel that I can do something about climate change”).
Studies on Climate Change Imagery: What Do We Know So Far?
A growing body of scholarly literature has analyzed prevalent climate change depictions in news media coverage, mainly taken from newspapers and magazines as well as television (for overviews, see Manzo, 2010a; O’Neill & Smith, 2014).
Visual Representations of Climate Change
Most studies employ content analysis to describe which images are used by the media to illustrate climate change. They focus on traditional news coverage such as the coverage of climate change in quality and tabloid newspapers and on television. Other mass media like radio, online content, or fictional and entertainment formats are rarely investigated (see Table 1 for an overview). Existing studies have established that five visual themes account for most of the climate change imagery used in news coverage:
Overview of Studies on Climate Change Imagery.
Note. IPCC = Intergovernmental Panel on Climate Change; SPM = summary for policy makers.
Images of climate change impacts and threats: One of the main themes in the visual discourse is climate change impacts, particularly disasters and risks (Rebich-Hespanha et al., 2015). Television news outlets in six countries—the United Kingdom, the United States, Australia, South Africa, India, and Singapore—focus on images that depict the environment and people as being threatened by climate change impacts (Lester & Cottle, 2009). The images used—rising sea levels and destructive weather events like floods—are spectacular and resonate in different cultural contexts (Lester & Cottle, 2009). In Swedish newspapers and on television, visual representations of climate change are often threatening as well, showing floods and human suffering from heat waves (Hoijer, 2010). Impactful imagery is one of the three main themes in the British press (Smith & Joffe, 2009) and among the five most common themes in climate change coverage in Swiss, Austrian, and German newspapers (Rüegg, 2015; see also Grittmann, 2013, 2014). Vulnerability and danger also appear as common themes in imagery for climate change action campaigns (Manzo, 2010b). Doyle (2007) shows that Greenpeace uses pictures to emphasize the impacts of climate change.
Nature themes: Nature is another major theme in climate change imagery, particularly pristine wilderness, flora, and fauna (Rebich-Hespanha et al., 2015). Manzo (2010a) identifies the polar bear as a particular icon of climate change, used in many commercial and political campaigns. For example, Greenpeace strategically used the image of the polar bear in Canada to represent climate change (Slocum, 2004). Legacy media, for example, in the U.S. news media, have used polar bears and vulnerable landscapes as prevalent visual representations of climate change (Rebich-Hespanha et al., 2015).
People/“talking heads”: Personifications of climate change in the form of “talking heads” is among the most common climate change visualizations in the U.S., British, and Australian press (O’Neill et al., 2013; cf. Smith & Joffe, 2009). This personification is particularly true for politicians (Rebich-Hespanha et al., 2015), especially during events such as the United Nations Climate Change Conferences (COPs). Eide (2012) analyzes visual imagery of the Copenhagen climate summit in 2009, finding that most images focused on people—not only on political leaders but also on victims of climate change and protesters. Another study of Canada’s main national newspapers’ coverage of climate change in 2008 notes the dominance of human imagery (DiFrancesco & Young, 2011). With regard to the German-speaking regions studied in this article, a content analysis of print and online coverage finds that images of people, particularly politicians, are the media’s most prominent visualizations in all three countries (Rüegg, 2015).
Graphs and models: Other studies analyze the graphic visualization of climate change data (models). Schneider (2009, 2012) shows that the hockeystick graph and the blue planet turning red are distinctive ways of visualizing climate change. Representations of climate change via graphs and tables are prominent in the British press (Smith & Joffe, 2009). Similarly, Walsh (2010) points out the significance of computer simulations of climate models for climate change imagery, and German and Swiss newspapers use infographics in their climate change coverage (Rüegg, 2015).
Carbon emissions/energy issues: A less dominant but important theme involves images related to carbon emissions (e.g., coal power plants) depicting the causes of climate change (Grittmann, 2013, 2014). Visualizations of energy efficiency, alternative energies, and energy prices also serve as visual frames in climate change communication in the United States (Rebich-Hespanha et al., 2015) and in campaign communications (Doyle, 2007).
In sum, these studies show that visual representations of climate change focus on people and nature being threatened by the impacts of climate change (cf. O’Neill & Nicholson-Cole, 2009). This focus is reflected in the other dominant visual themes of people and nature. Graphs and models are often used as well as images of carbon emissions and energy issues.
The literature also suggests that the use of imagery in climate change coverage is similar across different countries—from Anglo-American countries (United States, United Kingdom, Australia, Canada) to European (Sweden, Germany, Austria, Switzerland), African (South Africa), and Asian countries (Singapore, India)—as well as across different media such as newspapers or television. The media seem to create a mostly homogenous visual discourse of climate change as a transnational issue. This homogeneity is understandable when considering that media coverage on climate change in most countries is driven by the same international events (Schäfer, Ivanova, & Schmidt, 2014), that media organizations in different countries are often owned by the same corporations (e.g., Rupert Murdoch in the United Kingdom and United States, Bertelsmann AG in Europe), and that journalists often use the same agencies and databases to retrieve images for their coverage, particularly for climate change (Grittmann, 2012). As the slow and complex process of climate change is difficult to depict, journalists often rely on visual icons offered by international agencies like Getty Images (Hansen & Machin, 2008).
Perceptions of Climate Change Imagery
Compared to studies analyzing media visualizations of climate change, far fewer studies have scrutinized how audiences perceive these visualizations. Drawing on theories of media effects, research in the field of visual communication, particularly research on visual framing effects, has shown that images and the way they present specific content can influence individual perceptions of an issue (Geise & Baden, 2015; Gibson & Zillmann, 2000; Powell, Boomgaarden, de Swert, & de Vreese, 2015). It is notable, however, that such perceptions are socially and culturally embedded, that sociocultural contexts influence how audiences perceive media presentations, and that this may lead to differences in the perceptions of similar media representations (see the seminal study of Liebes & Katz, 1993; see also Clausen, 2004). Therefore, we study whether images of climate change, which are similar across different sociocultural contexts, are perceived similarly in these differing contexts.
Research on the perceptions of climate change imagery has established that, generally, visual themes dominating media coverage also resonate in people’s perceptions. In one of the first analyses, Nicholson-Cole (2005) focused on how U.K. citizens conceptualize climate change visually, demonstrating that individuals found that the easiest way to conceptualize the abstract issue of climate change was to visualize present and future impacts and the relationship to their personal lives (Nicholson-Cole, 2005). Similarly, Leiserowitz (2006) used a national U.S. survey to reconstruct the affective images people associate with climate change, with the most frequently mentioned images being melting ice and heat (Leiserowitz, 2006).
O’Neill and Hulme (2009) and O’Neill and Nicholson-Cole (2009) studied climate change images with respect to public engagement in the United Kingdom. They demonstrated that people were most drawn to personally relatable icons, such as depictions from their local environment (O’Neill & Hulme, 2009), while dramatic and frightening images were disengaging, making them feel unable to act (O’Neill & Nicholson-Cole, 2009).
In the most comprehensive study so far, O’Neill et al. (2013) analyzed the perceptions of climate change imagery in the United States, the United Kingdom, and Australia. They asked people to sort 40 climate change images with respect to whether the image made them “feel that climate change is important” (p. 415) and whether it made them feel that they “can do something about climate change” (p. 416). The selected pictures were regarded as a relevant set of statements reflecting the larger concourse of visual communication of climate change. O’Neill et al. found shared perceptions of climate change imagery by people in the United Kingdom, United States, and Australia. Again, the visual themes prominent in the media were reflected in participants’ perceptions: Images of climate change impacts were associated with “the sense of importance of the issue of climate change (saliency)” (p. 420), whereas “energy futures imagery promote[d] self-efficacy” (p. 419). At the same time, images of politicians and celebrities (“talking heads”) were not related to self-efficacy or salience perceptions; they could even undermine feelings of self-efficacy. Climate change graphs were not associated with salience or self-efficacy. All of these results were very consistent in the three countries.
These studies suggest that climate change imagery is associated with salience and efficacy perceptions, and that these perceptions are similar across countries. There are only few such studies yet, and they focus almost exclusively on Anglophone Western countries, namely, the United States, the United Kingdom, and Australia (see also Schäfer & Schlichting, 2014). Visual framing research, however, has not yet provided a broader empirical basis on to what extent perceptions of visual representations are similar in different cultural contexts. Therefore, we replicate O’Neill et al.’s (2013) study in a European context and investigate public perceptions of climate change imagery in Germany, Switzerland, and Austria. These countries are excellent cases to study whether perceptions of visual representations differ cross-culturally—not only because climate change imagery has rarely been studied in continental European countries but also because these countries differ from the Anglophone world in crucial dimensions. The elaboration and interpretation of images depend on individuals’ existing knowledge about and attitudes toward the issue presented (Geise & Baden, 2015), and such differences can be found between Anglophone and the three German-speaking countries: Attitudes toward climate change in Germany, Switzerland, and Austria differ compared not only to the United States but also to the United Kingdom and Australia (Metag, Füchslin, & Schäfer, 2015). The level of climate change skepticism in Europe is lower (Engels et al., 2013; European Commission, 2009), the existence of climate change has been accepted more widely, and a relatively broad consensus exists that political measures to fight it are necessary (Peters & Heinrichs, 2008; Schäfer, forthcoming; Weingart, Engels, & Pansegrau, 2000). Moreover, climate change affects countries like Australia more (e.g., through strong heat waves and floods) than European countries (DARA and the Climate Vulnerable Forum, 2012). Images of certain climate change impacts thus might evoke less feelings of climate change being a salient, important issue in regions where such impacts are less likely to manifest themselves. General environmental concern is higher in Germany, Switzerland, and Austria than in the United States and Great Britain (Franzen & Vogl, 2013). In addition, the individual responsibility of citizens to care for the environment is more established in these countries (Hadler & Haller, 2011), which might result in different perceptions of climate change imagery, in particular with regard to self-efficacy.
In light of these differences between German-speaking and Anglophone countries, the question arises as to whether perceptions of climate change imagery are similar or different. We will scrutinize this question, focusing on two crucial dimensions of image perceptions: perceptions of salience and self-efficacy. Both of these are important as media research on climate change has often investigated the link between media coverage and public engagement with climate change (e.g., Whitmarsh, O’Neill, & Lorenzoni, 2011), and public engagement can be achieved if the public perceives climate change as an important issue, that is, as salient, and if people also feel enabled to do something about it (self-efficacy; O’Neill & Nicholson-Cole, 2009). Since salience and self-efficacy are relevant preconditions of public engagement, we study whether the conventional perceptions of climate change images with regard to salience and self-efficacy also hold in German-speaking regions or if other, unanticipated perceptions of which images promote salience and which images promote self-efficacy emerge. Thus, the study contributes to theoretical and empirical research of cross-cultural perceptions of images by posing the following research question:
We follow O’Neill et al.’s (2013) approach and use the Q method to analyze which images people perceive as conveying the importance of climate change (salience) and which images convey a sense that they can do something about climate change (self-efficacy). We then compare the perceptions of climate change imagery in our German-speaking sample to the shared perceptions found in Anglo-American countries by O’Neill et al. (2013).
Methodology
As perceptions of images are individually constructed and often difficult to verbalize, a subjective approach to studying them must be taken (Nicholson-Cole, 2005). Therefore, the aforementioned studies often used qualitative methods, such as focus groups or semistructured interviews (Nicholson-Cole, 2005; O’Neill & Hulme, 2009). This study uses Q sorting, a methodology that is particularly helpful in the study of subjective perceptions and hidden meanings (McKeown & Thomas, 2013; Stephenson, 1953). Without depending on verbalization, Q methodology allows researchers to grasp associations that participants generate from images (Lobinger & Brantner, 2015).
Q methodology combines qualitative and quantitative procedures (Davis & Michelle, 2011). First, participants sort a number of predefined statements (in this case, images relating climate change) on a quasi-normal distribution grid (see Figure 1). Then, a statistical Q factor analysis, which is a factoring technique grouping participants instead of variables, is used to analyze the different sorts (McKeown & Thomas, 2013). By sorting different statements or images, the method makes it possible to transfer qualitative evaluations to a quantifiable scale (Geise & Kamps, 2012).
Q sorting grid.
The method has been applied in research on political attitudes and opinions, market research, environmental psychology, and gender research (Müller & Kals, 2004). It has also been used in media and audience studies (see Davis & Michelle, 2011, pp. 551f.; Lobinger & Brantner, 2015; Singer, 1997), including assessments of audience responses to images (e.g., Geise & Kamps, 2012; Lobinger & Brantner, 2015; O’Neill et al., 2013; O’Neill & Nicholson-Cole, 2009), some in the context of climate change (Hobson & Niemeyer, 2013; O’Neill et al., 2013).
In our study, participants sorted images related to climate change with respect to different variables. We supplemented the sorting procedure with a standardized questionnaire before and with open-ended interviews after the Q sorting session, allowing participants to expand on their sorting decisions.
Materials
The first step of a Q study is to develop a comprehensive array of relevant communicative content (McKeown & Thomas, 2013; Orchard, Fullwood, Morris, & Galbraith, 2014). In our case, we put together a set of images related to climate change. Ideally, a Q sample constitutes a “comprehensive but manageable representation of the concourse from which it is taken” (McKeown & Thomas, 2013, p. 23). In more structured samples, themes should be represented proportionally or according to external criteria, such as prevalence in real life. As we replicated the study by O’Neill et al. (2013), we got access to the images used in their study and followed their sample structure, which was based on a comprehensive media content analysis in the three most frequently researched countries—the United States, United Kingdom, and Australia. Given the strong parallels between climate change imagery in different countries revealed in the literature review, this structure should not compromise our sample for the Q sorts.
In addition, we checked whether O’Neill et al.’s (2013) set of images represented the pictures used in German-speaking media by drawing on the small number of existing studies on climate change imagery in the German-speaking world (Grittmann, 2013, 2014; Pleger, 2013; Rüegg, 2015). We also considered the currently unpublished findings of a large research project that coded German newspaper images of climate summit coverage from 2010 to 2013 (Wessler, Lück, & Wozniak, 2014; Wozniak, Lück, & Wessler, 2015). This step verified that the selection of images was largely appropriate, albeit some images had to be adjusted to the different national contexts, for example, by replacing the pictures of political leaders with comparable domestic ones, by replacing the image of the flooding map of an English river with a German one (River Elbe), or by substituting English with German words.
The final set included 40 images related to climate change (see Table 2). 1 It represents the visual discourse on climate change in the given countries and adheres to country-specific information. Images representing the theme of impacts and threats of climate change include the following: landslide, ski slopes with little snow, Inuit, car in snowstorm, low reservoir, cracked/dry earth, fighting bushfires, aerial views of floods, coral atolls, coastal erosion, and a person in the desert. Images conveying nature themes include coral reefs, icebergs, polar bear, glacier, the earth from space, volcano, and felling tropical forests. Doris Leuthardt/Angela Merkel/Werner Fayman, climate scientist, Al Gore, Bono, and Barack Obama represented pictures of people/talking heads. The temperature graph and flooding map (River Elbe) were climate change graphs and models. Images relating to the carbon emissions/energy issues theme included the electric car, meat counters/red meat for sale, protests in front of a coal power plant, tractor on a farm, home insulation, solar panels, wind farms, ecohouse, nuclear power plant, planes at airport, smokestacks, traffic jam, and fuel pump.
Images Used in the Study.
Participants
As Q methodology aims to uncover different viewpoints and does not strive for generalizability, sample size is not that important; some Q studies are even conducted on a single-case (n = 1) basis (McKeown & Thomas, 2013). In our study, however, 75 respondents from three comparable cities participated in the study: 25 in Hamburg (Germany), 25 in Zurich (Switzerland), and 25 in Vienna (Austria). A research design with 75 participants and 40 images is in line with Q study conventions. Participants were recruited in systematically selected, socioeconomically different neighborhoods (the selection was similar in all three cities). To recruit, we used supermarket postings and online platforms, and we approached people on the street. Each participant received a financial incentive for participation.
We aimed to include participants with different backgrounds by recruiting different age groups, sexes, and educational backgrounds. In total, 38 participants were female and 37 male; 28 were 18 to 29 years old, 25 were 30 to 49 years old, and 22 were over 49 years old. Of these, 21 had an academic degree, 28 a high school degree, and 26 a lower school degree. The questionnaire administered before each sorting session ensured that participants also varied in their views on climate change (“To what extent would you say you are interested in climate change?” 1 = not at all, 4 = very much; M = 3.08, SD = 0.67; “How concerned are you about climate change?” 1 = not at all concerned, 4 = very concerned; M = 3.08, SD = 0.77).
Procedure
Sorting sessions took place in Hamburg, Zurich, and Vienna in February 2014. Guided by a research assistant, each session took about 45 minutes. Participants were first asked to fill in the paper-and-pencil questionnaire and to read the sorting instructions. 2 The 40 images were printed on sorting cards and distributed to the participants stacked randomly. The sorting grid comprised a continuum from −4 (most disagree) to +4 (most agree). The distribution of images was forced, allowing more images to be placed in the middle categories than in the outer, extreme categories. For example, only two pictures could be ranked in the “most disagree” and two pictures in the “most agree” categories.
Participants undertook two Q sorts. First, they sorted the images based on the statement “This image makes me feel that climate change is important.” The second sorting was based on the statement “This image makes me feel I can do something about climate change.” The pictures were removed from the grid and mixed again between sorts. Participants were instructed to sort intuitively; there was no “right” sorting, and their subjective views were our main interest. After each sort, participants were interviewed about their sorting. In the interviews, particular attention was paid to pictures at the extremes of the distribution and to images about which participants felt strongly. The interviews were recorded, and photos of the sorts were taken as backup.
Analysis
The analysis of Q sorts relies on an inverted factor analysis (Q factoring), grouping participants instead of statements (McKeown & Thomas, 2013). Q sorts were analyzed using Advance Q 2.08 (Niemeyer, 2013), a program that calculates intercorrelations between Q sorts to group similar participants and allow for different adjustments during the analysis. 3 A principal component analysis was conducted with varimax rotation. The approach included all 75 participants together as we aimed to uncover shared perceptions of climate change imagery among the participants.
We conducted the analysis stepwise and aimed for a factor solution that accounted for the greatest number of uniquely defined responses by the smallest number of factors (Michelle, Davis, & Vladica, 2012), excluding “confounded” and statistically insignificant sorts (Michelle et al., 2012, p. 124). First, only factors with an eigenvalue over 1 were considered, and the level of confidence was .99. We proceeded with the analysis by reducing the maximum number of factors each time until achieving the final factor solution. This final factor solution was achieved when the number of factors still accounted for as much variance as possible but the factors were still distinct and highly interpretable. Compared to O’Neill et al. (2013), we not only interpreted the first factor that emerged and explained the largest part of variance representing the mainstream view (O’Neill et al., 2013, p. 416) but also considered the other emerging factors. The Q method is especially useful for observing perspectives deviating from the mainstream view. 4
Results
Salience
The first analytical dimension focuses on the extent to which different images promote the perception that climate change is an important issue. For this dimension, the analysis produced a strong first factor explaining most of the variance and a comparatively weak second factor. Table 3 displays the results for the salience sorts for all 75 participants. Factor scores are reported with the z scores for each factor converted back into an array of typical responses using the same range of responses presented to the participants (i.e., −4 to +4). For clarity, only the images ranked on the +4, +3, −3, and −4 positions for each factor are displayed in Tables 3 and 4.
Results for Salience Sorts.
Note. Factor scores are converted from z scores. Percentages are the explained variance for each factor. “Stmt” indicates the picture number.
Results for Self-Efficacy Sorts.
Note. Factor scores are converted from z scores. Percentages are the explained variance for each factor. “Stmt” indicates the picture number.
Climate Change Impacts
The climate change impacts factor explains most of the variance and includes often-used visual representations in the media, such as floods, people in the desert, landslides, polar bears, and drought. The pictures show objects affected by climate change or the impacts themselves. Flooding is often perceived as a strong and imminent threat, perhaps due to the 2013 floods of the River Danube in Austria and Germany (“This is what affects us ourselves, in Germany, these floods”; P27, f, >49, acad., GER). 5 Flooding is also strongly related to climate change (“With the flooding, for example, it is clear: Climate change!”; P29, m, 30-49, low, GER). Flooding is a strong image that appears often on television (Lester & Cottle, 2009), and the polar bear is one of the most typical pictures associated with global warming (cf. Manzo 2010a): “The polar is simply the image of climate change. You always think first of the polar bear which does not have any ice to walk on anymore” (P11, f, 18-29, high, CH). Pictures associated with drought and other climate change impacts (landslides) are salient because they depict threats to humans: “These pictures just evoked the strongest reaction because there is a person in them who is affected by a drought” (P9, m, >49, acad., CH) or “This is what can evoke danger. Like a catastrophe which might happen. It is also something that can make you scared” (P23, m, 30-49, acad., CH). Participants also made clear that they recognized the images from the mass media. Participant 53 (m, 18-29, acad., AUS) summarizes, “Pictures which remind me most of climate change. These are exactly those pictures which I relate to climate change and which are used in media.” In contrast, the church congregation, deckchairs in the sun, and pictures of political leaders do not seem to convey the importance of climate change as an issue. As a German participant pointed out, “The church just does not have anything to do with climate change in my view” (P27, f, >49, acad., GER).
Respondents do not seem to trust that politicians will do something to fight climate change. An Austrian participant made this point clear when explaining why he ranked the picture of Werner Faymann, the Austrian chancellor, as the least salient: “Politicians come off worst in this respect. Because their statements are just not trustworthy” (P62, m, >49, low, AUS). The images of the ecohouse and home insulation do not seem to be related to perceptions of salience as some participants did not recognize how the pictures related to climate change (ecohouse: “[. . .] this is a nice house. But I cannot identify whether it is built in a climate-friendly way” [P35, f, 18-29, low, GER]). As for home insulation, one participant stated, “I had to think hard about what this actually shows until one gets that the guy is insulating his roof. First I thought, there are bales of straw or something like that on the picture” (P36, m, 30-49, high, GER).
Causes of Climate Change
The second factor includes causes for environmental problems with images of smokestacks, deforestation, protests in front of a coal power station, and traffic jams. A respondent pointed out that these images are “negative images not only of the consequences but of the causes, related to the cars [. . .] and the emissions. They have a strong impact on me” (P12, f, 30-49, low, CH). Respondents noted that there might be a connection between the causes, such as carbon emissions and consequences: “I think that climate change mostly has to do with trees dying and that there is polluted air. And these are the most important aspects of global warming in my view” (P33, f, 30-49, high, GER). As the images of smokestacks and traffic jam reveal, carbon emissions are perceived as an environmental problem and as a cause for climate change. Participant 62 illustrates this position well: “The carbon emissions, air pollution, industrial areas, factories—this was very dominant in my view” (m, >49, low, AUS). Pictures of protests in front of a coal power station and traffic jams illustrate humanity’s immense energy needs and the related carbon emissions, while pictures of deforestation illustrate how the problem of carbon emissions is continuously intensified by commercial interests.
Similar to the first factor, the church congregation, deckchairs, and home insulation are perceived as least conveying the importance of climate change. Regarding the image of the volcano and the car in the snow, many participants did not relate a volcanic eruption or snowfall to climate change: “I couldn’t make sense of the volcano with respect to climate change. I didn’t see a connection there; whether volcanoes are erupting more often now due to global warming” (P26, m, 18-29, acad., GER). Snowfall was not perceived as having anything to do with climate change: “Snow does not really stand for climate change; snow has existed forever” (P43, f, 30-49, low, GER).
In sum, images of climate change impacts as well as causes of global warming are related to perceptions of salience. Among this imagery, visual representations of who is affected and threatened by global warming play an important role (animals, biodiversity, and humankind). Flooding or a person in the desert raise awareness in people that climate change is an important issue because they feel that someone or something will be negatively affected.
Self-Efficacy
The second analytical dimension focused on the extent to which different images promote the perception that audience members themselves could do something about climate change. Table 4 displays the results for this self-efficacy dimension. Again, we find one strong first factor and a second, weaker factor.
Renewable Energies
This factor depicts solutions to mitigate climate change using clean energies, thereby replacing fossil fuels and protecting the climate. The wind park, solar panels, and electric car represent renewables, while the traffic jam illustrates the need for alternative energies. It seems clear to participants that people can do something about climate change if they use these renewable energies: “These are things which you need to be responsible for yourself. [. . .] That you pay attention to how much energy you are using. That you rely on solar and wind energy more” (P19, f, 18-29, high, CH).
Images showing individual or consumer opportunities to reduce carbon emissions also promote self-efficacy. By abstaining from eating meat, for example, people can fight global warming: “This is exactly where I have the easiest choice. If I want less carbon emission I have to think about: Do I want to eat a lot of meat, do I want to pay attention to where the meat comes from or not? (P27, f, >49, acad., GER)”.
Individual responsibility is reflected in images demonstrating a choice to participants in terms of what they can abstain from or change in their daily lives: “Do I buy certain products, am I driving a car, what does my consumption look like, these are things that I can actually personally influence” (P35, f, 18-29, low, GER). As another participant put it, “I use my car less. This affects me because with my behavior I can contribute to not producing so many emissions [. . .]” (P57, m, 30-49, acad., AUS).
In many cases, images ranking low in self-efficacy depict climate change impacts or natural disasters. The landslide, flooding, and iceberg images undermine feelings of self-efficacy. Also, the image of the polar bear does not seem to convey to the respondents that they can do something about climate change: “Yes, this is something where I think: This shows the impacts of climate change but you can’t do anything about it” (P53, m, 18-28, acad., AUS). In addition, imagery that relates to scientific research and statistics, such as temperature graphs, does not seem to be associated with self-efficacy.
Mobility
The mobility factor tackles the problem of traffic and the daily need for gasoline and energy, as exemplified by images of the electric car, traffic jams, fuel pump, and planes at the airport. Respondents perceive the electric car as a means of self-efficacy and a potential solution to the problem of enhanced mobility in modern societies: “I drive a car myself and I think one can consider to stop using conventional petrol and how I could use a car with alternative energies” (P47, m, 18-29, acad., GER).
The factor includes holiday travel as well as routine mobility. Holiday travels relate to further and longer journeys, while routine mobility refers to everyday mobility. The planes at the airport convey that in terms of vacations or longer journeys, people can do something about climate change: “I thought about what I can do myself. What affects myself. And holidays and using the plane, I can affect that myself” (P23, m, 30-49, acad., CH). Relating more to everyday actions, routine mobility is depicted by images of traffic, cars needing gas, and various forms of mobility, such as the electric car. One respondent said this about the traffic jam image: “This shows me that I should use the car less” (P15, m, 18-29, acad., CH). Solar panels also play a role in the mobility factor by presenting solutions to the energy crisis: “I would support alternative energies such as the windmill or solar power or green electricity” (P59, f, 30-49, high, AUS).
In this factor, the volcano eruption is a natural disaster that related least to feelings of self-efficacy. Identifiable “talking heads,” such as political leaders, were often ranked as less efficacious, for example, Barack Obama, Al Gore, or U2’s Bono. The church congregation also was not perceived as promoting the idea that people can do something about climate change. The connection between the church and climate change was unclear to many participants: “I put the image of the church there because I do not have a personal relation to it, and the picture doesn’t give me the feeling that I should do something” (P31, f, 18-29, high, GER).
Overall, perceptions of self-efficacy are reflected in images of renewable energies, carbon emissions, mobility, and traveling in the context of climate change. This kind of imagery relates to people’s lifestyles and their personal choices in everyday life.
Discussion and Conclusion
Imagery is an important and widely used facet of media reporting on climate change, and certain visual representations seem to be associated with particular audience perceptions (O’Neill et al., 2013). To date, however, studies have mostly analyzed the Anglo-American world (Schäfer & Schlichting, 2014). Going beyond this limitation, our study contributes to visual communication research by analyzing whether these perceptions are cross-culturally similar. It analyzed perceptions of climate change imagery in the major German-speaking countries: Germany, Switzerland, and Austria. Based on Q sorting methodology, we found that certain images are related to respondents’ feeling that climate change is important (salience), and that other images are associated with respondents’ perceptions that they can do something about climate change (self-efficacy).
First, visual representations of climate change impacts are perceived as conveying the importance of global warming, and these images are very often used in the Anglophone (e.g., O’Neill et al., 2013) and German-language media (Rüegg, 2015). Aerial views of floods and the desert seem to be particularly impressive, attracting people’s attention and creating fear.
At the same time, images of climate change impacts may undermine perceptions of self-efficacy. They make climate change appear as an overwhelming, forceful natural development, and participants think that they, as individuals, are helpless and cannot do anything to stop it.
The set of images associated with self-efficacy is notably distinct from images related to salience, and it contains pictures that are used less often in the media. It includes images of clean and renewable energies, forms of mobility, and lifestyle and consumption choices; in essence, the images depict ways to reduce carbon emissions.
Imageries of identifiable people and politicians are not related to perceptions of salience or self-efficacy. Politicians are not perceived as conveying the idea that climate change is an important issue, nor do they convey that individuals can do anything about climate change.
These results need to be further substantiated and refined, for example, by distinguishing different subgroups based on gender, age, education, or attitudes about climate change or by extending the analysis to respondents from rural areas. Nonetheless, a number of conclusions can already be drawn. One the one hand, we find some, albeit small, differences between our results and those of O’Neill et al.’s (2013) for Australia, the United States, and the United Kingdom. Comparing the two studies, the picture of the coral reef that ranked high for salience in O’Neill et al.’s (2013) study does not appear to be associated with salience perceptions in our study—maybe because of the greater distance between German-speaking countries and coral reefs. By contrast, felling forests is more strongly related to climate change being an important issue than it seemed to be for respondents in the United States, the United Kingdom, and Australia. Pictures of home insulation or protest in front of a coal power plant relate differently to self-efficacy. They were important in O’Neill et al.’s study, but they do not play much of a role for feelings of self-efficacy in our study. Some respondents did not associate the home insulation image with energy efficiency.
This indicates that perceptions of climate change imagery are domesticized to some extent. Overall, however, our results are remarkably similar to O’Neill et al. (2013). Although people in the analyzed German-speaking countries differ in their attitudes toward climate change from people in the United States, the United Kingdom, or Australia, their perceptions of climate change imagery seem to be rather comparable. This suggests that climate change imagery, and potentially imagery of other transnational issues as well, is perceived in similar ways cross-culturally, transcending geographical and linguistic boundaries.
These results are in line with the broader research into the visual communication of climate change. That images of strong climate change impacts such as floods are not linked to perceptions of self-efficacy, for example, mirrors the finding that fear-inducing pictures were ineffective in evoking Britons’ engagement with climate change (O’Neill & Nicholson-Cole, 2009). Also similar to Germany, Switzerland, and Austria, images of identifiable people, particularly politicians, were not related to perceptions of salience and barely related to self-efficacy perceptions in the United States, United Kingdom, and Australia (O’Neill et al., 2013). Pictures of politicians being only slightly related to perceptions of salience or self-efficacy is an interesting finding because such images are among the most common in the media (Eide, 2012; O’Neill, 2013; Smith & Joffe, 2009). It seems that the most common pictures used in the media either entirely fail to evoke the feeling that climate change is important or fail to make people believe they can do something about it, which is the case for “talking heads.” Alternatively, they are associated only with high salience perceptions but not with perceptions of self-efficacy—as is the case for impact imagery. Imagery highlighting self-efficacy appears less often in the media.
Normatively, these findings raise interesting points. The imagery predominantly used by the media does not seem to be suitable to raise perceptions of salience and self-efficacy. Even if some of the frequently used pictures—like those of climate change impacts—are associated with climate change being an important issue, images conveying the feeling of self-efficacy are barely used in the media. While mass media imagery of climate change is able to draw attention to the issue, it does not provide options for action, at least not through visual communication. While this is understandable given the media’s attempts to maximize audience appeal and attract their attention, one may raise the question of the normative function of media coverage. If one advocates the position that journalists are supposed to engage people in taking action against climate change—which would be in line with recent demand for more “constructive news” (Haagerup, 2015)—then current climate change imagery is ill-suited. However, one could argue that by using images that run against common public perceptions of climate change, journalists can challenge these mainstream perceptions. Offering new perspectives can also be regarded as journalists’ responsibility. In contrast to journalists, climate change campaigners aim to raise awareness of the issue, and thus our results can be used to assess whether their campaigns use the most effective images to raise awareness about global warming.
The results may also be valuable in reflecting upon the imagery used in science communication; for example, they might bring into question the effectiveness of the graphs and models used when covering the Intergovernmental Panel on Climate Change report or other scientific studies on climate change (Mahony & Hulme, 2014; Schneider, 2012). Given the fact that most people—at least in German-speaking countries—no longer question the problem of climate change itself, a change in the climate change imagery employed by the media and other communicators might contribute to motivating people to become more active in containing it.
In any case, this research demonstrates that it may be useful for future studies to differentiate between different dimensions of perceptions, such as salience and self-efficacy. It would also be fruitful to study the perceptions of media producers like journalists and photographers and compare them with audience perceptions. Since salience and self-efficacy are crucial for public engagement, a further step could be to quantitatively assess, for example, in an experimental design, whether perceptions evoked by these images have an effect on other relevant dimensions for engagement with climate change, such as participation in climate action campaigns.
Footnotes
Acknowledgements
The authors would like to thank Saffron O’Neill for granting access to the images and materials used in their study and Simon Niemeyer for providing his Q software and advice on its use. We would also like to thank the anonymous reviewers for their helpful comments on the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection was partly funded by the German Research Council DFG through the Cluster of Excellence “Integrated Climate System Analysis and Prediction (CliSAP)’’, EXC 177, University of Hamburg.