Appeal to negative emotions in US corporate media. Fox News and its counterparts

Table of contents

1. Introduction

2. Theoretical Groundwork
2.1 Negative emotions and EASI theory
2.2 The roots of (negative) emotions
2.3 The construction of reality

3. Empirical evidence for Fox News’ impact and Hypothesises

4. The networks
4.1 CNN
4.2 Fox News Channel
4.3 MSNBC

5. Methodology

6. Results
6.1 Thematic Codes
6.2 Singular Codes

7. Conclusion

8. References

9. Table of figures

10. Annex

1. Introduction

The question about the role emotions play in making political decisions is one of central importance for democracies and it is alive ever since the very first Greek democratic city states emerged. It is inextricably connected with the degree of rational thinking that we believe our brains are capable of, with the two often being displayed as somewhat antagonistic forces fighting for control over our decision making. The discourse finds its origins in Plato’s claim that reason should always win over passion as the latter would have harmful influence on the functioning of society, causing ill-conceived and narrow-minded decisions for its members (Scherer 1995). After a phase of little interest for this question in which the power of decision makers was given through god and/or tradition and their decisions were therefore not to be judged, the question remerged with American independence and the French revolution. The idea of citizens being able to make rational decisions became one of the main arguments against religion-based paternalism and monarchy as it is an important anthropological presumption for the concept of states based on articles of partnership. This presumption can also be found in the federalist papers and was a central thought upheld by America's founding fathers paving the way for today’s modern democracies (Westen and Hofmann 2012). However, with scientific findings especially from the last thirty years, this idea of man began to crumble, giving a more sophisticated view on the power that emotions have over our decision making. Insightful research such as Kahneman’s prospect theory has debunked the concept of homo economicus and shown that rational decision making is more of an exception and most decisions of people’s everyday life are made unconsciously (Kahneman 2012). These findings have severe consequences for our understanding of how democracies work. In an ideal modern democracy, candidates should try to convince citizens that they have the best concept to solve their problems and that they are most fit to represent them over a period of time, creating an open marketplace of ideas that people can choose from. But following Westen’s argumentation this is not how political discourse works, instead he describes a “marketplace of emotions” which he claims to play a much more important role in democratic elections (Westen and Hofmann 2012). The importance of emotions in politics is no secret to politicians, at most differing in the place value they give to emotional appeals in their speeches and election campaigns. However, politicians are far from being the only sources of public opinion. As for America, which this paper will focus on, the public usually receives their information about current political events through newspapers, radio and television (Shearer 2018). Indeed, even the most basic models in communication science, for example Lasswell’s model of communication (“Who says what in which channel to whom with what effect?” (Bentele, Brosius, and Jarren 2013:182)) are well aware of the fact that reality will always be distorted somehow on its way from the communicator to the audience, may it be intentional or not (Bryson 1948). Having influence on how political events are depicted in the news, combined with a way of communicating that strongly affects peoples believes, gives American news networks tremendous power over the country’s political reality. In this paper I will argue that one of the most effective ways to work on peoples believes is the constant appeal to emotions and the combination of these emotions with certain topics. Television news channels have been, and still are of great interest for researchers regarding the effects that they have on people and politics as they are to this day the main news source for American citizens, even with social media and various other new news sources around (Shearer 2018). Following this interest, I will take a look at the role which emotions play in the language that the three major American cable news channels CNN, MSNBC and Fox News display. Based on Salmela and von Scheve’s work on emotions in right-wing political populism and further research on the persuasive power of emotions, this study will concentrate on Fox News, which stands out both in its audience size and ideological orientation. To do so, a total of 36 show transcripts over the course of one year is examined through a text analysis covering frequent topics on the three networks and how they are connected to certain emotions. Corresponding to previous research and the interest of the hypothesises tested in this paper, the analysis addresses negative emotions exclusively. The goal of this text analysis and paper is to test whether the assumption, that on Fox News both general appeals to negative emotions and the connection of negative emotions with various entities is more frequent than on the other networks, can be empirically supported. A discussion of existing literature, the methodology, expectations and results will be presented in the following.

2. Theoretical Groundwork

2.1 Negative emotions and EASI theory

As for the theoretical groundwork of this paper, thematic framing, negative emotions and their roots hold central roles. According to Ekman’s broadly accepted categorization there are seven basic emotions that universally occur in all cultures and can be measured empirically: anger, fear, surprise, sadness, disgust, contempt and happiness (Ekman and Cordaro 2011). For this paper, anger, contempt and fear will be of predominant interest. This limitation has several reasons, first of all researchers believe “that individuals are motivated to act politically on the basis of what are often negative emotions, including anger, fear, hatred and disgust towards groups or ideas that are discursively marked out as different or "Other." Indeed, in mediated narratives, as in everyday talk, we tend to see a preponderance of negative emotions - such as fear, anger and worry - over more positive ones - such as hope and love. (Wahl-Jorgensen 2019:11).

Secondly, it comes with the nature of the populist right-wing rhetoric that Fox News employs. For this I’d like to quote Salmela and von Scheve referring to Ernesto Laclaus’ work on populist rhetoric:

“emotions and feelings figure centrally in the rhetoric of right-wing populist parties that frame issues like immigration, national culture and employment in terms of emotions like fear, anxiety, anger and feelings of powerlessness. Prospective right-wing voters are known to experience these emotions in these and related contexts and the open appeal to these very emotions is part and parcel of the populism of the new right (Salmela and Scheve 2017:572).

To summarize, negative emotions are not only more effective in motivating political activism but also more salient in both right-wing populist language and subjective experiences of individuals who hold such political views. The reason for why these emotions predominate in prospective right-wing voters can be thought both ways, people might be drawn to right-wing populism because they already feel angry, fearful and powerless or populist rhetoric might be eliciting these emotions first. A balanced answer must not disregard either one of the approaches and this paper will therefore take both into account, even though some literature suggests the latter to be the less important effect (Cassino 2016). However, this doesn’t yet give an answer to why individuals would feel (more) angry or fearful just because they are exposed to rhetoric that employs fear and anger. The emotions as social information (EASI) theory can close this gap. The theory “posits that emotional expressions shape social influence by triggering affective reactions and/or inferential processes in observers” (Harkins et al. 2014:1). Thus, individuals see an emotional expression towards an object by someone else as an information on how they themselves should feel about that object. This then can result in attitude change and social actions by the observer. Van Kleef introduces two moderators influencing the effect that an emotional expression has on the observer. First, the degree of information processing describes “the observer’s motivation and ability to process the information conveyed by these expressions” (Harkins et al. 2014:5) and secondly the perceived appropriateness of an expressed emotion. The latter is determined by social- contextual factors such as cultural norms, the way in which the emotion is expressed and the expressors status (Harkins et al. 2014). Experiments have proven effects on individuals regardless of whether the emotional expression is being presented through written text, audio or facial expressions (van Kleef, van den Berg, and Heerdink 2015). To recap once again, emotions, especially negative ones, affect people’s political beliefs and actions. EASI theory offers a viable explanation for why political agitators appeal to emotions and give them a certain direction. Furthermore, quoted literature claimed that such negative emotional expressions are especially common among voters of right-wing parties and corresponding rhetoric. In the following I will outline common emotional states in western societies and how they can be exploited.

2.2 The roots of (negative) emotions

In their paper “Emotional roots of right wing political-populism” Salmela and von Scheve discuss whether socioeconomic factors can provide satisfactory explanation for the success that right-wing parties had in western democracies throughout the last ten years. In their, but contemporary sociological research in general, socioeconomic explanations are strongly connected with the issues of globalization and neoliberalism. This approach focuses on a number of changes that 21st-century post-industrial global capitalism has brought upon western societies. One of these is the “increased mobility of jobs, capital and people” (Salmela and Scheve 2017:569) which on the one hand made for stronger competition between manufacturing places with many western companies relocating their production sites to third world and developing countries for cheap labour and in turn increased immigration to western countries adding even more pressure on low- and medium-skilled workers. This increased mobility comes together with comprehensive privatization and economic deregulation. Even though the authors focused on Europe when describing these processes their implications on societal issues hold universal validity, only difference being that deregulation and privatization started earlier in the US under the Reagan administration in the 1980s. A time in which only Great Britain under Thatcher witnessed comparable changes in the west (Lütz 2002; Tingle 1988). Next to austerity and the deconstruction of state-led welfare a neoliberal agenda has great implications on a cultural and micro-sociological level, namely the advancement of individualization. According to Baumann, individualization “refers to the way in which identity is transformed from a ‘given’ into a ‘task’, and that individuals are encouraged to take responsibility for this task” (Dawson 2013:19; Bauman 2000). Putting these two factors together means that individuals in modern societies face more insecurity and pressure in their work-life (some groups of course to a greater extent than others) while at the same time facing a culture that puts great emphasis on seeing every individual as solely responsible for his or her career and life. Such a mindset will inevitably sooner or later lead up to a failure to live up to society's and one’s own expectations for at least some individuals (Hamann 2009). The socioeconomic based explanation thus suggests that it is only rational for pressured individuals to hold contempt towards immigrants and other entities who seem to stand in the way of their work-related success. However, a socioeconomic approach that does not go beyond this point, leaves many questions unanswered. First of all, contempt towards immigrants and similar out-groups is not the only logical conclusion from a (perceived) precarious situation in modern society. Just like the common right-wing interpretation of reality, a left-wing interpretation of such a situation is feasible, for example blaming certain neoliberal policies or supranational institutions. It is therefore necessary to dive deeper into the potential roots of emotional states and which psychological mechanisms may lie behind them. Salmela and von Scheve claim that the general ground on which emotions are based on is the same for both left-wing- and right-wing populist interpretations, meaning a mixture of individualized responsibility and precariousness. The important difference being how individuals process their (perceived) precariousness and whether they repress or acknowledge feelings of shame elicited by their situation. The authors also give a possible explanation for why some people either repress or acknowledge their shame. They point to European countries that were strongly affected by the 2008 financial crisis and where shared consequences such as job loss and hardship through austerity policies led to the impression that single individuals cannot be made responsible for their situation. This resulted in the rise of populist left-wing parties in countries such as Greece and Spain while countries that weren’t affected as hard rather saw the strengthening of populist right-wing parties (Salmela and Scheve 2017). This shows that the social reality in which individuals find themselves has strong influence on which ideological interpretation people link to their emotions. But reality of course is not singular and can be constructed in numerous ways.

2.3 The construction of reality

As stated, the socioeconomic approach can offer an explanation for a significant number of individuals expressing beliefs that adhere to right-wing ideology, but it bears a number of problems. Minkenberg got to the heart of it when stating that all “‘objectifying’ interpretations, which postulate a direct relationship between social and economic change and individual political behaviour must be treated carefully. For they neglect the dimension of political mediation and the subjective perception of the problems involved - [...] they overlook the fact that a social and political construction of reality is involved. Various empirical studies show that neither the actual level of unemployment nor the immediate presence or influx of immigrants correlates with a growth in right-wing attitudes or voting behaviour.” (Minkenberg 2000:182)

Furthermore, apart from macro-level indicators Salmela and Scheve point out that “even those who profit from contemporary social and economic developments may perceive themselves as threatened by [...] various out-groups and minorities” (Salmela and Scheve 2017:570). Disregarding whether people are actually threatened or only perceive themselves to be so, it has to be acknowledged that the situation in which individuals find themselves in, is open for ideological interpretation which again is linked to the afore mentioned social construction of reality. Next to more direct forms of social interaction, mass media plays an important part in the construction of such a reality (McQuail 1983). Taking these considerations plus the earlier introduced EASI theory into account and to give a specific example, this means that for an individual’s opinion it is more relevant to see how a news anchor on Fox News expresses his emotion towards immigrants than how the influx of immigrants might actually affect his or her situation in life. Obviously, all forms of media and all television channels construct some version of reality, but regarding the American media landscape, Fox News (FN) plays a special role.

3. Empirical evidence for Fox News’ impact and Hypothesises

I have now outlined potential emotional states of individuals in contemporary societies and how they are affected by socioeconomic factors. This was followed by the assertion that the factors towards which individuals link their emotional states strongly depend on how reality is being constructed around them and which interpretations are available. In the beginning I had already outlined the ongoing importance of televised news in American society and pointed to the extraordinary role that FN plays in it. However, until now I had only mentioned that a significant proportion of Americans watches FN and that tv news in general play an important part in the construction of reality. In the following I will therefore present a number of studies that give empirical evidence for the effect that Fox News has on its audience.

FN first proved to be influential right after its establishment in 1996. The fact that the network was not introduced at once but step by step turned out to be very helpful for researchers covering FN’ early effects. A study that analysed vote shares from 1996 to 2000 in presidential and senate elections was able to show FN’ influence by comparing towns where the network was already available during elections with those where it had not been introduced yet. The towns in which Fox News was being broadcasted saw an increase of 0.4 to 0.7 percentage points in votes for the Republican party compared to towns in which FN was not yet available (DellaVigna and Kaplan 2007). This significant effect was also supported by a more recent study in which the authors could show “that watching the Fox News Channel for [...] additional 2.5 minutes per week increases the vote share of the Republican presidential candidate by 0.3 percentage points among voters [...]" (Martin and Yurukoglu 2017:2566). Considering this effect on people’s voting behaviour they estimate that the Republican party would have had 3.59 (2004) and 6.34 (2008) percentage points less in the presidential elections without Fox News - margins that can easily change an elections’ outcome in the United States. They also tried to find a similar effect for watching MSNBC, however they found the effect to be “an imprecise zero” (Martin and Yurukoglu 2017:2566), which gives even more weight to the question for why Fox News seems to be so persuasive in comparison to other networks. What these two studies show so far is that FN has the power to make people vote for a certain party that otherwise wouldn’t have done so. However, it doesn’t give any information about whether Fox News changed people’s attitudes or just motivated people to go out and vote according to political attitudes they already had in the first place. Thus, following Dan Cassino’s assumption that “ideological media isn't changing the direction of attitudes, though it may be changing their magnitude” (Cassino 2016:7-8), I’d like to point to a study that scrutinizes the effect that Fox News has on attitudes of registered republican voters. For this study more than 3000 US voters were surveyed on their news watching behaviour and their attitudes towards political and cultural topics.

Figure 1: Ideology of Republican voters by FN viewership

Abbildung in dieser Leseprobe nicht enthalten

Figure 1 shows that Republicans who use Fox News as a news source more often self-identify as “very conservative” and more seldom as “moderate” or just “conservative” in comparison to non-Fox News viewers. Regarding specific attitudes it can be summarized that Fox News viewers hold more extreme views (meaning more to the right of the political spectrum) towards all covered topics including among other things gun regulations, wealth taxes and political correctness. Topics that are also often covered on FN (Ray 2019).

But even for individuals who do not hold views that are more or less close to those projected on Fox News, effects could be found. A study that looked at FN’ impact on singular attitudes was able to show effects on democratic voters. The study examined the relation between preferred news sources and attitudes towards Mexican immigrants. More precisely, differences in being exposed to either Fox News or CNN were analysed. The authors came to the conclusion that “watching FOX News was associated with negative perceptions of Mexican immigrants and higher support for restrictive immigration policies” (Gil de Zuniga, Correa, and Valenzuela 2012:610), this effect also showed for “liberal Democrats who [...] had more anti­immigrant attitudes than liberal Democrats who did not [watch Fox News] (Gil de Zuniga, Correa, and Valenzuela 2012:610). In congruence with the earlier quoted study in which no significant effects on people watching MSNBC could be found, the authors where unable to find any relationship between being exposed to CNN and people’s attitudes towards Mexican immigrants, neither positive nor negative (Gil de Zuniga, Correa, and Valenzuela 2012).

As these studies give evidence for the impact that Fox News has on its audience, there is one central question arising: Why is FN so successful in influencing its viewers believes? The answer to this question I assume is closely connected to what was outlined in the beginning of this paper with regard to emotions, for I believe that a central reason for Fox News’ relative persuasiveness lies in the way in which the network embeds emotional expressions in its reporting. Hence, there are three hypothesises that shall be tested empirically in the following:

- Hypothesis 1: Fox News more often covers topics concerning out-groups and minorities compared to other cable news channels
- Hypothesis 2: Fox News more often appeals to negative emotions in its programming compared to other cable news channels
- Hypothesis 3: Fox News more often appeals to negative emotions when covering out­groups compared to other cable news channels

Since there might be some studies that examine the general sentiment of reporting or the quantity of certain topics in American cable news, but none looking specifically at the use of 11 emotional language, I have decided to use data collected on my own in order to test the hypothesises via a qualitative text-analysis of show transcripts from the three major cable news channels. Methodology and results will be presented in the following.

4. The networks

Before proceeding with the results and methodology I will give a firm overview of the three TV-channels that are being examined. For that matter I will shortly present each of their histories, their relevance for American society and their viewership.

Media in the United States underwent several fundamental shifts over the last decades. In his book “Post-Broadcast Democracy” Markus Prior goes back to the 1970s to describe a society in which television was available to virtually everyone but with almost no channels to choose out of. The few available channels were quite homogeneous as they had to appeal to the widest possible audience. This also applied to the news on television which could not afford to strongly lean towards a specific political ideology that might only be suitable for a minority of viewers. Fast forward to the 2000s, the average viewer could choose out of roughly 100 channels of which several provided 24-hour news coverage (Prior 2007). Using more recent data, Cassino states that nearly all of the cable packages that are available today include “at least three 24-hour cable news channels - CNN, Fox News and MSNBC [...]" (Cassino 2016:3). He claims that this diversification of cable news has similar consequences as the increased usage of the internet as a news source, as Americans now have the possibility to choose a channel that fits and reassures them of their political views (Cassino 2016).

4.1 CNN

CNN was founded by Ted Turner in 1980 as the first news-only 24-hour cable channel and remained number one in cable news network ratings until 2002 when it was overtaken by the Fox News Channel (Reagan 2010). Especially throughout the 90s and early 2000s it was extremely influential. Their then unique form of nonstop coverage and its impact on American society and policy makers even earned it its own term in media studies (CNN effect) (Gilboa 2005). However, as other networks began to adopt this model their once distinctive role in the American media landscape shrunk and as of 2018, they are the third news-only cable network in total viewer numbers, behind MSNBC and FN (see Figure 2). Since May 2018 the network belongs to the telecommunication company AT&T (AT&T Inc. 2018).

4.2 Fox News Channel

Fox News was founded by the Australian media mogul Rupert Murdoch in 1996 who put the former republican political adviser Roger Ailes in charge of establishing the network. The network was introduced as a conservative counterweight against a so-called liberal media bias, a business model that proved highly successful with Fox News today by far outpacing its rivals both in revenue and viewer numbers (Dickinson 2011; Cassino 2016). The New York Times described FN’ news coverage during the Bush administration as a “new sort of TV journalism that casts aside traditional notions of objectivity, holds contempt for dissent and eschews the skepticism of government as mainstream journalism’s core.” (Rutenberg 2003). But even though the networks core concept can definitely be described as appealing to Americans that hold political views right of the center it also has to be acknowledged that they were the “first to develop more dynamic audio and visual presentations of the news” (Morris 2005:60) with innovations such as the news ticker, or dramatic visual effects accompanying the news presentation. In total, all these factors made FN become the overall number one news source among republican voters. Hochschild describes the network as an entity standing “next to industry, state government, church, and the regular media as an extra pillar of political culture all its own” (Hochschild and Toren 2017:126; Ray 2019).

Figure 2: 2018 Cable Networks, Total Viewers 2018 Cable Networks, Total Viewers

Abbildung in dieser Leseprobe nicht enthalten

(Moraes 2018)

4.3 MSNBC

MSNBC was brought to life almost simultaneously with FN in July 1996 as a partnership project between Microsoft and NBC. It is now often described as somewhat of a liberal antithesis to the conservative Fox News channel, but this analogy should be treated with caution for various reasons: First, contrary to FN, MSNBC’s political tone was not at all evident throughout its first years. In fact, many conservative hosts that now appear on FN, started or spent periods of their careers at MSNBC (e.g. Laura Ingraham, Tucker Carlson, Ann Coulter et cetera) (Cassino 2016). Secondly, even though “MSNBC is certainly partisan and traffics in outrage and opinion, its reporting—even on its prime-time talk shows—has a much clearer relationship with facts than does coverage on Fox” (Jacob L. Nelson 2019:1).

Recapitulating each network’s ideological bias, one is inclined to think that MSNBC’s ideological direction stems mainly out of an effort to generate as much revenue as possible which made them adopt the FN model after its success on the other side of the political spectrum became evident. For FN, an ideological motivation next to a monetary based, seems more likely taking its strict political course, its personnel and the explicit political activism from its owner Rupert Murdoch into consideration (McSmith 2016). Not to forget CNN, which currently stands a little bit outside of the partisan race between FN and MSNBC. Even though it is seen as a left-leaning network (which is at least true for its audience) CNN has recently put a lot of emphasis on appearing as unbiased as possible and to avoid too one-sided programming (Hashmi et al. 2012). This effort also seems to show in the data collected for this paper. However, I won’t pursue the question of motivation behind bias any further as it is more of an issue of journalistic investigation rather than scientific research and the motive behind the network’s actions are of no relevance as the effect on the audience remains the same.

5. Methodology

The three most popular cable news channels, Fox News, CNN and MSNBC will be examined. While the focus of this paper lies on Fox News, all three channels are being examined to the same extent to guarantee a solid comparability. Out of each network the three most watched shows were picked, they are as follows:

Fox News:

- Sean Hannity
- Tucker Carlson Tonight
- The Ingraham Angle

CNN:

- Cuomo Prime Time
- Anderson Cooper 360
- Erin Burnett Outfront

MSNBC:

- Rachel Maddow Show
- The Last Word with Lawrence O'Donnell
- All in with Chris Hayes (Maglio 2018)

To avoid the data to be biased by a certain topic that was dominating reports during a specific time, four dates throughout the year 2018 were picked via the random calendar date generator. All 36 show transcripts from all nine shows that are being analysed are from the same four dates to further increase comparability. However, on some dates not all nine shows were broadcasted, hence, the missing shows were replaced by the one aired chronologically closest, usually the day before or after¹. As first tries using available coding-dictionaries for sentiment analysis and opinion mining had proven to be too inaccurate, I chose to go through all transcripts sentence by sentence in order to detect and correctly code all expressions that are of interest for the analysis². Transcripts were coded on two levels: First they were separated into what will be called thematic codes in the following. Those give account of which topic dominated over a certain passage of the transcript. They cover at least a few paragraphs, in most cases numerous pages, depending on how often topics switched throughout one show. The selection of topics is on the one hand the result of what turned out to be reoccurring themes throughout the analysis and on the other hand the result of expectations I had beforehand. Two make the categorization definite the code “No specific topic” was added and assigned when none of the other codes were fitting and the content was predominantly apolitical. This would for example be banter or small talk between hosts. Figure 4 shows all thematic codes that were assigned throughout the 36 transcripts together with the percentage they covered on each network.

Not only passages but also single remarks and expressions were coded. Those will be called singular codes in the following. A sentence could be coded multiple times with different codes or not coded at all if none of the singular codes were fitting. Singular codes theoretically could also be used to see which topics were frequent on which network, but thematic codes already sufficiently cover this area. More interestingly, singular codes give insight into relations between topics themselves and relations between topics and emotions. The proximity³ of different singular codes can be analysed and potential differences between networks explored. For example, how likely is the expression of anger close to a reference to the Trump administration? The more often these two codes appear close to each other, the more plausible is a negative bias towards this topic on the network. This approach is of course not without fault, as the simple observation of proximity does not prove that an emotional expression is in fact directed towards that code. However, the more cases of singular codes appearing together are found, the more likely is an intentional connection between the two. As the one-dimensional observation of two codes and their ties can in some cases leave important factors out of the picture, I will also employ so-called code maps which allow a more intuitive interpretation of the results. They can give an idea whether an emotional expression was actually directed towards the code suggested by a first look at the results or maybe rather another entity mentioned within the same paragraph. This can be done by not only looking at two codes (emotional expression and (out-)group), but several codes at once. The map visualizes relations between codes by positioning them according to the frequency with which they appeared together in the transcripts. The more often codes appeared together the closer they are on the map and the thicker the lines that connect them.

Next to emotional expressions, singular codes referring to certain (out)-groups are most important for this analysis. Earlier quoted literature named the following as common out­groups in right-wing discourse which were then established as singular codes in the analysis: Immigrants (Immigration); elites (establishment/elite); the left (democrats/left); the media (mainstream media) and foreigners, in this case Muslims⁴ (Muslims/Islamic terrorism). In order to compare the tone of reporting on FN to MSNBC and CNN, groups that are not typical out-groups in right-wing discourse were also coded. They are the left-wing discourse counterparts to what is covered on Fox News: The right (republicans/right) and conservative media (conservative media), the rest works equally on all networks. Figure 3 shows all singular codes applied throughout the analysis with their frequency. To make the whole coding process more transparent and understandable, a commented screenshot from the analysis program (MAXQDA 2018) was added to the annex which should make all steps described above traceable (App. 7).

Figure/Table 3: Singular codes and their frequency

Abbildung in dieser Leseprobe nicht enthalten

MAXQDA Output; own calculations

I had stated in the beginning that anger, contempt and fear will be the main emotions that are being searched for during the analysis. However, looking at the table above, one will realize that while fear has its own category, anger, contempt and disgust are merged into one. This is due to two reasons: First, classifying emotions will always inherit a certain degree of subjectivity, but while fear and anger are fairly distinguishable, with anger and contempt it showed to be especially difficult. It was often unclear which of the two emotions was to be seen dominant in a sentence, or one could easily interpret it as expressing both. Same goes for disgust, but expressions of disgust in the transcripts were so few (around ten) that I decided to merge them with anger/contempt anyways. Secondly, at least for the hypothesises stated in this paper, I believe a precise differentiation between the three wouldn’t result in any further findings. A final remark regarding the mutual exclusiveness of the employed codes: In general, sentences or parts of sentences were often coded with numerous singular codes which is in par with the methodology as relationships between codes are examined by analysing their proximity. However, there are a few codes that could in some cases describe the same, or be unclear without context, those I approached as follows: The code “Trump administration” overlaps with “republicans/right” thus, if Trump or someone from his administration who also happens to be Republican was mentioned, they were coded with “Trump administration” and if they were explicitly referred to as Republican, but are still a part of the administration, they were assigned both codes. The terms for the codes “conservative media” and “mainstream media” might be misleading. The code “conservative media” was mostly assigned on CNN and MSNBC and in most cases specifically referred to Fox News. Mainstream media on the other hand is a term often used on FN to describe other cable networks, which can be irritating because Fox News as the most watched cable network is objectively speaking part of the mainstream media. However, on Fox News it usually refers to media entities standing left to FN on the political spectrum.

6. Results

6.1 Thematic Codes

Figure/Table 4: Thematic codes coverage per network in percent

Abbildung in dieser Leseprobe nicht enthalten

MAXQDA Output; own calculations

Figure 5: Amount of coverage for thematic codes concerning out-groups, FN vs. MSNBC; one-tailed t-test Fox News CNN

Abbildung in dieser Leseprobe nicht enthalten

Figure 6: Amount of coverage for thematic codes concerning out-groups, FN vs. MSNBC; one-tailed t-test⁵

Abbildung in dieser Leseprobe nicht enthalten

First, I’d like to breakdown the results regarding thematic codes. Those give an idea of what topics networks focused on, thus indicating their biases and priorities. It is therefore also the most fitting method to test Hypothesis 1. In Figure 4 the amount of text that each thematic code covers, is being displayed for each network in percent. Figures 5 and 6 show whether differences between networks were significant for all thematic codes regarding Hypothesis 1.

For example, in Figure 4 the value for “Trump administration” in the CNN line shows 32%. This means that 32% of the text from the CNN transcripts have been identified to be dominated by that topic. Vice versa for all other topics and networks. In some cases, expectations turned out to be right in others they did not. For instance, the thematic codes “climate change” and “education” were set up in the assumption that they would play some part on the supposedly left leaning MSNBC and to some extent CNN. However, the topics barely came up at all in the transcripts, with the only exceptions appearing on Fox News instead. On the other side of the spectrum I had expected numerous segments on Fox News to devote themselves to what they call the “mainstream media” and the “Washington establishment”. Both topics did show up every now and then, but to a far lesser extent than expected and barely ever so consistent that one could call a whole passage dominated by it. Also noticeable is the extremely high share that the Trump Administration and the overall coverage of Trump had on MSNBC and CNN. Both networks dedicated more than 50% of their reporting to the Trump administration, the investigations against him or just Trump (and his family) as a person. Further worth mentioning is the almost non-existent coverage of Republicans and the Trump administration on Fox News. While the latter forms the most frequent topic on both CNN and MSNBC, the two topics together only make up 1% of coverage on Fox News. Most frequently covered on Fox News is the immigration theme instead, followed by the Russia investigation against Trump (equally with democrats/ left) which however mostly circles around supposed wrongdoings of the political left. Regarding Hypothesis 1 the results give a mixed answer: The immigration theme was covered to a far greater extent on Fox News than on the other networks and it is also the only significant difference for both networks (see Figure 5&6). Furthermore, the topics “the media”, “establishment” and “nationality”, all mentioned in literature as common themes in right-wing discourse, exclusively came up on Fox News, however, with only one percent each, they are barley worth mentioning. Also relevant is the share of the topic “Ethnicity/Race” which could also be an indicator for coverage of minorities and out-groups. CNN received the highest count with 14% versus 6% on Fox News. In the t- tests the thematic codes “nationality”, “gender roles” and “ethnicity/race” were added to the “Aspects of identity” code to make a comparison easier. Interestingly the mean value for percentage of text covered by this merged code was higher on CNN than on FN however the difference is not significant. On the contrary “Aspects of identity” covered significantly more text on Fox News compared to MSNBC where it was virtually non-existent (see Figure 6). The reason for this strong discrepancy between CNN and MSNBC in this matter is due to a story on Elizabeth Warren⁶ who at that time made a controversially discussed claim to be of Native American descent which was heavily covered on both Fox and CNN but barely on MSNBC (Gajanan 2018).

In conclusion, the results neither allow for the hypothesis to be rejected nor to be supported with certainty. The numbers on the immigration theme, which probably encompasses the most important of all out-groups in right-wing discourse, speak in favour of the hypothesis, as it shows the biggest difference and is the only significant result on both networks compared with Fox News. All other results can’t be seen as evidence for either direction. I’d therefore hope for future research, maybe a more extensive content analysis of right-wing media, especially to clarify whether immigrants steadily dominate out-group discourse or if maybe, had I picked another timeframe for the analysis, establishment or the mainstream media would have dominated the shows contents instead.⁷

6.2 Singular Codes

Figure/Table 7: Relations between topic and emotion: Fox News vs. CNN; one-tailed t-test Fox News CNN

Abbildung in dieser Leseprobe nicht enthalten

Figure/Table 8: Relations between topic and emotion: Fox News vs. MSNBC; one-tailed t-test

Abbildung in dieser Leseprobe nicht enthalten

Putting the codes “worries/fear” and “anger/contempt/disgust” together, Fox News totals at 464 codes for negative emotions, CNN at 390 and MSNBC at 366. This gap further widens when considering that Fox News’ transcripts made up for the least amount of text (see Figure 4), thus if codes were equally distributed should also make up for the least amount of codes for negative emotions. However, one-tailed t-tests showed that differences were neither significant when comparing Fox News (M =38.67, SD =17.55) with CNN (M =32.5, SD =13.7), t (21) = 0.96, p = .17. nor when comparing Fox News with MSNBC (M =30.5, SD =13.6), t (21) = 1.27, p = .10. Results therefore do not support Hypothesis 2, even though the general trend points into the suggested direction. When looking at the two codes representing anger and fear separately, it turns out that frequencies for “worries/fear” were significantly higher on FN (M =13.75, SD =9.58) compared to CNN (M =5.67, SD =7.02), t (20) = 2.36, p = .01 and especially MSNBC (M =4.83, SD = 4.91), t (16) = 2.87, p < .00. But considering the original aim the results do not allow for the hypothesis to be maintained. In Hypothesis 3 I had stated that out-groups are presented in connection with negative emotions on FN more often than on the other channels. Figure 9 shows the singular codes that appeared close⁸ to an expression of a negative emotion for each network with their frequency (Appendices 1, 2 and 3 in the annex show the exact numbers). Two peaks immediately catch the eye when looking at the graph: Immigration and democrats/left. Both topics were presented with a more than three times higher amount of negative emotions on Fox News than on both MSNBC and CNN, both results are significant (see Fig. 7&8). Significantly higher numbers of negative emotions were also observed in connection with remarks to “establishment/elite”. All other results displayed in the graph below and the tables above were not significant (see Fig. 7,8 & 9).

The only two exceptions where Fox News does not draw in the highest numbers are “conservative media” and “republicans/right” which is in par with the intuitive assumption one would have, as one concerns the ideology of the network and the other concerns the network itself. As virtually no remarks to “conservative media” could be found on FN and vice versa for “mainstream media” on CNN and MSNBC, the two topics were summed up under the term “media” in order to compare the frequencies between networks, but differences were not significant (see Fig. 7&8). Either way the comparison between the two is debateable as the “conservative media” theme misses the populist element of “mainstream media” as the latter basically discredits all media entities except for one. Regarding “republicans/right” scores where higher on CNN (M =19.25, SD=14.91) compared to FN (M =14.25, SD =9.22), t (18) = 0.99 p = .17 but the difference was not significant. MSNBC’s (M =12.08, SD =38.72) score was even lower than FN’, however once again the difference was nonsignificant t (19) = 0.67, p = .25. None of the topics observed showed significantly higher numbers of negative emotional expressions on CNN or MSNBC compared to Fox News. Another interesting takeaway from the data is the fact that negative emotional expressions regarding the left and the right are almost equal on CNN and MSNBC, very much in contrast to Fox News with 607 expressions regarding the left and 171 regarding the right.

Figure 9: Negative emotions by topic and network

Abbildung in dieser Leseprobe nicht enthalten

As already mentioned, proximity does not automatically mean an intended connection between a code and an emotion. As for the graph presented above all results correspond to the intuitive assumptions one would have knowing about the ideological biases the networks have. However, even though the results already support Hypothesis 3, I’d like to visualize results further with the earlier mentioned code maps. Figure 10 shows such a map for all important singular codes⁹ from Fox News’ transcripts (for exact numbers see App. 4).

Noticeable is how negative emotional expressions form one big block with the left, immigration and the right. The right, even though it stands a little bit apart from the other codes already, appears to be the only connection not fitting into the expected FN bias. However, when observing this conjunction, a likely explanation came up: The code “anger/contempt/disgust” is the second most frequent code that “republicans/right” appeared with together (109) but “democrats/left” is by far the most frequent (193).

Abbildung in dieser Leseprobe nicht enthalten

Thus, a logical assumption would be, that when Democrats where being covered overwhelmingly negative, Republicans were then presented as the “good” counterpart. Looking at instances where a combination of these codes appeared, this assumption very often turned out to be true, for example in this statement on “Tucker Carlson Tonight”:

Figure 11: Tucker Carlson Tonight 26 Feb 2018; Screenshot MAXQDA

Abbildung in dieser Leseprobe nicht enthalten

(Carlson 2018)

Interestingly it seems like Fox News not only frames their political targets with negative emotions but also with each other, considering how often the codes “democrats/left” and “immigration” appeared with together (227 times). This is on the one hand a very efficient way of attacking political enemies and on the other hand in part simply owed to the fact that they are the two most frequent codes on FN and therefore will automatically appear together every now and then. Moving on to CNN transcripts, a very different picture shows (Figure 12, for exact numbers App. 5). Trump and his administration very much build the core of news coverage on CNN and as the map shows, it was overwhelmingly negative. The right also appeared together with many expressions of “anger/contempt/disgust” and here a similar structure to that seen on Fox News shows, with again many intersections of the right and left. Immigration on CNN is predominantly connected with the right and the Trump administration which suggests that emotional expressions appearing in the realm of immigration are rather directed towards actions the administration has taken in that context, than towards immigrants themselves as for example on this instance:

Figure 12: Code Map CNN

Abbildung in dieser Leseprobe nicht enthalten

Figure 13: Cuomo Prime Time 20 Dec 2018; Screenshot MAXQDA

Abbildung in dieser Leseprobe nicht enthalten

Furthermore, interesting is the fact that the media theme came up more frequently on CNN compared to Fox News (63 times conservative media versus 49 times mainstream media). However, when it came up, coverage remained far more negative on Fox than on CNN (see Figure 9 and App. 1 & 2).

Finally, Figure 14 shows the code map for MSNBC transcripts. Most striking is the degree to which “Trump administration/Trump” and “anger/contempt/disgust” set themselves apart from the other codes. Trumps persona seems to constitute the core of coverage on MSNBC, even more than it was the case on CNN, while negative emotions are almost exclusively reserved to him and his administration. While “anger/contempt/disgust” are quite frequent emotional expressions on MSNBC, the code “worries/fear” was assigned the fewest on the network and with the exception of “republicans/right” not connected to any of the measured (out-) groups. And even here the number of instances (16) is marginal compared to those observed on CNN and Fox News. Another peculiarity that could already be witnessed on the other two networks, especially CNN is the strong tie between “republicans/right” and “democrats/left” which reaches its purest form on MSNBC. As negative emotional expressions towards the right are only slightly higher than towards the left, one could see their closeness and simultaneous distance to other codes as an indicator for a relatively impartial coverage of political topics on the network. However, this impression does not really uphold when bringing the coverage on the Trump administration back into the picture.

Abbildung in dieser Leseprobe nicht enthalten

Figure 14: Code Map MSNBC

7. Conclusion

As I had stated in the beginning, democracy in part legitimizes itself through the supposition that citizens are capable of rational decision making and will use this capability in a political context. While this assumption was not contested in general, I outlined a relatively easy way to circumvent this capability which is the appeal to emotions. As one of the pillars of mass communication in American society the cable news network Fox News was introduced together with various studies demonstrating the significant effect that Fox News has on people regularly watching it. The goal of the analysis was to find out whether a more frequent appeal to negative emotions on Fox News could be a viable explanation for the extraordinary persuasiveness of the network. Literature especially suggested the framing of out-groups and minorities with negative emotions as an important component in right-wing discourse, thus making up a second focus in the analysis and hypothesises. The first hypothesis laid the ground to answer this question by checking whether out-groups and topics regarding minorities were brought up more often on FN. While this turned out to be true for most topics in the transcripts, immigration was the only theme that was covered significantly more often on Fox News compared to the other networks. The hypothesis was therefore neither specifically rejected nor supported by the data. A similar picture showed with the second hypothesis: Even though the number of negative emotional expressions was slightly higher on Fox, differences were not significant. However, the number of appeals to fear was significantly higher on FN compared to the other channels, pointing to the extraordinary role that appeals to fear play in Fox News’ coverage. Eventually, the third hypothesis outlined how differently out-groups were related to negative emotions on the examined networks. Results showed that negative emotional expressions appeared significantly more often when immigration, the left and the so-called elite were covered on Fox News, thus showing strong support for the hypothesis. Numbers regarding the media and Muslims also showed higher frequencies for Fox News, differences were not significant though.

While the data predominantly looked promising and in favour of the hypothesises it was in some cases unable to deliver clear and significant results. Thus, I’d like to address some peculiarities that this analysis leaves unaddressed and unanswered that might be interesting for future research. First is the time period that the transcripts were taken from which defining element was the Trump presidency. As most negative expressions on MSNBC and CNN are directed towards Trump one must ask himself whether such a negative coverage of a president is normal or rather due to the polarizing nature of this particular president. For this it would be interesting to compare how the Obama and Bush Jr. presidency were covered by each network during their times. Second is the number of transcripts analysed. Considering limited resources and the workload of a manual text analysis, 36 showed to be the most appropriate number of transcripts, but of course results would be more meaningful and, in some cases, might have turned significant with a higher number of analysed transcripts. One last comment concerning the extreme uniformity in the coverage within networks: Each channel had about two or three go to topics which remained consistent throughout the year. Other issues¹⁰ which inherit the problem that they cannot compete with the thrill and flashiness of a raging president or immigrants storming the country didn’t find place in any of the network’s programming. Thus, the distinct sensationalism of American cable news networks is something which future analyses might also want to consider.

Now, what conclusions can be drawn from these results, for American media and politics but also democracies in general? Fox News focuses of coverage in 2018 were immigration, democrats, anger and fear. And while immigration undoubtedly is an important topic for the USA and their citizens it was covered disproportionately and in no way informative but instead exclusively negative and many times in a fear and hate mongering way. Also, the power of Fox News is often dismissed by saying that it may be partisan and favouring republicans in its reporting, but democrats have their own mouthpiece in cable news with MSNBC. Next to the fact that FN has a much bigger viewership, the analysis showed that FN’ reporting is biased to a degree that is in no way comparable to MSNBC let alone CNN. Because FN does not comply with rules of ethical journalism, right-wing populists, conservatives and neoliberalists can communicate their ideas a lot more effectively to the public. To a degree that stands in no relation to the value that these ideas have for the general weal of society as some would argue. If the public asks itself how this very real problem for its political system can be tackled in the future, it should take a look into the past. In 1987 the Reagan administration eliminated the so-called fairness doctrine which had been put in place 1949 by the Federal Communications Commission and among other things was designed to ensure that media platforms present more than one perspective on public issues (Honig 2019). And while it definitely would have to be strongly revised in order to fit the ever-evolving reality of media, it shows that American lawmakers are aware of the problem and it is everything but impossible to find an answer to it.

8. References

AT&T Inc. 2018. AT&T Completes Acquisition of Time Warner Inc. Dallas.

Bauman, Zygmunt. 2000. Liquid Modernity. Hoboken: Wiley.

Bentele, Günter, Hans-Bernd Brosius, and Otfried Jarren. 2013. Lexikon Kommunikations- und Medienwissenschaft. Wiesbaden: Springer Fachmedien Wiesbaden.

Bryson, Lyman (ed.). 1948. The Communication of Ideas: A Series of Addresses. New York: Harper.

Carlson, Tucker. 2018. Tucker Carlson Tonight. Fox News Channel.

Cassino, Dan. 2016. Fox News and American politics. How one channel shapes American politics and society. Routledge studies in political psychology, vol. 3. New York, London: Routledge Taylor & Francis Group. Cuomo, Chris. 2018. Cuomo Prime Time. Warner Media.

Dawson, Matt. 2013. Late Modernity, Individualization and Socialism. An Associational Critique of Neoliberalism. New York: Palgrave Macmillan.

DellaVigna, S., and E. Kaplan. 2007. “The Fox News Effect: Media Bias and Voting”. The Quarterly Journal of Economics 122 (3):1187-234. doi: 10.1162/qjec.122.3.1187.

Dickinson, Tim. 2011. “How Roger Ailes Built the Fox News Fear Factory”. Rolling Stone. 9 June 2011. Retrieved October 29, 2019 (https://www.rollingstone.com/politics/politics-news/how-roger- ailes-built-the-fox-news-fear-factory-244652/).

Ekman, Paul, and Daniel Cordaro. 2011. “What is Meant by Calling Emotions Basic”. Emotion Review 3 (4):364-70. doi: 10.1177/1754073911410740.

Gajanan, Mahita. 2018. “Cherokee Nation Calls Elizabeth Warren's DNA Test 'Inappropriate and Wrong'”. Time. 15 August 2018. Retrieved March 18, 2020 (https://time.com/5425427/cherokee- nation-responds-elizabeth-warrens-dna-test/).

Gil de Züniga, Homero, Teresa Correa, and Sebastian Valenzuela. 2012. “Selective Exposure to Cable News and Immigration in the U.S.: The Relationship Between FOX News, CNN, and Attitudes Toward Mexican Immigrants”. Journal of Broadcasting & Electronic Media 56 (4):597-615. doi: 10.1080/08838151.2012.732138.

Gilboa, Eytan. 2005. “The CNN Effect: The Search for a Communication Theory of International Relations”. Political Communication 22 (1):27-44. doi: 10.1080/10584600590908429.

Hamann, Trent H. 2009. “Neoliberalism, Governmentality, and Ethics”. Foucault Studies:37. doi: 10.22439/fs.v0i0.2471.

Harkins, Stephen G., Kipling D. Williams, Jerry Burger, and Gerben van Kleef. 2014. Emotions as agents of social influence: Insights from Emotions as Social Information (EASI) theory, vol. 1. Oxford University Press.

Hashmi, Ali, Shaina Humphries, Maureen LaForge, and Jean Song. 2012. “Audience Report on CNN”. Retrieved November 17, 2019 (http://web.media.mit.edu/~a_hashmi/portfolio/audience_report_cnn_ali_hashmi_medill.pdf).

Hochschild, Arlie Russell and Suzanne Toren. 2017. Strangers in their own land. Anger and mourning on the American Right. [Ashland, Oregon]: Brilliance Audio.

Honig, Jonathan Andrew Stewart. 2019. “Public Policies on Broadcast and the Fairness Doctrine: History, Effects, and Implications for the Future”. PUBLIC POLICY AND ADMINISTRATION REVIEW 7 (1). doi: 10.15640/ppar.v7n1a1.

Jacob L. Nelson. 2019. “What is Fox News? Researchers want to know”. Columbia Journalism Review (https://www.cjr.org/tow_center/fox-news-partisan-progaganda-research.php).

Kahneman, Daniel. 2012. Thinking, fast and slow. London: Penguin Books.

Lütz, Susanne. 2002. Der Staat und die Globalisierung von Finanzmärkten. Regulative Politik in Deutschland, Großbritannien und den USA. Zugl.: Hagen, Fernuniv., Habil.-Schr., 2001. Schriften des Max-Planck-Instituts für Gesellschaftsforschung Köln, vol. 43. Frankfurt am Main: Campus- Verl.

Maglio, Tony. 2018. “35 Top Cable News Shows of 2018: From ‘Hannity’ to ‘The Situation Room’”. The Wrap. 12 December 2018. Retrieved March 8, 2020 (https://www.thewrap.com/top-cable- news-shows-hannity-maddow/).

Martin, Gregory J., and Ali Yurukoglu. 2017. “Bias in Cable News: Persuasion and Polarization”. American Economic Review 107 (9):2565-99. doi: 10.1257/aer.20160812.

McQuail, Denis. 1983. Mass communication theory. An introduction. London: SAGE Publications Ltd.

McSmith, Andy. 2016. “Iraq and the Rupert Murdoch connection: The media mogul's network of pro­war campaigners”. The Independent. 7 July 2016. Retrieved November 5, 2019 (https://www.independent.co.uk/news/media/press/chilcot-inquiry-report-iraq-war-rupert- murdoch-connection-a7125786.html).

Minkenberg, Michael. 2000. “The Renewal of the Radical Right: Between Modernity and Anti­modernity”. Government and Opposition 35 (2):170-88. doi: 10.1111/1477-7053.00022.

Moraes, Lisa de. 2018. “Fox News Channel Tops Basic Cable In 2018; Sean Hannity Leads Lineup”. Deadline. 12 December 2018. Retrieved October 28, 2019 (https://deadline.com/2018/12/fox- news-channel-basic-cable-top-network-2018-sean-hannity-rachel-maddow-ratings- 1202518853/).

Morris, Jonathan S. 2005. “The Fox News Factor”. Harvard International Journal of Press/Politics 10 (3):56-79. doi: 10.1177/1081180X05279264.

Prior, Markus. 2007. Post-broadcast democracy. How media choice increases inequality in political involvement and polarizes elections. Cambridge studies in public opinion and political psychology. Cambridge: Cambridge University Press.

Ray, John. 2019. “The Fox News Bubble”. Data for Progress. Retrieved November 17, 2019 (https://www.dataforprogress.org/blog/2019/3/23/the-fox-news-bubble).

Reagan, Jilian. 2010. “CHART OF THE DAY: How Fox Conquered Cable News”. Business Insider. 11 January 2010. Retrieved October 29, 2019 (https://www.businessinsider.com/chart-of-the-day- fox-cnn-msnbc-ratings-2010-1?IR=T).

Rutenberg, Jim. 2003. “War or No, News on Cable Already Provides the Drama”. The New York Times. 15 January 2003. Retrieved October 30, 2019 (https://www.nytimes.com/2003/01/15/business/media-business-advertising-war-no-cable- already-provides-drama.html).

Salmela, Mikko, and Christian von Scheve. 2017. “Emotional roots of right-wing political populism”. Social Science Information 56 (4):567-95. doi: 10.1177/0539018417734419.

Scherer, Klaus R. 1995. Plato's legacy: Relationships between cognition, emotion and motivation. Geneva Studies in Emotion and Communication.

Shearer, Elisa. 2018. “Social media outpaces print newspapers in the U.S. as a news source”. Pew Research Center (https://www.pewresearch.org/fact-tank/2018/12/10/social-media-outpaces- print-newspapers-in-the-u-s-as-a-news-source/).

Tingle, Michal. 1988. “Privatization and the Reagan Administration: Ideology and Application”. Yale Law & Policy Review. van Kleef, Gerben A., Helma van den Berg, and Marc W. Heerdink. 2015. “The persuasive power of emotions: Effects of emotional expressions on attitude formation and change”. The Journal of applied psychology 100 (4):1124-42. doi: 10.1037/apl0000003.

Wahl-Jorgensen, Karin. 2019. Emotions, media and politics. Contemporary political communication. Cambridge, UK, Medford, MA: polity.

Westen, Drew and Niklas Hofmann. 2012. Das politische Gehirn, 1st edn. Edition Unseld, vol. 44. Berlin: Suhrkamp.

9. Table of figures

Figure 1: Ideology of Republican voters by FN viewership

Figure 2: 2018 Cable Networks, Total Viewers

Figure/Table 3: Singular codes and their frequency

Figure/Table 4: Thematic codes coverage per network in percent

Figure/Table 5: Amount of coverage for thematic codes concerning out-groups, FN vs. MSNBC; one-tailed t-test

Figure/Table 6: Amount of coverage for thematic codes concerning out-groups, FN vs. MSNBC; one-tailed t-test

Figure/Table 7: Relations between topic and emotion: Fox News vs. CNN; one-tailed t-test

Figure/Table 8: Relations between topic and emotion: Fox News vs. MSNBC; one-tailed t-test

Figure 9: Negative emotions by topic and network

Figure 10: Code Map Fox News

Figure 11: Tucker Carlson Tonight 26 Feb 2018; Screenshot MAXQDA

Figure 12: Code Map CNN

Figure 13: Cuomo Prime Time 20 Dec 2018; Screenshot MAXQDA

Figure 14: Code Map MSNBC

10. Annex

Abbildung in dieser Leseprobe nicht enthalten

Appendix 1: Code relations Fox News: emotional expressions

Abbildung in dieser Leseprobe nicht enthalten

Appendix 2: Code relations CNN: emotional expressions

Abbildung in dieser Leseprobe nicht enthalten

Appendix 3: Code relations MSNBC: emotional expressions

Abbildung in dieser Leseprobe nicht enthalten

Appendix 4: Code relations from Code Map Figure 12 Fox News

Abbildung in dieser Leseprobe nicht enthalten

Appendix 5: Code relations from Code Map Figure 9 CNN

Abbildung in dieser Leseprobe nicht enthalten

Appendix 6: Code relations from Code Map Figure 14 MSNBC

Abbildung in dieser Leseprobe nicht enthalten

Appendix 7: Screenshot from the used data analysis program MAXQDA 2018 Blue numbers display paragraph count

Abbildung in dieser Leseprobe nicht enthalten

[...]

---

¹ Only exception being CNN’s “Cuomo Prime Time” as it was launched step-by-step in 2018, thus ran irregularly. For the two missing dates the two closest available were picked instead.

² I am aware that this approach underlies a certain degree of subjectivity, however I believe it remains the best possibility with the resources at hand.

³ A code is close to another if it appears either two paragraphs before or after the observed code. In the program used for this analysis a paragraph usually encompasses one line. An example is displayed in App. 7.

⁴ Of course, Islam is not a nationality/ethnicity, but none of the networks distinguished between Arabs and Muslims. The only other regularly mentioned foreigners were Hispanics who were covered by the Immigration code due to nature of the debate

⁵ The codes “nationality”, “ethnicity/race” and “gender roles” are sub codes to “aspects of identity”. They have been listed each to provide further insight, however in the rest of the analysis they will be treated as one under the term “aspects of identity”

⁶ A likely presidential candidate for the democratic party in 2018.

⁷ Remarks to mainstream media on FN were compared with remarks to conservative media on CNN and MSNBC.

⁸ Within two paragraphs before or after

⁹ At least twenty appearances in total and at least ten conjunctions with at least one other code were set as a threshold for all code maps

¹⁰ such as the American health care crisis or climate change