Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CLIMATE SCIENCE (climatescience.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 20 October 2017

Content Analysis in Climate Change Communication

Summary and Keywords

Content analysis is one of the most frequently used methods in climate change communication research. Studies implementing content analysis investigate how climate change is presented in mass media or other communication content.

Quantitative content analysis develops a standardized codebook to code content systematically, which then allows for statistical analysis. Qualitative analysis relies on interpretative methods and a closer reading of the material, often using hermeneutic approaches and taking linguistic features of the text more into account than quantitative analysis. While quantitative analysis—particularly if conducted automatically—can comprise larger samples, qualitative analysis usually entails smaller samples, as it is more detailed. Different types of material—whether online content, campaign material, or climate change imagery—bring about different challenges when studied through content analysis that need to be considered when drawing samples of the material for content analysis. To evaluate the quality of a content analysis measures for reliability and validity are used.

Key themes in content analyses of climate change communication are the media’s attention to climate change and the different points of view on global warming as an issue being present in the media coverage. Challenges for content analysis as a method for assessing climate change communication arise from the lack of comparability of the various studies that exist. Multimodal approaches are developed to better adhere to both textual and visual content simultaneously in content analyses of climate change communication.

Keywords: content analysis, climate change, manual content analysis, automated content analysis, discourse analysis, frame analyses, imagery, online content, reliability

Content analysis is a widely used method for investigating climate change communication and media portrayals. This article gives an introduction of content analysis as an empirical method and of the major research questions assessed through content analysis of climate change communication. It introduces different content analyses that are used in climate change communication research and portrays typical analytical strategies. It provides detailed information on what needs to be considered when specific types of material are studied as well as an overview of key themes and categories in content analyses of climate change communication. Challenges of the method and criteria to evaluate content analyses are discussed. The article will not discuss results of the various content analyses in detail as these results are presented in other articles in the encyclopedia (see for example “Elite Newspaper and Magazine Coverage of Climate Change,” “TV and Cable News Coverage of Climate Change,” and “Blog, Twitter, and Other Social Media Depictions of Climate Change”).

Aims of Content Analysis Methods in Climate Change Communication

Climate change communication and the media coverage of global warming are crucial for people’s awareness and opinions about the issue. Most people do not experience climate change firsthand but have to rely on communicative content to develop an understanding of climate change.

The question thus arises how to assess to what extent and in what ways media outlets and other means of communication cover climate change. Content analysis is a method to describe and analyze communicative content and message characteristics. Traditionally—and still in many cases today—it is implemented for the analysis of textual content, but it can also be applied to other material like images, video, or film (Krippendorff, 2013).

Content analyses of climate change communication deal with questions such as whether the media coverage of climate change differs between countries, how it develops over time, and in what way climate change is presented in textual or visual outlets (e.g., which perspectives on climate change as an issue are dominant and which kind of actors are given a voice). The method can be used to investigate to what extent the coverage of climate change differs between media outlets and other communication materials (e.g., between conservative and liberal media, broadsheet and tabloid newspapers, or legacy and online media).

Content analysis can be carried out in a standardized quantitative way as well as in an interpretative, qualitative manner. This article as well as other meta-analyses of media representations of climate change demonstrate that there is a roughly equal number of studies using quantitative and qualitative analysis (Schäfer & Schlichting, 2014; Wozniak, Lück, & Wessler, 2014).

The research aims outlined previously can be assessed by solely relying on content analysis as a method, which is the main focus of this article.1 In addition, content analysis data can be used in mixed-methods studies, comparing content with survey data of public opinion of climate change (e.g., Feldman, Maibach, Roser-Renouf, & Leiserowitz, 2012). The combination of these methods aims at understanding to what extent the media content on climate change that people use is linked to their views on climate change, such as their acceptance of global warming. Recently, other mixed-method approaches have been realized, such as combining content and network analysis (e.g., Williams, McMurray, Kurz, & Hugo Lambert, 2015). These approaches use content analysis to assess what people post on social media about climate change while taking its network character into account.

Content Analysis Designs in Climate Change Communication Research

One can distinguish between quantitative and qualitative approaches to content analysis. The quantitative approach is rooted in the traditional quantitative newspaper analysis (Krippendorff, 2013), aiming to describe textual content systematically and objectively (Neuendorf, 2002), and to make “replicable and valid inferences from texts (or other meaningful matter) to the context of their use” (Krippendorff, 2013, p. 24). Qualitative approaches are concerned with (re-)constructing the meaning of climate change content and thus rely on “a context-sensitive and in-depth exploration” (Olausson, 2010, p. 141) using interpretative techniques.

Quantitative Content Analysis Designs

Manual Content Analysis with Human Coding

Most content analyses of climate change communication are quantitative analyses conducted manually, that is, carried out by human coders. In such an analysis, researchers follow certain steps (see Figure 1) based on their research questions and hypotheses that have been theoretically derived from the climate change communication literature ex ante.

Content Analysis in Climate Change CommunicationClick to view larger

Figure 1. Flowchart of typical processes of quantitative content analysis

Source: Adapted from Neuendorf (2002, pp. 50–51) and Rössler (2005, p. 35).

First, they need to decide on their sampling strategy (e.g., to select articles randomly) and their units of coding (e.g., a newspaper article dealing with climate change). Based on considerations of which constructs they want to measure (e.g., how global warming is “framed”), the researchers develop a codebook with regard to their research question. If human coding is used, coders—the researchers themselves or others, often student assistants—are then trained in applying the codebook to the texts (e.g., Ahchong & Dodds, 2012). The aim of this training is that all coders use the codebook reliably in the same fashion, that they code the same categories for the same texts and parts of the texts such as single statements. After a pilot coding test, the codebook is re-assessed and refined. Then a reliability test is conducted to assure its reliable measurement (see also section “Evaluative Techniques and Challenges of Content Analysis as a Method”). If the reliability test yields sufficient scores, the coding can start.

A content analysis differentiates between sampling units and coding units (Krippendorff, 2013). The sampling unit comprises the criteria based on which items are included in the analysis (e.g., all articles on climate change in the Washington Post). Coding units are units that are distinguished for coding, such as single articles, statements in a text, images etc.

With regards to identifying sampling units for content analyses, different aspects can be taken into account. One decision to be made is which country is studied. Most of the research on climate change communication currently focuses on North America or Europe (Schäfer & Schlichting, 2014). There are many studies on Anglophone countries such as the United States (e.g., Boykoff & Boykoff, 2004), Canada (e.g., Ahchong & Dodds, 2012; Young & Dugas, 2012), and Australia (e.g., Bacon & Nash, 2012). By now, a content analysis of climate change communication exists for most European countries: France (Painter & Ashe, 2012), Netherlands (Dirikx & Gelders, 2010), Germany (Grundmann, 2007; Schäfer, Ivanova, & Schmidt, 2014), Ireland (Wagner & Payne, 2015), and Sweden (Berglez, Höijer, & Olausson, 2009; Olausson, 2010; Shehata & Hopmann, 2012); see also “Climate Change Communication in Switzerland,” “Climate Change Communication in Norway,” and “Climate Change Communication in Spain.” Because countries in the Global South are under-researched, but differently and often more strongly affected by the impacts of climate change, their media coverage has increasingly been analyzed recently: for example, India (Billett, 2010), Brazil (Painter & Ashe, 2012), countries in southeast Asia (Nash & Bacon, 2006), and other non-industrialized countries (Shanahan, 2009). Some books provide a global perspective on climate change coverage through collecting accounts of studies on climate change communication in numerous countries and on different continents (Boyce & Lewis, 2009; Eide & Kunelius, 2012; Eide, Kunelius, & Kumpu, 2010).

The other category to be decided when conducting a content analysis is the time period of analysis. In the case of manual content analysis, the amount of material usually needs to be limited to be able to be coded with the naturally limited resources of a research project and a certain number of coders. One way to restrict the time period under investigation is to focus on events like the publication of the Intergovernmental Panel on Climate Change (IPCC)’s Assessment Reports (e.g., O’Neill, Williams, Kurz, Wiersma, & Boykoff, 2015) or the Conference of the Parties (COP) conferences (e.g., Eide et al., 2010).

Before the development of the codebook and the beginning of the coding, the material for the content analysis needs to be collected. Most studies retrieve their material from (online) databases. As the major share of content analyses of climate change coverage is based on texts, mainly from print media, the common databases are LexisNexis, ProQuest, and Factiva (Antilla, 2010; Boykoff, 2012; Boykoff & Boykoff, 2004). Texts are retrieved from these databases using search terms like “global warming” or “climate change” (for an overview, see Schmidt, Ivanova, & Schäfer, 2013). In case of more specific research questions (e.g., studying the coverage of the COP conference) search terms matching these research aims are used (e.g., “IPCC” or “greenhouse gas” [e.g., Billett, 2010]).

Most studies focus on print media coverage of climate change. Among those, the majority analyzes elite press coverage (e.g., Ahchong & Dodds, 2012; Antilla, 2010; Bell, 1994; Billett, 2010; Boykoff & Boykoff, 2004), while only few include tabloids (e.g., Boykoff, 2008b). Compared to print media, the number of studies dealing with climate change coverage on television is smaller (e.g., Boykoff, 2007, 2008a; Boykoff, 2012; Krøvel, 2011; Painter, 2011; Ungar, 1999). Their focus is often on news broadcasts, such as BBC World News Tonight, CBS Evening News, or NBC Nightly News. Material from these news shows is retrieved through the Vanderbilt University Television News Archive (offering abstracts of newscasts) (e.g., Boykoff, 2008a; Ungar, 1999) or through Lexis-Nexis, which offers transcripts of some U.S. television channels such as Fox News, CNN, and MSNBC (e.g., Feldman, Maibach, Roser-Renouf, & Leiserowitz, 2012). Which newspapers and news broadcasts researchers select for their analysis is often based on readership and viewing figures that indicate the reach of a media outlet.

While sample size differs from analysis to analysis, the state of research allows for some general statements. Usually, the more specific the research aim and the issue under study, the smaller or more restricted the sample will be (e.g., 60 websites of climate change organizations [Jun, 2011]). There are different sampling strategies on how to draw a sample of articles if coding all news items of the population exceeds the resources of a research project. One strategy is to code only articles from the front page (Asplund, Hjerpe, & Wibeck, 2013). Some studies focus only on specific events, others cover large time periods, for example studying media content on climate change from the 1970s or even earlier up to now (e.g., Liu, Lindquist, & Vedlitz, 2011; Rebich-Hespanha et al., 2014; Ungar, 1999). Studies covering a longer time frame reveal that early media coverage of climate change was predominantly scientific but has recently shifted to be more politicized and now even plays a role in popular culture with celebrities appearing in climate change coverage as advocates or skeptics (Anderson, 2011). Another sampling strategy is random sampling. Researchers draw a random sample of the material to be studied, for example, taking every third article (Shaw, 2013). A very common random stratified sampling strategy is using a constructed week (identifying all Mondays in the period under study and then randomly selecting a number of Mondays, identifying all Tuesdays and selecting a sub-sample randomly, and so on) because this takes the cyclical nature of media content into account (e.g., the science section in a newspaper may appear only on Thursdays). The type of medium analyzed can also influence the sample for the content analysis because specific material such as popular books for a mainstream audience might make purposive sampling necessary (Shaw, 2013).

Studies using quantitative content analysis with manual coding can thus offer a quantifiable assessment of the extent to which certain aspects of climate change are covered in the text under study. Such accounts are replicable and reliable, and the codebook can be used for the analysis of numerous texts. However, they do not provide a deeper and interpretive analysis of specific parts of the texts, and the sample size is restricted.

Automated Content Analysis

Compared to manual content analyses of climate change communication, automated content analysis does not use human coders, proceeds automatically based on software, and can handle larger amounts of material. Therefore, it has been used to analyze online communication, for example, where large data collections can be analyzed (Fløttum, Gjesdal, Gjerstad, Koteyko, & Salway, 2014; Kirilenko & Stepchenkova, 2014). A second advantage of an automated content analysis is that coding procedures conducted by a computer are very reliable (Krippendorff, 2013).

Typical approaches of computational content analysis are text mining and dictionary approaches. Text mining is a method for searching large amounts of texts for specific information (e.g., certain textual attributes). In dictionary approaches, texts are searched for word families based on dictionaries and rules that are fed to the computer software (Krippendorff, 2013; Scharkow, 2013). Figure 1 shows the dictionary approach as one possible way automatic content analysis can proceed. After deciding on the sampling and the coding units the researcher needs to feed the computer a coding scheme or a dictionary and needs to develop a codebook that makes the coding transparent for other researchers. In contrast to manual analysis, the training of coders, pilot coding, and reliability checks are mostly obsolete with automatic coding. Following a dictionary approach, Ivanova (2015) analyzes climate change coverage in 11 countries over 15 years with respect to their mentioning of transnational governance, other countries, and transnational collectives (Ivanova, 2015).

More recently, machine-based learning offers an inductive approach to automated content analysis as it uses an algorithm that is first trained with data of the same kind, which is then automatically analyzed after the training (Scharkow, 2013). That means that this approach first includes manual coding of the documents used for training the algorithm. Afterwards, the trained algorithm codes the entire content automatically.

In climate change communication, a type of analysis that is typically done using automated content analysis is frequency analysis (see also the section on “Frequency Analysis”). Frequency analyses are conducted automatically using search strings (Schmidt, Ivanova, & Schäfer, 2013). Other more complicated ways of automated analysis such as latent dirichlet allocation, which is a statistical model used for estimating the frequency of words in texts, or machine-based learning are able to identify frequent words in texts as well as patterns of similarity by analyzing the way these words are used in the text (Kirilenko, Stepchenkova, Romsdahl, & Mattis, 2012). Frequent words can then be further analyzed through factor analysis to identify topics in the texts (Kirilenko et al., 2012).

Corpus linguistic studies conduct more complex automated analyses (see “Linguistic Analysis Approaches for Assessing Climate Change Communication”). They use different kinds of software to search the Web (e.g., blogs, websites, RSS feeds) and extract certain word combinations (Koteyko, 2010; Koteyko, Thelwall, & Nerlich, 2010) or identify lexical bundles or patterns (Fløttum et al., 2014) in the content they find on climate change. These automated corpus linguistic studies often use keywords and collocations, or sequences of words or terms that have become established, as analytical tools to identify word patterns (Koteyko, Jaspal, & Nerlich, 2013).

Certain types of online data like Twitter data contain useful information such as the geographical location of a tweet (Kirilenko & Stepchenkova, 2014). Automated content analyses of tweets on climate change are thus able to assess not only the contents of the tweets but also their spatial and temporal distribution (Kirilenko & Stepchenkova, 2014).

Computer-based content analysis is currently able to identify word frequencies and patterns of words and thus can infer topics. However, the main limitation of an automated analysis of climate change communication is the difficulty of assessing in more detail what the texts are actually about regarding climate change, for example, what frames are used. Therefore, some studies combine manual and computer-assisted content analysis in order to benefit from the advantages of both types of analysis. In such cases, usually a smaller sample is drawn for manual coding out of the larger sample that has been coded automatically (Ivanova, 2015; Lörcher & Neverla, 2015).

In contrast to manual content analysis of print media or television content, the samples of digital content can be quite large because the analysis can be conducted automatically. Exemplary samples are a blog corpus comprising 1.5 million English blog posts (Fløttum, Gjesdal, Gjerstad, Koteyko, & Salway, 2014) or the analysis of 1.8 million Twitter messages worldwide on climate change (Kirilenko & Stepchenkova, 2014). To draw the samples from Twitter, researchers usually use the Twitter API—an Application Programming Interface for retrieving data from Twitter—to collect the tweets (Kirilenko & Stepchenkova, 2014).

Of course, traditional news media content can be analyzed in large samples if it is available in digital form through databases like LexisNexis. For example one study analyzed more than 100,000 stories on climate change in newspapers from the United Kingdom, United States, Germany, and France and found that climate change advocates are more prominent in media coverage than climate skeptics but that climate skeptics are more prominent in U.S. media than in European media (Grundmann & Scott, 2014).

The strength of automated content analysis thus lies in its very reliable analysis of enormous amounts of texts. Its weakness is that it cannot reconstruct more detailed and deeper, interpretative meanings inherent in texts nor can it conduct very complicated coding procedures.

Qualitative Content Analysis Designs

Qualitative content analysis relies on a closer reading of the material under study and derives prevalent themes from it through interpretative methods. It uses common qualitative techniques, for example, paraphrasing the content of climate change communication.

In general, qualitative content analysis offers a closer analysis of specific stories and different parts of the text that considers details such as sentence structure, lexicon, and linguistic expressions (Bell, 1994). Because the analysis is more detailed and requires a closer reading than quantitative analysis, it covers less material, perhaps only one month of newspaper coverage (Bacon & Nash, 2012). The type of analysis is driven by theories from social sciences, literary theories, or critical scholarship, for example using Bourdieu’s field theory to identify themes in climate change communication material (Bacon & Nash, 2012) or critical discourse analysis (CDA) (Weiss & Wodak, 2007). Also, the analysis of sub-topics and themes (e.g., Asplund, Hjerpe, & Wibeck, 2013) as well as framing analysis of climate change are often done qualitatively by deducing the frames inductively from the material (Krøvel, 2011).

One of the most common types of qualitative content analysis is discourse analysis (see also “Discourse Analysis Approaches for Assessing Climate Change Communication and Media Representations”). As discourse analysis is not a strictly defined term, many studies can be found under this label, still slightly varying in their approaches (Gavin & Marshall, 2011), for example, social constructivist analysis or rhetorical analysis. What most discourse analysis studies have in common is that they take the language—on different levels of abstractness—in which content about climate change is presented very closely into account (Gavin & Marshall, 2011). This can be done by identifying core terms in the documents under study and analyzing when they appear (e.g., Hammerstad & Boas, 2015) or by using metaphor maps (e.g., Ison, Allan, & Collins, 2015). The aim of such types of analysis is also to discover different story lines and by whom these story lines are propagated (Zannakis, 2015).

Critical discourse analysis also considers the context of an article (Olausson, 2010)—particularly the “cognitive, social, historic, and cultural contexts” (Urquijo, Stefano, & La Calle, 2015, p. 279)—as well as its language and tone (Boykoff, 2008b). Studies deal with the structure of parts of texts and so-called “critical discourse moments,” which are relevant climate change–related events (often the publication of the Intergovernmental Panel on Climate Change [IPCC] reports or international summits) (e.g., Carvalho & Burgess, 2005). Most discourse analyses of climate change focus on texts, but some also analyze visual material (from television) and take the relationship between textual and visual content into account (e.g., Olausson, 2010). Often, discourse analysis does not restrict itself to media content but covers a wide range of material, such as government bills, committee reports, interviews, and parliamentary debates (Zannakis, 2015).

Data analysis of qualitative content analysis is being conducted with specific programs such as Atlas.ti (Shaw, 2013; Urquijo, De Stefano, & La Calle, 2015). This software is able to sort different analytical categories that are derived from the material itself (e.g., linguistic and institutional mechanisms relating to water securitization in laws) (Urquijo et al., 2015) and then to further condense these categories.

In sum, qualitative content analysis studies of climate change offer an interpretative and detailed analysis of smaller numbers of texts, reconstructing meaning and often taking linguistic features into account. They are less standardized compared to quantitative content analysis and usually do not proceed with a pre-defined codebook; therefore, they do not typically produce quantifiable results of climate change content. However, they allow for a deeper analysis while taking context and genesis into account.

Typical Analytical Techniques in Content Analyses of Climate Change Communication

There are various analytical techniques that can be used in content analysis. They range from different methods of statistical analysis (descriptive and multivariate statistics) to different thematic analyses. Frequency analysis, framing analysis, and narrative analysis are three of the most frequently employed analytical techniques in content analyses of climate change communication.

Frequency Analysis

A very common and basic form of content analysis is used by studies dealing with issue salience and issue attention to climate change (e.g., Ahchong & Dodds, 2012). Issue attention to climate change is measured by counting news items (e.g., number of articles in print media, number of news stories on television) dealing with climate change over a certain period of time (e.g., Asplund, Hjerpe, & Wibeck, 2013). Therefore, such studies are called frequency or resonance analyses of climate change communication.

Schmidt, Ivanova, and Schäfer (2013) provide an overview and show that there are a number of studies that have investigated issue attention to climate change. These studies analyze media coverage in (usually national) newspapers all over the world but still with a focus on industrialized countries (Schmidt et al., 2013). From a methodological perspective these studies and the issue attention they measure are difficult to compare, as they often use different search terms, for example, “climate change” or “global warming” in the United Kingdom (Boykoff & Mansfield, 2008) or “climate change” and “forest” in Sweden (Kleinschmit & Sjöstedt, 2014).

There are some research projects that monitor climate change media coverage constantly over time, such as the Media and Climate Change Observatory (MeCCO) by the University of Colorado at Boulder, which monitors 50 news sources across 25 countries in seven different regions around the world (Gifford et al., 2016). The monitoring is done through accessing archives such as LexisNexis, ProQuest, and Factiva (using the search terms “climate change or global warming”), and its main outcome is the presentation of the number of articles per month per source.

The analytical technique used in these studies is count analysis. If different news outlets or points in time are compared, cross tabulations are used. Typical results of frequency analyses of climate change are that the media coverage of climate change is increasing (Asplund et al., 2013; Neverla & Schäfer, 2012) but to varying degrees. In order to take the development of the media coverage in general into account, Schmidt and colleagues (2013) used the share of climate change coverage in relation to the entire media coverage to determine its significance. Using correlation analysis or time series, studies dealing with such frequencies can also investigate through which external factors media attention for climate change is driven. They come to the conclusion that international events (climate summits) and NGO public relations efforts are substantial drivers of issue attention (Schäfer, Ivanova, & Schmidt, 2014).

Framing Analysis

Framing analysis attempts to investigate which aspects of climate change are presented in climate change communication and what kinds of perspectives on climate change prevail (see “Frame Analysis Approaches for Assessing Climate Change Communication and Media Portrayals” and “Framing, Discourses, and Metaphors in Media Representations of Climate Change”). Framing means to emphasize certain aspects of an issue, and to present it from a specific point of view. Aspects that constitute a frame—according to the definition by Entman (1993), which is widely used—are the problem definition of the issue, who is responsible (attribution of responsibility), what the solutions are, and how it is evaluated (Entman, 1993). Frames are used strategically by social movements and other actors (Benford & Snow, 2000) in order to establish their favorite point of view on an issue. Against this theoretical background, many content analysis studies of climate change deal with what kind of frames prevail in communication about climate change (Asplund et al., 2013; Boykoff, 2008b; Dirikx & Gelders, 2010; Wagner & Payne, 2015).

Such framing studies use different approaches to detect frames in media coverage; for an overview of methods of framing analysis see Matthes and Kohring (2008). Qualitative studies derive frames from the material in a mostly interpretative, inductive fashion (e.g., Antilla, 2005; Krøvel, 2011; Ransan-Cooper, Farbotko, McNamara, Thornton, & Chevalier, 2015; Takahashi, 2011). Different levels of analysis can be taken into account in qualitative framing analysis, from macro-level structures that include the thematic structure of an article about climate change to more micro-level structures consisting of syntactic and linguistic aspects, for example, the coherence of parts of the text or the choice of words (Olausson, 2009). Studies that follow strategies of discourse analysis consider the language used and specific words in order to identify frames (Foust & O’Shannon Murphy, 2009; Olausson, 2009; Ransan-Cooper, Farbotko, McNamara, Thornton, & Chevalier, 2015). Frames that are derived in these qualitative studies are then described in much detail and can be of very different thematic focus.

Quantitative framing analyses also differ in their approach. A very common strategy is to make use of generic frames developed by Semetko and Valkenburg (2000), who found that five frames occur frequently in news media independent of the topic at hand: the responsibility frame, conflict frame, (economic) consequences frame, human interest frame, and morality frame. With respect to climate change, researchers code how often these frames occur in media coverage (Dirikx & Gelders, 2010). A frame is coded based on a list of coding questions for the different frames. For example, the question “Does the story suggest that some level of government has the ability to alleviate the problem?” is answered to operationalize the attribution of responsibility frame (Dirikx & Gelders, 2010, p. 736). Such studies show which perspective on climate change prevails in the content under study, also differentiating between different time periods and news outlets. This approach is deductive as it defines the frames beforehand and the content analysis aims to discover these frames in the material. Many studies follow this approach, not only using the generic frames by Semetko and Valkenburg (2000), but also introducing other deductive frames related to climate change, such as the frame “science and technology” (Wagner & Payne, 2015, p. 8) or the “political economic” frame (Boykoff, 2008b, p. 556). One of the first studies to do this was Trumbo’s (1996) framing analysis of the climate change debate resulting in the “climate change frame,” which is challenged by several other frames, such as the “scientific uncertainty frame” (Shehata & Hopmann, 2012). This deductive set of frames appears across different countries and media outlets (Nisbet, 2009; O’Neill, Williams, Kurz,Wiersma, & Boykoff, 2015). Researchers can adapt these frames and use them for their analyses of climate change coverage.

This approach is deductive and does not generate frames from the material itself—but other framing studies do exactly this. They do not code holistic frames but instead take an inductive approach by coding different frame elements and then merging these elements after the coding through statistical analysis. For example, coding categories for frame elements entail codes on science and impacts and codes on the attribution of responsibility (Billett, 2010). Other studies make use of the concept of idea elements that constitute a frame, code these idea elements, and then merge them into a frame (Kaiser & Rhomberg, 2015). Statistical methods such as cluster analysis can be used in order to find patterns among the frame elements, and these patterns are then interpreted as frames (Wozniak, Lück, & Wessler, 2014).

Narrative Analysis

A typical analytical strategy used mostly in qualitative content analyses is narrative analysis (see also “Agenda Building, Narratives, and Attention Cycles in Climate Change News Coverage” and “Narrative Persuasion and Storytelling as Climate Communication Strategies”). It focuses on the narrative structure of texts and “involves reconstruction of the composition of the narrative” (Neuendorf, 2002, p. 5), thus following an interpretative paradigm (Wozniak et al., 2014). Narrative structure means that events, subjects, objects, abstract concepts, and actions are integrated in a coherent story (Viehöver, 2012, p. 178). Narratives thus serve as organizing principles to construct meaning.

In essence, studies using narrative analysis in the context of climate change communication are investigating what kind of storylines, actors, and themes are predominant in texts on climate change. This also includes taking into account what stylistic elements are used in stories about climate change, for example to create suspense, emotion, or resolution (Wozniak et al., 2014). Most studies following a narrative approach examine print media or texts, but some also include television (e.g., Krøvel, 2011). Some of these studies formulate questions that guide the narrative analysis, as Krøvel (2011) does in his analysis of the media coverage of the Conference of the Parties (COP) in Bali in 2007. Examples of his guiding questions are: “Who are constructed as ‘protagonists’ and ‘opponents’ in the negotiations at Bali? What is implicitly or explicitly constructed as the ‘goal’?” (Krøvel, 2011, p. 93). Most narrative analyses are based on qualitative content analysis, but studies that combine narration and quantitative content analysis exist, such as McComas and Shanahan’s (1999) investigation of which narrative factors in the media coverage of climate change explain cyclical changes in that coverage.

Studies using this type of analytical approach typically find that different narratives in climate change communication exist and that these differ between countries. Viehöver (2012) identifies six different narratives in the German climate change debate between 1970 and 2011, ranging from the “global greenhouse effect as anthropogenic catastrophe” to the “story of the nuclear winter” (Viehöver, 2012, pp. 190–209). The “apocalyptic” narrative of climate change constitutes a dominant storyline in the United States (Smith, 2012), particularly in the elite and popular press (Foust & O’Shannon Murphy, 2009).

These different analytical techniques are used to answer typical questions in climate change coverage (cf. “Aims of Content Analysis Methods in Climate Change Communication”) and often come along with either quantitative or qualitative content analyses (cf. “Content Analysis Designs in Climate Change Communication Research”). To summarize, Table 1 provides an overview of typical aims of content analyses in climate change communication, the associated methods and samples, typical findings, as well as studies that can serve as examples for the respective research aim and method.

Table 1. Overview of typical aims, methods, and findings in content analyses of climate change communication

Typical aims/research questions

Method

Samples

Typical findings

Exemplary studies

Does climate change coverage change over time?

Quantitative analysis (manual or automated), frequency analysis

Mostly newspaper coverage (often elite press)

Coverage of climate change has increased over the last decades

Aykut, Comby, and Guillemot (2012) and Liu et al. (2011)

Does climate change coverage differ between countries or media outlets?

Quantitative analysis (manual or automated), frequency and/or frame analysis

Mostly newspaper coverage (often elite press)

Strong differences between countries (e.g., attention lower in German-speaking countries than in Anglophone countries), differences in amount of coverage between media outlets

Boykoff (2007) and Schmidt et al. (2013)

What kind of linguistic corpus is associated with climate change?

Quantitative analysis (often automated)

Often online content

Identification of clusters of textual compounds

Koteyko (2010) and Koteyko et al. (2010)

Which themes are embedded in climate change discourse?

Qualitative analysis (manual), discourse analysis

All kinds of content (print media, television, official documents)

Construction of identity or climate responsibility through climate change discourse

Olausson (2010) and Zannakis (2015)

Which perspectives on climate change are presented?

Quantitative or qualitative analysis (manual), frame analysis

All kinds of content, but mostly print media

Different frames: e.g., science and technology frame, scientific uncertainty frame

Dirikx and Gelders (2010), Olausson (2009), and Trumbo (1996)

What storylines appear in climate change communication?

Qualitative analysis (manual), narrative analysis

Textual content (mostly print media)

Different narratives: e.g., apocalyptic narrative

Krøvel (2011), Smith (2012), and Viehöver (2012)

Specifics of the Material Used in Content Analysis

As shown, most content analysis of climate change communication analyzes legacy media. Although less frequent, content analyses have also been used to scrutinize online content, campaign communication, official science and policy documents, and images.

Online Content

When online content is under investigation in climate change communication research (see “Blog, Twitter, and Other Social Media Depictions of Climate Change”), the definition of the sample is more complex, as the amount of Web content on climate change is vast (e.g., Schäfer, 2012, p. 6). When analyzing this content, one strategy for researchers is to undertake a Google search with the same search terms they would use for print media archives. They take the search results list and analyze the first results appearing (for example, the first 10 pages from the Google search), arguing that the public barely looks at search results beyond the first search page (Gavin & Marshall, 2011). This method ensures that all possible Web content is considered, yet one has to take the algorithms into account, which influence results of a Google search. When the aim is to investigate climate change communication on specific websites, for example websites of climate change organizations, a possible way to put together the sample of websites is through indexes and link collections provided by other websites (Jun, 2011).

No matter in what way the sample is put together, cleaning the sample is important when studying online communication on climate change, as the results of online searches need to be reassessed regarding repetitions and non-active websites (Jun, 2011). Researchers also have to face the problem that some online content changes quickly (e.g., news results) and need to deal with the problem of reliability if they access the content at different times. What is usually done is that web content is downloaded at one pre-determined point of time so that all content is gathered at the same time and will not be changed after. Searches from different computers may produce different search results as the user IP differs, which must also be considered.

One of the first studies investigating online communication about climate change studied the websites of conservative think tanks from 1990 to 1997, including documents that were available for download on these websites such as speech transcripts or book sections (McCright & Dunlap, 2000). Another frequent research strategy is to code both online content and mass media content about climate change (Mix & Waldo, 2015) or to analyze both print media content and the respective online outlets of these print media (e.g., BBC News online) (Shaw, 2013). Specifically with climate change communication, online content of science-related websites, such as the online news sections of scientific journals (Nature and Science) can be of interest to researchers (Nielsen & Kjaergaard, 2011). Online content is often studied using corpus linguistic procedures, as linguistic structures can be analyzed using automated software. One study investigated what kind of compounds can be found regarding the term “carbon” in RSS feeds and with which other textual elements these linguistic compounds co-occur (Koteyko, 2010; Koteyko, Thelwall, & Nerlich, 2010).

More recently, content analysis has been used to investigate climate change communication in Web 2.0 applications such as social networks or blogs. Blogs with explicit relation to climate change are often under study (Fløttum, Gjesdal, Gjerstad, Koteyko, & Salway, 2014), and campaign communication Facebook posts of strategic actors are analyzed with respect to their content about climate change (Katz-Kimchi & Manosevitch, 2015). Content analysis of the discussion of climate change on microblogging services like Twitter have recently become a trending topic in climate change communication research (Kirilenko & Stepchenkova, 2014; Rauchfleisch, 2015). Apart from this textual content, online videos from YouTube are integrated in the content analysis when analyzing political speeches (Eckersley, 2015). Compared to the climate change coverage of mass media—no matter whether it appears online or in the traditional format—content on blogs, microblogs and social networks includes communication by laypeople as opposed to journalists, and in the interactive Web 2.0 sphere can be commented on by other online users. These comments again can be included in content analyses of climate change communication (Koteyko, Jaspal, & Nerlich, 2013).

A recent issue when studying online content is the consideration of astroturfing in campaign communication. Astroturfing means that Web content is created because corporations pay a public relations firm to promote their desired claims through false grassroots campaigns and coalitions on the Internet (Mix & Waldo, 2015). The challenge for researchers is to identify this type of content and to decide how to treat it in their analysis.

Campaign Communication

Campaigners are important actors in climate change communication as they follow specific and strategic aims with their communicative actions. As campaigns play an important role for a strongly debated issue like climate change, studies use content analysis to study campaigns or material of campaign organizations, such as their public relations efforts (Jun, 2011). This content of campaign communication is often coded with regard to the tactic the campaign organization employs, for example, what the strategic aim of a Facebook post by Greenpeace is (Katz-Kimchi & Manosevitch, 2015). Campaign communication often changes over time, thus when trying to follow how a campaign emerged and developed often both traditional mass media content and Web content are analyzed at the same time to compare the two channels (e.g., Mix & Waldo, 2015).

Scientific and Policy Documents

In addition to mass media and journalistic content as well as campaign communication, other types of material on climate change, which are often aimed at a more specialized audience, such as scientists or politicians, can be the focus of content analyses. One of the most prominent of these are official reports by institutions, such as the Human and World Development Reports (Gasper, Portocarrero, & St. Clair, 2013; Fløttum & Dahl, 2012). Researchers need to take into account for their content analysis that different reports have different strategic aims resulting from their varied institutional and organizational backgrounds and thus might differ in how they deal with climate change (Fløttum & Dahl, 2012).

A common approach in this field is the analysis of Intergovernmental Panel on Climate Change (IPCC) reports or documents from the Conference of the Parties (COP) conferences. Bassett and Fogelman (2013), for example, study the role of the adaptation concept in these reports while Fløttum and Dahl (2011) focus on claims on complexity and uncertainty of climate change and Vihersalo (2008) on political statements that were uttered at climate conferences. This type of content is very closely related to scientific research on climate change. Some studies therefore compare the IPCC reports and scientific literature with the media coverage of climate change, for example with respect to specific scientific findings such as the projections of the sea level rise (Rick, Boykoff, & Pielke Jr., 2011). This content analytical approach follows the implicit assumption that mass media should cover scientific research reflecting the content of scientific documents (figures, probabilities, projections) as exactly and correctly as possible without further interpretation. Although aiming more at a mainstream audience, even popular books on climate change (Shaw, 2013) and special interest media like Swedish farm magazines (Asplund, Hjerpe, & Wibeck, 2013) have been used for content analysis.

Policy material is another type of content that is analyzed with regard to climate change. Policy makers issue documents or statements on climate change that can be assessed through content analysis, such as the UK government’s Ministry of Defence (MOD) strategic defense reviews and its Cabinet Office national security strategies (Hammerstad & Boas, 2015). In the United States, the Lexis-Nexis Congressional Publications archive provides access to congressional testimonies that can be used for content analysis (Liu, Vedlitz, Stoutenborough, & Robinson, 2015). Depending on the researchers’ disciplines, documents about even more specific topics, such as UK and Australian water management planning documents (Ison, Allan, & Collins, 2015), food security policy documents in Nepal (Nagoda, 2015), actual laws addressing water securitization (Urquijo, De Stefano, & La Calle, 2015), or low-carbon city plans and their statements on emissions and the environment (Zhou et al., 2015), can be under investigation in climate change communication research.

Content analyses of these documents are basically conducted in the same way mass media content is analyzed: a codebook is developed that fits the specifics of the documents under study. Some studies take linguistic approaches to analyze these reports, by investigating the use of specific words (“can,” “important,” “but,” . . .) (Fløttum & Dahl, 2012), while others take a qualitative approach and investigate these reports from a contextual and structural perspective (Gasper et al., 2013).

Images

By now there are quite a few studies dealing with content analysis of images in climate change communication (see overview in Metag, Schäfer, Füchslin, Barsuhn, & Kleinen-von Königslöw, 2016; O’Neill and Smith, 2014) (see also “Visual Depictions and Media Images of Climate Change and Their Effects on Audience Perceptions,” “Affective Imagery, Social Representations, and Public Perceptions of Climate Change,” and “Methods for Assessing Visual Images and Depictions of Climate Change”). One of the first decisions that must be made for a content analysis of images concerns the sampling, that is, out of which material the pictures are taken. Most of the studies take the images they analyze from print media (e.g., DiFrancesco & Young, 2011), some of them also analyze the articles the images accompany (DiFrancesco & Young, 2011) and take the image captions into consideration (O’Neill, 2013). If television coverage of climate change is under study and the visual content is incorporated, researchers rely on news items as their basis of analysis (e.g., Hoijer, 2010; Lester & Cottle, 2009). However, it is not always clear how the visuals are then analytically demarcated from each other and what is treated as a single image. Images used in campaign material, such as material published by Greenpeace (Doyle, 2007; Slocum, 2004) or other NGOs such as Oxfam or Christian Aid (Manzo, 2010), have also been the focus of visual content analysis.

A content analysis of visual communication brings about different challenges compared to the analysis of textual content because visual representations of climate change include a less manifest, connotative meaning. Thus, the development of a standardized codebook is more complex and qualitative approaches are often used. In general, content analyses of climate change imagery are often descriptive and focus on specific cases of climate change imagery being used (e.g., Manzo, 2010; Slocum, 2004). If a quantitative content analysis of images is conducted a codebook is developed, similar to textual analyses, which entails categories such as the image theme (often with main category and sub category) (DiFrancesco & Young, 2011; Eide, 2012; O’Neill, 2013; Smith & Joffe, 2009) and, if the image is taken from newspaper articles, variables like newspaper name and article publication date are coded (O’Neill, 2013). Images can be coded with regard to their function as symbolic, iconic, or spectacular (Lester & Cottle, 2009) and according to their content (e.g., environment/risk). Content analyses of images of climate change have established that the five most common image themes are climate change impacts and threats, nature themes, people or talking heads, graphs and models, and carbon emissions/energy issues (Metag, Schäfer, Füchslin, Barsuhn, & Kleinen-von Königslöw, 2016).

Standardized content analyses of larger samples of images pose challenges because the databases through which newspaper articles are accessed (LexisNexis, ProQuest) often do not provide the images belonging to the print media articles (O’Neill, 2013). Images are therefore often retrieved through the online search portals of newspapers (O’Neill, 2013), which is feasible if print newspapers and their online versions do not differ significantly.

Qualitative approaches to climate change imagery are particularly used to explore images in more detail and to identify visual frames. Visual frames in climate change imagery are difficult to capture systematically (O’Neill, 2013). O’Neill (2013), for instance, analyzes the images qualitatively with regard to their denotative, connotative, and ideological content.

Key Categories in Content Analyses of Climate Change Communication

From the many studies using content analysis of climate change communication emerge some key themes and the variables that are coded in order to capture these themes. One of these is issue attention; studies investigating how much attention climate change receives in mass media over time, at specific events, in different countries, in different outlets. This is not limited to climate change as a general issue but includes specific aspects of climate change, such as extreme weather events (Ungar, 1999). Analyses with issue attention as a key theme also in some cases investigate whether there are correlations between real-word indicators (e.g., weather data) and issue salience of global warming in the media coverage (Ungar, 1999).

When a codebook for a content analysis is developed it usually consists of formal as well as topical coding categories. Formal categories include characteristics describing the formal appearance of the coding unit, such as the date on which an article dealing with climate change appeared. The most common topical categories are the main topic of the news item, the main actors who get a voice in the news item, and the opinion on climate change as an issue.

Grasping the viewpoint or stance on climate change is another key theme in content analysis of climate change coverage. Variables are trying to discern whether articles, television news items, or single statements accept or dismiss climate change (e.g., Feldman, Maibach,Roser-Renouf, & Leiserowitz, 2012). These coding categories may include what type of statements are made about the scientific consensus about climate change, the certainty of the existence of climate change, and the human causes of climate change (Feldman et al., 2012). When operationalizing different viewpoints on climate change in their codebooks, many studies draw on established coding schemes such as the one by Boykoff and Boykoff (2004). It focuses on the extent to which different statements in the text yield different viewpoints on climate change. Statements are coded according to whether they “(1) presented the viewpoint that anthropogenic global warming (distinct from natural variations) accounts for all climate changes, (2) presented multiple viewpoints, but emphasized that anthropogenic contributions, distinct from yet still in combination with natural variation, significantly contribute to climate changes (most accurately communicating the dominant view from climate science), (3) gave ‘a balanced account’ surrounding existence and non-existence of anthropogenic climate change, and (4) presented multiple viewpoints but emphasized the claim that anthropogenic component contributes negligently to changes in the climate” (Boykoff, 2008a, p. 5). Coding categories may also include specific aspects such as the two degree limit (Shaw, 2013), sea level rise (Rick et al., 2011), or climategate2(Nerlich, 2010). Variables capturing different perspectives and attitudes toward climate change expressed in media portrayals are then often used to discover frames in the climate change coverage (e.g., Asplund, Hjerpe, & Wibeck, 2013; Krøvel, 2011).

Analyses of official documents require (partly) different categories than analyses of journalistic content. For example, a study on congressional testimonies of scientists’ views on climate change codes the scientists’ background information (e.g., scientific background and institutional affiliation) as well as the scientists’ stance on global warming with respect to its existence and human causation (Liu, Vedlitz, Stoutenborough, & Robinson, 2015).

Evaluative Techniques and Challenges of Content Analysis as a Method

Assessing the reliability and validity of content analyses belongs to the methodological standard evaluations. “A research procedure is reliable when it responds to the same phenomena in the same way regardless of the circumstances of its implementation” (Krippendorff, 2013, p. 267). Validity, on the other hand, assesses whether the research instrument actually measures the construct it is supposed to measure.

In a manual content analysis, the most important measure is intercoder reliability. Reliability scores indicate to what extent different coders agree in their application of the coding categories, that is, to what extent they code the same variables for the same material. A reliability test is thus a test for measurement consistency. Commonly, when conducting a standardized content analysis, reliability tests are conducted and reliability scores are documented when the study is published. There are different types of reliability coefficients used in content analyses of climate change communication that are based on different kind of mathematical formulas, including Krippendorff’s alpha (Billett, 2010) and Scott’s Pi (Katz-Kimchi & Manosevitch, 2015; Nielsen & Kjaergaard, 2011).

Best practice for studies using content analysis as a method is to document the size of the sample that was used for the reliability test, which kind of coefficient was used to measure reliability, and the coefficient for each variable. Most studies in the field of content analyses of climate change communication document at least an overall reliability coefficient in their publications. However, there are studies that do not document a reliability coefficient at all. And those studies that provide only overall reliability coefficients may lack information on the coefficients for each variable. This information should be provided because the overall reliability score gives only an average of all coding categories, leaving open whether each category itself actually yields sufficient reliability. Another problem is studies that provide a figure but do not indicate which kind of reliability coefficient they used (e.g., Jun, 2011). Only some studies document actual coding categories or explain them in detail—studies that do this well are Billett (2010) and Liu, Lindquist, and Vedlitz (2011)—and provide sufficient reliability scores. An example for a comprehensible and adequate documentation of variables and reliability coefficients is the analysis by Nielsen and Kjaergaard (2011).

Validity is more complex to measure than reliability. One can distinguish between external validity—meaning the generalizability of a content analysis—and internal validity, meaning the closeness of the theoretical construct and its measurement (Neuendorf, 2002). This internal validity can be differentiated in criterion, content, and construct validity. Criterion validity tries to assess the validity of a construct measured in a content analysis through a comparison with other constructs not measured in the analysis while content validity evaluates whether the analysis reflects all aspects of a construct (Neuendorf, 2002) (all aspects of the issue of climate change). Finally, construct validity is estimated by testing whether the measure employed in the content analysis relates to other measures as it was theoretically expected (Neuendorf, 2002); for example, measures of alarmist perspectives on global warming should be positively related to measures of actions against climate change. Whether a content analysis measures the construct it is supposed to measure can be assessed by providing a coding scheme that transparently describes the variables, definitions, and rules for coding (Potter & Levine‐Donnerstein, 1999).

From a methodological point of view, different types of content analyses offer different advantages and disadvantages and bring about varying challenges. While qualitative content analysis takes specificities and more detailed relations and connections of statements into account, quantitative analysis offers a more generalizable overview of climate change communication. One challenge of content analysis as a method for assessing climate change communication concerns the comparability of different studies using content analysis. Not only do these studies start with different research questions and aims, they also differ with regard to the analyzed time frames and media (cf. Schmidt, Ivanova, & Schäfer, 2013). And even studies of the media coverage of climate change that had been planned as comparative studies from the beginning often face problems, such as measures that are not equivalent so that they compare indicators that measure slightly different constructs (cf. Schmidt et al., 2013). Schmidt and colleagues (2013, p. 1234) give the example that in some studies absolute numbers of newspaper articles on climate change are compared although these numbers are influenced by other factors that are not comparable, such as the size of the newspaper.

A challenge in the field of automated content analysis is that although the analysis of large text corpora can provide information on attention cycles and word combinations—word combinations are sometimes regarded as topics (e.g., Kirilenko, Stepchenkova, Romsdahl, & Mattis, 2012)—it is not yet fully able to investigate details on how climate change is framed by what kind of actors because it often cannot provide sufficient information on the actual meaning of the word combinations retrieved. There is, however, substantial research in this area suggesting innovative designs to analyze frames automatically. Ivanova (2015) identifies 100 topics in climate change media coverage based on latent dirichlet allocation procedures and then codes these topics according to Entman’s (1993) framing definition.

Another challenge in the field affects the different modes in which climate change communication can appear. Most content analyses focus on either textual content or images. Recent research suggests drawing these different foci together in a multimodal research design (Wozniak, Lück, & Wessler, 2014) to better adhere to the different modes climate change can be presented in.

Many content analyses are based on the assumption that the content of climate change communication affects public perceptions of climate change to some extent. However, there is only a limited number of studies drawing content analyses of climate change communication and audience research together (e.g., Feldman, Maibach, Roser-Renouf, & Leiserowitz, 2012). As it is often a very complex and demanding undertaking to gather content analysis data as well as survey data on different audiences and connect these data sets sufficiently, reliable results on the effects of media coverage of climate change on audience perceptions are scarce. Research in this field still needs to adhere to this challenge.

Acknowledgments

I am thankful to Mike S. Schäfer, Matthew Nisbet and the anonymous reviewers for valuable comments on this article.

References

Ahchong, K., & Dodds, R. (2012). Anthropogenic climate change coverage in two Canadian newspapers, the Toronto Star and the Globe and Mail, from 1988 to 2007. Environmental Science & Policy, 15, 48–59.Find this resource:

Anderson, A. (2011). Sources, media, and modes of climate change communication: The role of celebrities. WIREs Climate Change, 2(4), 535–546.Find this resource:

Antilla, L. (2005). Climate of scepticism: US newspaper coverage of the science of climate change. Global Environmental Change, 15(4), 338–352.Find this resource:

Antilla, L. (2010). Self-censorship and science: A geographical review of media coverage of climate tipping points. Public Understanding of Science, 19(2), 240–256.Find this resource:

Asplund, T., Hjerpe, M., & Wibeck, V. (2013). Framings and coverage of climate change in Swedish specialized farming magazines. Climatic Change, 117(1–2), 197–209.Find this resource:

Aykut, S. C., Comby, J.-B., & Guillemot, H. (2012). Climate change controversies in French mass media 1990–2010. Journalism Studies, 13(2), 157–174.Find this resource:

Bacon, W., & Nash, C. (2012). Playing the media game: The relative (in)visibility of coal industry interests in media reporting of coal as a climate change issue in Australia. Journalism Studies, 13(2), 243–258.Find this resource:

Bassett, T. J., & Fogelman, C. (2013). Déjà vu or something new?: The adaptation concept in the climate change literature. Geoforum, 48, 42–53.Find this resource:

Bell, A. (1994). Climate of opinion—public and media discourse on the global environment. Discourse & Society, 5(1), 33–64.Find this resource:

Benford, R. D., & Snow, D. A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1), 611–639.Find this resource:

Berglez, P., Höijer, B., & Olausson, U. (2009). Individualisation and nationalisation of the climate issue: Two ideological horizons in Swedish news media. In T. Boyce & J. Lewis (Eds.), Climate change and the media (pp. 211–223). New York: Peter Lang.Find this resource:

Billett, S. (2010). Dividing climate change: Global warming in the Indian mass media. Climatic Change, 99(1), 1–16.Find this resource:

Boyce, T., & Lewis, J. (Eds.). (2009). Global crises and the media: Vol. 5. Climate change and the media. New York: Lang.Find this resource:

Boykoff, J. (2012). US media coverage of the Cancún climate change conference. Ps: Political Science & Politics, 45(2), 251–258.Find this resource:

Boykoff, M. T. (2007). From convergence to contention: United States mass media representations of anthropogenic climate change science. Transactions of the Institute of British Geographers, 32(4), 477–489.Find this resource:

Boykoff, M.T. (2008a). Lost in translation?: United States television news coverage of anthropogenic climate change, 1995–2004. Climatic Change, 86, 1–11.Find this resource:

Boykoff, M. T. (2008b). The cultural politics of climate change discourse in UK tabloids. Political Geography, 27(5), 549–569.Find this resource:

Boykoff, M. T., & Boykoff, J. M. (2004). Balance as bias: Global warming and the US prestige press. Global Environmental Change, 14(2), 125–136.Find this resource:

Boykoff, M. T., & Mansfield, M. (2008). “Ye Olde Hot Aire”: Reporting on human contributions to climate change in the UK tabloid press. Environmental Research Letters, 3(2), 8.Find this resource:

Carvalho, A., & Burgess, J. (2005). Cultural circuits of climate change in UK broadsheet newspapers, 1985–2003. Risk Analysis, 25(6), 1457–1469.Find this resource:

DiFrancesco, D. A., & Young, N. (2011). Seeing climate change: The visual construction of global warming in Canadian national print media. Cultural Geographies, 18(4), 517–536.Find this resource:

Dirikx, A., & Gelders, D. (2010). To frame is to explain: A deductive frame-analysis of Dutch and French climate change coverage during the annual UN Conferences of the Parties. Public Understanding of Science, 19(6), 732–742.Find this resource:

Doyle, J. (2007). Picturing the clima(c)tic: Greenpeace and the representational politics of climate change communication. Science as Culture, 16(2), 129–150.Find this resource:

Eckersley, R. (2015). National identities, international roles, and the legitimation of climate leadership: Germany and Norway compared. Environmental Politics, 25(1), 180–201.Find this resource:

Eide, E. (2012). Visualizing a global crisis: Constructing climate, future and present. Conflict & Communication Online, 11(2), 1–16.Find this resource:

Eide, E., & Kunelius, R. (2012). Media meets climate: The global challenge for journalism. Göteborg, Sweden: Nordicom.Find this resource:

Eide, E., Kunelius, R., & Kumpu, V. (2010). Global climate, local journalisms: A transnational study of how media make sense of climate summits. Global journalism research series 3. Bochum, Germany: Projectverlag.Find this resource:

Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58.Find this resource:

Feldman, L., Maibach, E. W., Roser-Renouf, C., & Leiserowitz, A. (2012). Climate on cable: The nature and impact of global warming coverage on Fox News, CNN, and MSNBC. International Journal of Press/Politics, 17(1), 3–31.Find this resource:

Fløttum, K., & Dahl, T. (2011). Climate change discourse: Scientific claims in a policy setting. Fachsprache, 3–4, 205–219.Find this resource:

Fløttum, K., & Dahl, T. (2012). Different contexts, different “stories”? A linguistic comparison of two development reports on climate change. Language & Communication, 32(1), 14–23.Find this resource:

Fløttum, K., Gjesdal, A. M., Gjerstad, Ø., Koteyko, N., & Salway, A. (2014). Representations of the future in English language blogs on climate change. Global Environmental Change, 29, 213–222.Find this resource:

Foust, C. R., & O’Shannon Murphy, W. (2009). Revealing and reframing apocalyptic tragedy in global warming discourse. Environmental Communication, 3(2), 151–167.Find this resource:

Gasper, D., Portocarrero, A. V., & St. Clair, A. L. (2013). The framing of climate change and development: A comparative analysis of the Human Development Report 2007/8 and the World Development Report 2010. Global Environmental Change, 23(1), 28–39.Find this resource:

Gavin, N. T., & Marshall, T. (2011). Mediated climate change in Britain: Scepticism on the web and on television around Copenhagen. Global Environmental Change-Human and Policy Dimensions, 21(3), 1035–1044.Find this resource:

Gifford, L., Luedecke, G., McAllister, L., Nacu-Schmidt, A., Andrews, K., Boykoff, M., and Daly, M. (2016). World Newspaper Coverage of Climate Change or Global Warming, 2004–2016. Center for Science and Technology Policy Research, Cooperative Institute for Research in Environmental Sciences, University of Colorado, Web. [22.07.2016].

Gordon, J. C., Deines, T., & Havice, J. (2010). Global warming coverage in the media: Trends in a Mexico City newspaper. Science Communication, 32(2), 143–170.Find this resource:

Grundmann, R. (2007). Climate change and knowledge politics. Environmental Politics, 16(3), 414–432.Find this resource:

Grundmann, R., & Scott, M. (2014). Disputed climate science in the media: Do countries matter?. Public Understanding of Science, 23(2), 220–235.Find this resource:

Hammerstad, A., & Boas, I. (2015). National security risks? Uncertainty, austerity and other logics of risk in the UK government’s National Security Strategy. Cooperation and Conflict, 50(4), 475–491.Find this resource:

Hoijer, B. (2010). Emotional anchoring and objectification in the media reporting on climate change. Public Understanding of Science, 19(6), 717–731.Find this resource:

Ison, R., Allan, C., & Collins, K. (2015). Reframing water governance praxis: Does reflection on metaphors have a role?. Environment and Planning C: Government and Policy, 33(6), 1697–1713.Find this resource:

Ivanova, A. (2015). Transnationalisierung von Öffentlichkeiten: Eine länderübergreifende Längsschnittanalyse der Berichterstattungzum Klimawandel in führenden Tageszeitungen. Dissertation. Hamburg, Germany: University of Hamburg.Find this resource:

Jun, J. (2011). How climate change organizations utilize websites for public relations. Public Relations Review, 37(3), 245–249.Find this resource:

Kaiser, J., & Rhomberg, M. (2015). Questioning the doubt: Climate skepticism in German newspaper reporting on COP17. Environmental Communication, 1–19.Find this resource:

Katz-Kimchi, M., & Manosevitch, I. (2015). Mobilizing Facebook users against Facebook’s energy policy: The case of Greenpeace Unfriend Coal Campaign. Environmental Communication, 9(2), 248–267.Find this resource:

Kirilenko, A., Stepchenkova, S., Romsdahl, R., & Mattis, K. (2012). Computer-assisted analysis of public discourse: A case study of the precautionary principle in the US and UK press. Quality & Quantity, 46, 501–522.Find this resource:

Kirilenko, A. P., & Stepchenkova, S. O. (2014). Public microblogging on climate change: One year of Twitter worldwide. Global Environmental Change, 26, 171–182.Find this resource:

Kleinschmit, D., & Sjöstedt, V. (2014). Between science and politics: Swedish newspaper reporting on forests in a changing climate. Environmental Science & Policy, 35, 117–127.Find this resource:

Koteyko, N. (2010). Mining the Internet for linguistic and social data: An analysis of “carbon compounds” in Web feeds. Discourse & Society, 21(6), 655–674.Find this resource:

Koteyko, N., Jaspal, R., & Nerlich, B. (2013). Climate change and “climategate” in online reader comments: A mixed methods study. Geographical Journal, 179, 74–86.Find this resource:

Koteyko, N., Thelwall, M., & Nerlich, B. (2010). From carbon markets to carbon morality: Creative compounds as framing devices in online discourses on climate change mitigation. Science Communication, 32(1), 25–54.Find this resource:

Krippendorff, K. (2013). Content analysis: An introduction to its methodology (3d ed.). Los Angeles: SAGE.Find this resource:

Krøvel, R. (2011). Journalistic narratives of success and failure at the Bali Climate Change Conference in 2007. Intercultural Communication Studies, 20(2), 89–104.Find this resource:

Lester, L., & Cottle, S. (2009). Visualizing climate change: Television news and ecological citizenship. International Journal of Communication, 3, 920–936.Find this resource:

Liu, X., Lindquist, E., & Vedlitz, A. (2011). Explaining media and congressional attention to global climate change, 1969–2005: An empirical test of agenda-setting theory. Political Research Quarterly, 64(2), 405–419.Find this resource:

Liu, X., Vedlitz, A., Stoutenborough, J. W., & Robinson, S. (2015). Scientists’ views and positions on global warming and climate change: A content analysis of congressional testimonies. Climatic Change, 131(4), 487–503.Find this resource:

Lörcher, I., & Neverla, I. (2015). The dynamics of issue attention in online communication on climate change. Media and Communication, 3(1), 17.Find this resource:

Manzo, K. (2010). Imaging vulnerability: The iconography of climate change. Area, 42(1), 96–107.Find this resource:

Matthes, J., & Kohring, M. (2008). The content analysis of media frames: Toward improving reliability and validity. Journal of Communication, 58(2), 258–279.Find this resource:

McComas, K., & Shanahan, J. (1999). Telling stories about global climate change—measuring the impact of narratives on issue cycles. Communication Research, 26(1), 30–57.Find this resource:

McCright, A. M., & Dunlap, R. E. (2000). Challenging global warming as a social problem: An analysis of the conservative movement’s counter-claims. Social Problems, 47(4), 499–522.Find this resource:

Metag, J., Schäfer, M. S., Füchslin, T., Barsuhn, T., & Kleinen-von Königslöw, K. (2016). Perceptions of climate change imagery: Evoked salience and self-efficacy in Germany, Switzerland, and Austria. Science Communication, 38(2), 197–227.Find this resource:

Mix, T. L., & Waldo, K. G. (2015). Know(ing) your power: Risk society, Astroturf campaigns, and the battle over the Red Rock coal-fired plant. Sociological Quarterly, 56(1), 125–151.Find this resource:

Nagoda, S. (2015). New discourses but same old development approaches? Climate change adaptation policies, chronic food insecurity and development interventions in northwestern Nepal. Global Environmental Change, 35, 570–579.Find this resource:

Nash, C., & Bacon, W. (2006). Reporting sustainability in the English-language press of Southeast Asia. Pacific Journalism Review, 12(2), 106–135.Find this resource:

Nerlich, B. (2010). “Climategate”: Paradoxical metaphors and political paralysis. Environmental Values, 19(4), 419–442.Find this resource:

Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks, CA: SAGE.Find this resource:

Neverla, I., & Schäfer, M. S. (Eds.). (2012). Das Medien-Klima. Wiesbaden, Germany: Springer VS.Find this resource:

Nielsen, K. H., & Kjaergaard, R. S. (2011). News coverage of climate change in Nature News and ScienceNOW during 2007. Environmental Communication—a Journal of Nature and Culture, 5(1), 25–44.Find this resource:

Nisbet, M. C. (2009). Communicating climate change: Why frames matter for public engagement. Environment: Science and Policy for Sustainable Development, 51(2), 12–23.Find this resource:

Olausson, U. (2009). Global warming—global responsibility? Media frames of collective action and scientific certainty. Public Understanding of Science, 18(4), 421–436.Find this resource:

Olausson, U. (2010). Towards a European identity? The news media and the case of climate change. European Journal of Communication, 25(2), 138–152.Find this resource:

O’Neill, S., Williams, H. T. P., Kurz, T., Wiersma, B., & Boykoff, M. (2015). Dominant frames in legacy and social media coverage of the IPCC Fifth Assessment Report. Nature Climate Change, 5(4), 380–385.Find this resource:

O’Neill, S. J. (2013). Image matters: Climate change imagery in US, UK and Australian newspapers. Geoforum, 49, 10–19.Find this resource:

O’Neill, S. J., & Smith, N. (2014). Climate change and visual imagery. Wiley Interdisciplinary Reviews: Climate Change, 5(1), 73–87.Find this resource:

Painter, J. (2011). Poles apart: The international reporting on climate scepticism. London: Reuters Institute for the Study of Journalism.Find this resource:

Painter, J., & Ashe, T. (2012). Cross-national comparison of the presence of climate scepticism in the print media in six countries, 2007–10. Environmental Research Letters, 7(4), 1–8.Find this resource:

Potter, W. J., & Levine‐Donnerstein, D. (1999). Rethinking validity and reliability in content analysis. Journal of Applied Communication Research, 27(3), 258–284.Find this resource:

Ransan-Cooper, H., Farbotko, C., McNamara, K. E., Thornton, F., & Chevalier, E. (2015). Being(s) framed: The means and ends of framing environmental migrants. Global Environmental Change, 35, 106–115.Find this resource:

Rauchfleisch, A. (2015). Two weeks on Twitter: COP21, smoking heads and tweets from outer space.

Rebich-Hespanha, S., Rice, R. E., Montello, D. R., Retzloff, S., Tien, S., & Hespanha, J. P. (2014). Image themes and frames in US print news stories about climate change. Environmental Communication, 9(4), 491–519.Find this resource:

Rick, U. K., Boykoff, M. T., & Pielke, R. A., Jr. (2011). Effective media reporting of sea level rise projections: 1989–2009. Environmental Research Letters, 6(1).Find this resource:

Rössler, P. (2005). Inhaltsanalyse:UTB. Konstanz, Germany: UVK-Verl.-Ges.Find this resource:

Rowe, D. (2011). Comparing newspaper coverage of climate change during election campaigns in the United States, Canada and Australia. Mass Communications—Dissertations. Paper 87.Find this resource:

Sampei, Y., & Aoyagi-Usui, M. (2009). Mass-media coverage, its influence on public awareness of climate-change issues, and implications for Japan’s national campaign to reduce greenhouse gas emissions. Global Environmental Change, 19(2), 203–212.Find this resource:

Schäfer, M. (2012). Online Communication on Climate Change and Climate Politics. A Literature Review. Wiley’s Interdisciplinary Reviews (WIREs): Climate Change, 3(6), 527–543.Find this resource:

Schäfer, M. S., Ivanova, A., & Schmidt, A. (2014). What drives media attention for climate change? Explaining issue attention in Australian, German and Indian print media from 1996 to 2010. International Communication Gazette, 76(2), 152–176.Find this resource:

Schäfer, M. S., & Schlichting, I. (2014). Media representations of climate change: A meta-analysis of the research field. Environmental Communication, 8(2), 142–160.Find this resource:

Scharkow, M. (2013). Thematic content analysis using supervised machine learning: An empirical evaluation using German online news. Quality & Quantity, 47(2), 761–773.Find this resource:

Schmidt, A., Ivanova, A., & Schäfer, M. S. (2013). Media attention for climate change around the world: A comparative analysis of newspaper coverage in 27 countries. Global Environmental Change, 23(5), 1233–1248.Find this resource:

Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication, 50(2), 93–109.Find this resource:

Shanahan, M. (2009). Time to adapt? Media coverage of climate change in nonindustrialised countries. In T. Boyce & J. Lewis (Eds.), Global crises and the media: Vol. 5. Climate change and the media (pp. 145–157). New York, NY: Lang.Find this resource:

Shaw, C. (2013). Choosing a dangerous limit for climate change: Public representations of the decision making process. Global Environmental Change, 23(2), 563–571.Find this resource:

Shehata, A., & Hopmann, D. (2012). Framing climate change: A study of US and Swedish press coverage of global warming. Journalism Studies, 13, 175–192.Find this resource:

Slocum, R. (2004). Polar bears and energy-efficient lightbulbs: Strategies to bring climate change home. Environment and Planning D: Society and Space, 22(3), 413–438.Find this resource:

Smith, N. W., & Joffe, H. (2009). Climate change in the British press: The role of the visual. Journal of Risk Research, 12(5), 647–663.Find this resource:

Smith, P. (2012). Narrating global warming. In J. C. Alexander, R. N. Jacobs, & P. Smith (Eds.), The Oxford handbook of cultural sociology (pp. 745–760). New York: Oxford University Press.Find this resource:

Takahashi, B. (2011). Framing and sources: A study of mass media coverage of climate change in Peru during the V ALCUE. Public Understanding of Science, 20(4), 543–557.Find this resource:

Tolan, S. (2007). Coverage of climate change in Chinese media. Human Development Report Office, Occasional Paper 2007/38. New York: UNDP.Find this resource:

Trumbo, C. (1996). Constructing climate change: Claims and frames in US news coverage of an environmental issue. Public Understanding of Science, 5(3), 269–283.Find this resource:

Ungar, S. (1999). Is strange weather in the air? A study of U.S. national network news coverage of extreme weather events. Climatic Change, 41(2), 133–150.Find this resource:

Urquijo, J., De Stefano, L., & La Calle, A. (2015). Drought and exceptional laws in Spain: The official water discourse. International Environmental Agreements: Politics, Law and Economics, 15(3), 273–292.Find this resource:

Viehöver, W. (2012). Öffentliche Erzählungen und der globale Wandel des Klimas. In M. Arnold, G. Dressel, & W. Viehöver (Eds.), Erzählungen im Öffentlichen (pp. 173–215). Wiesbaden, Germany: VS Verlag für Sozialwissenschaften.Find this resource:

Vihersalo, M. (2008). Framing climate change in Montreal 2005: An environmental justice perspective. In A. Carvalho (Ed.), Communicating climate change: Discourses, mediations and perceptions. Braga, Portugal: Centro de Estudos de Comunicação e Sociedade (CECS) Universidade do Minho.Find this resource:

Wagner, P., & Payne, D. (2015). Trends, frames and discourse networks: Analysing the coverage of climate change in Irish newspapers. Irish Journal of Sociology.

Weiss, G., & Wodak, R. (2007). Critical discourse analysis: Theory and interdisciplinarity. Basingstoke, U.K.: Palgrave Macmillan.Find this resource:

Williams, H. T., McMurray, J. R., Kurz, T., & Hugo Lambert, F. (2015). Network analysis reveals open forums and echo chambers in social media discussions of climate change. Global Environmental Change, 32, 126–138.Find this resource:

Wozniak, A., Lück, J., & Wessler, H. (2014). Frames, stories, and images: The advantages of a multimodal approach in comparative media content research on climate change. Environmental Communication, 9(4), 469–490.Find this resource:

Young, N., & Dugas, E. (2012). Comparing climate change coverage in Canadian English- and French-language print media: Environmental values, media cultures, and the narration of global warming. Canadian Journal of Sociology—Cahiers Canadiens De Sociologie, 37(1), 25–54.Find this resource:

Zannakis, M. (2015). The blending of discourses in Sweden’s “urge to go ahead” in climate politics. International Environmental Agreements: Politics, Law and Economics, 15(2), 217–236.Find this resource:

Zhou, G., Singh, J., Wu, J., Sinha, R., Laurenti, R., & Frostell, B. (2015). Evaluating low-carbon city initiatives from the DPSIR framework perspective. Habitat International, 50, 289–299.Find this resource:

Notes:

(1.) Content analytical methods can also be used for meta-analyses of scientific publications on climate change communication (e.g., Schäfer & Schlichting, 2014).

(2.) Climategate was a controversy that arose when emails from the Climatic Research Unit (CRU) at the University of East Anglia were hacked in 2009. Climate change skeptics argued that the emails revealed that climate change was a scientific conspiracy.