The Role of U.S. Media in Shaping the American
Public’s Definition of Terrorism
MONICA NATES | 11790415
MASTER THESIS | POLITICAL SCIENCE: INTERNATIONAL RELATIONS
SUPERVISOR: DR. SEIKI TANAKA
Abstract
The aim of this thesis is to examine the role of the U.S. media in shaping the American public’s definition of terrorism. In the literature on media effects, agenda setting, framing, and motivated reasoning are the most often cited factors in shaping overall opinions (Craft & Wanta, 2004; Chong & Druckman 2007; Lodge and Taber, 2000; Scheufele & Tewksbury, 2007). However, they do not fully explain why there are differing definitions of terrorism among the public that consumes television news media. In order to investigate what else contributes to the public’s differing perceptions of terrorism, this study examines the role of partisan media outlets (Fox News and MSNBC), and more specifically their selective coverage and labeling of certain violent attacks. This thesis is based on mixed methods: first, a qualitative content analysis of these two networks was done by examining their coverage of terrorist attacks gathered from the Global Terrorism Database (GTD) second, it largely relies on a survey of 200 participants, which was conducted with the use of Amazon’s Mechanical Turk (MTurk). This study linked media consumption to the level of agreement with the preferred networks labeling of terrorism. The findings implicate that the media’s partisan stance on terrorism coverage does impact the public’s definition of terrorism. This study contributes to broader research on the relationship between media effects and public opinion, particularly in regards to terrorism.
1. Introduction ... 1 2. Literature Review ... 4 2.1 Defining Terrorism ... 5 2.2 Agenda Setting ... 6 2.3 Framing ... 8 2.4 Motivated Reasoning ... 9
2.5 Gap in the Literature ... 10
3. Theoretical Framework ... 11
3.1 Competing Theories ... 11
3.2 Argument ... 13
4. Content Analysis ... 17
4.1 Sampling Procedure and Data Overview ... 18
5. Analysis ... 22
5.1 Research Design ... 22
5.2 Population and Sample ... 23
5.3 Methods ... 23 5.3.1 Materials ... 23 5.3.2. Procedure ... 24 5.4 Survey Design ... 24 5.5 Independent Variables ... 25 5.5.1 Scenarios ... 25 5.6 Dependent Variable ... 28
5.6.1 Agreement with News Media Labeling ... 28
5.6.2 Disagreement with News Media Labeling ... 28
5.7 Control Variables ... 28
5.7.1 Age and Gender ... 28
5.7.2 Race ... 28
5.7.3 Level of Education and Employment ... 28
5.7.4 Political party affiliation ... 29
6.2 Survey Results ... 29
6.2 Descriptive characteristics ... 29
6.3 T-Test Analysis ... 30
6.4 Chi-Square Test Analysis ... 30
6.5 Alternative Explanations ... 32
7. Discussion ... 34
7.1 Limitations and Recommendations ... 36
8.0 Conclusion ... 38
Bibliography ... 39
Appendix ... 49
Appendix 1: Survey Sample ... 49
Appendix 2: Survey Responses ... 58
Appendix 3: Participant Analysis ... 60
1. Introduction
In February of 2017 President Trump’s administration made a claim that the media overlooks reporting on some terrorist attacks and released a list of allegedly underreported attacks (Blake, 2017). The perpetrators of these events were overwhelmingly Muslim. While reporters and academics were quick to counter with proof that these attacks were actually extensively covered (Graham, 2017), research does show that the media does not cover some attacks at all and others receive disproportionate coverage (Kearns, Betus, and Lemieux, 2018). The news media, for the purpose of this thesis referring solely to television, decides which types of attacks receive more coverage and the impact of this on a person’s perception of what constitutes a terrorist attack deserves further research. The Trump administration’s comments, in fact, reflect this. A key factor of their labeling an attack as terrorism was the perpetrators were Muslim but in today’s environment the public’s conversation on what constitutes a terrorist attack is evolving to include more elements.
The attacks that are labeled as terrorism need to be more clearly understood because as I will show, television news audiences are divided in how they label attacks as terrorism according to the media they consume. With the increase in mass shootings in recent years and hate crimes, referred to as right-wing terrorism now, drawing more attention, audiences are being exposed to a variety of violent attacks, which has led to the call for a more evolved assessment of these events from the media (DeFoster, 2017; “A Dark and Constant Rage”). The purpose of this quantitative study is to examine the impact of partisan media coverage of violent attacks in the United States in a high choice environment, meaning a new era in which options to watch the news are abundant, on its viewership’s definition of terrorism.
Identifying violent events in the United States as acts of terrorism in today’s fragmented and partisan media climate has become more contentious than ever before. Conflicting coverage on events such as Charleston, South Carolina; San Bernardino, California; Orlando, Florida and several others has sparked many debates on how to classify such violent incidents (DeFoster 2017). The actors and the actions of these events are described in the media in a subjective manner (Huff & Kertzer 2018) because of the considerable leeway the media has in the way it chooses to cover an incident. In the case of Charleston, for example, the perpetrator and his crime were presented in the media in a way that divided the way the public classified this attack.
To certain television news channels it was a mass murder, but to others it met all the criteria of a domestic terror attack. Adding to the confusion of what constitutes a terrorist attack is the fact that the term ‘terrorism’ has never reached a unanimous definition amongst scholars and policy makers (Gibbs 1989; Nacos 2007; Schlesinger 1981; Schmid 1983). The goal of this thesis emerges from this confusion. This thesis aims to understand why there is a division among the American public when labeling violent attacks as terrorism. The research question of this thesis is, therefore, the following:
What effect do partisan media in a high choice environment have on how people identify violent attacks in the United States?
Understanding the role that the media plays in shaping the way we think about terrorism will provide a way to identify violent events in a way that is more consistent. There is a wealth of research on public reaction acting as an instrument that links violent incidents to political consequences, everything from influencing elections to increasing support for right wing parties and changing voting behavior (Bali 2007; Berrebi & Klor 2008; Getmansky & Zeitzoff 2014). Therefore, because terrorism receives widespread media coverage (Nacos, Bloch-Elkon, and Shapiro 2011) it is important to examine the effect media coverage, which provides different narratives, has on a person’s understanding of what constitutes a terroristic event. Gaining insight on the causality of the media, will allow a more accurate measurement of the direct effect the media has on audiences. This will provide leverage on understanding if public opinion can be affected in the short term or if attitudes are shaped over a long period of time.
The media environment that exists today is vastly different than in years before and it is worth exploring the role this new normal has in shaping public opinion. Never has the public had more choice when it comes to television news options, so it is important that we understand what people see and hear. The average U.S. home now receives 189 channels, a record high and a jump from just 2008 when the average was 129 channels (Nielsen 2014). This growth in choice has led to fragmentation within television news media (Prior 2007) and cables news networks are now stratified along ideological lines (Coe et al. 2008; Groseclose & Milyo 2005; Morris 2005; Mullainathan & Shleifer 2005). A high choice media environment, a term used to explain the way in which news is now delivered to viewers, is a new factor that leads individuals to choose ideological media and that leads to the observed variance in how the public defines terrorism. A high choice media environment (HCME) is the situation we now exist in where individuals have
a wealth of information and diversified content and thus vastly different accounts of the same event can be viewed on Fox News and MSNBC for example (Chen, 2012).
These media outlets give opposing perspectives and opinions about why their view is the superior one and why their argument should carry the most weight. However, despite these differences channels like Fox and MSNBC share two important features in common. First, they both offer mostly one-sided arguments that leave no doubt as to where they stand on the issue. Second, programs on these channels offer sharp critiques of the other side (Levendusky 2013). Essentially, in a HCME, partisan outlets such as these provide a biased take on issues and reinforce a message that might make viewers more polarized, since they never hear a balance argument (ibid).
There are several classic media and mass communication effects theories stating that public opinion is responsive to the dominant messages provided by the news media (Chong & Druckman 2007, Iyengar & Kinder 1987, Scheufele & Tewksbury 2007). The basic premise is that in order for a social system to prioritize the problems it faces, a rank among different issues emerges in the form of an agenda. This “agenda-setting” process explains why an issue that is in contention is able to gain more attention than another (Dearing & Rogers, 1996); this thesis will explore the classifications of violent attacks within the partisan media. Determining the agendas of the television news channels will be crucial to answering the research question of this paper because if will prove which types of violent attacks get the most coverage. By proving this, then we will be able to test if public opinion is responsive to the dominant messages provided by the news source a person chooses to watch most often. Furthermore, media attention to various political issues or “framing” certain aspects of an issue makes those aspects more important and gives a particular meaning (Iyengar, p. 11, 2010). Therefore, indexing the way a news channel frames their agenda through a content analysis will be crucial to gaining a further understanding of why the public’s view of terrorism varies in relation to the news source that is most watched.
I am arguing that “motivated reasoning,” which is the level of acceptance given to persuasive political information based on preexisting views (Zaller 1992, Lodge & Taber 2013) along with the divisive agenda-setting and framing of a partisan HCME affects the impact of the television news media on the public’s definition of terrorism. Because the options that have emerged in the HCME are partisan, viewers will take their preexisting views and be able to find ample media choices that align with their political opinions. Therefore, a product of the HCME
is that the person’s definition of terrorism ends up aligning with the media source they most consume and strengthen the existing attitudes they have about terrorism.
A product of agenda setting is the viewer is given what to think about when watching coverage of a violent attack. The average viewer can align their news consumption with their preexisting views by watching one version on MSNBC or another version on Fox News. Increasingly, they are doing just that by accessing the partisan media that is constantly available to them (Iyengar & Hahn 2009; Stroud 2011). As such, television news media has become a reflection of one’s political beliefs and not just a place to gather information. The media can reinforce the public’s preexisting views by having a specialized agenda that frames terrorism in terms that appeal to their audience. Hence media influence on the way the public labels certain types of violent attacks, as terrorism will be uncovered through these mechanisms.
To briefly summarize, I will examine why people define terrorism differently than others and specifically look at the effect of their exposure to a HCME and pre-existing partisan leanings that could align them with a specific television network. My literature review will examine the agenda-setting, framing, and motivated reasoning works and identify how this study will add to the existing terrorism and media research. The following section will be my theoretical framework that argues the way a news channel labels a violent attack is a product of a HCME that now forces the media to employ partisan agenda-setting and framing and directly influences their audiences. In order to prove these claims a qualitative content analysis of Fox News and MSNBC’s coverage and labeling of violent attacks will follow. The results, show these two channels have a similar stance on which attacks are covered, which elements of an attack are most focused on, but are distinct in which attacks are labeled as terrorism. The content analysis informs the survey that tests if these elements do all work to influence how the public defines terrorism. I find that the amount of television consumed is correlated with level of agreement with a media source. However, there is no evidence that Fox News or MSNBC are more capable of shaping the public’s definition of terrorism.
2. Literature Review
The main aspect of this research is to examine the potential influence of the U.S. television news media, specifically how it affects the way people perceive and define violent attacks as either terrorist attacks or not. This chapter reviews the literature on what explains
media effects, mainly agenda-setting and framing, as they go hand in hand, and motivated reasoning. These are all major factors in understanding the link between media exposure and the way terrorism is interpreted.
2.1 Defining Terrorism
The government and policy definitions of terrorism have been generally narrowed down to mean “premeditated, politically motivated violence perpetrated against non-combatant groups by subnational groups or clandestine agents (22 U.S. Code, n.d.). Among scholars, however, defining terrorism is contentious as some find the pursuit of a definition not necessary to study the phenomenon itself while others find an all-inclusive definition crucial (DeFoster, 2017, p. 20). As such, there is no consensus on a definition, which in turn leads to studies such as Huff and Kertzer’s (2018) that look to the public to gain and understanding of how terrorism works. Their work focuses on how members of mass public understand the term. This approach concentrates on the standards that are used to label an attack as terrorism. They find that the public’s understanding of terrorism is influenced just as much by the subjective narrative around an attack as it is by the objective facts on the ground (ibid).
Several studies have been conducted on the factors that shape a person’s perception of a violent event and whether it is defined as terrorism. The perception of threat and fear as it is related to media consumption will be key to explaining that relationship. Research shows that fear of crime is related to overall media consumption, resonance of news reports, and how credible a person believes it to be (Nellis & Savage, 2012). Ferraro (1995) and Warr (1994) suggest that the strongest predicators of fear are gender, age, and race; which coupled with association of victimization and fear, people become disproportionately afraid due to information provided by informal sources, like the media (Lane & Meeker 2003; Skogan 1987; Stafford & Galle 1984).
Studies also find the frequency of media consumption that overemphasizes violence and other extreme events, leads viewers to a distorted worldview that stems more from media consumption than reality (Gerbner, 1969). It is logical then that studies have come to look at the role that media plays in terms of the fear people have about terrorism. The role that race plays as related to terrorism fear and risk is also telling, as it highlights a key divide for and individual’s level of threat and thus their likelihood of defining a violent attack as terrorism. Nellis and Savage (2012) found in their study that Whites rated the likelihood of a future terrorist attack to
be significantly lower than minorities, which is consistent with other findings of minorities’ greater fear of crime than nonminorities (Ferraro 1995, O’Keefe 1984).
Several authors before Nellis and Savage that looked at how media elevates perceptions of risk and fear of crime, found that fear is related to the overall amount of media consumption, resonance of news reports, how much attention a person pays to the news, and how credible they believe it to be (Eschholz 1997; Chiricos et al. 2000; Gerbner 1969). As, Gallup polls suggest that Americans are now afraid of terrorism (Carlson 2004), it is therefore essential to accurately gauge public fear levels that the media has created and what role that possibly plays in defining it. In terms of creating fear of terrorism, the Nellis and Savage (2012) shrewdly point out that overall terrorism is a rare phenomenon that does happen in most neighborhoods, so the public has to rely exclusively on the media for terrorism-related information. Coupled with above-mentioned findings that people intentionally watch news related to topics they seek out, it is not surprising that the authors found a statistically significant relationship between hours spent watching television news, resonance, attention and credibility and fear of terrorism.
2.2 Agenda Setting
Scholarly articles on the effects of media on public opinion can be divided in many ways but generally posit that public opinion is responsive to dominant messages provided by the news media (Chong & Druckman, 2007; Iyengar & Kinder, 1987; Scheufele & Tewksbury, 2007). An agenda-setting process determines the messages that the news media provides. The overall agenda-setting process is made up of the media agenda, the public agenda, and the policy agenda, and how they relate to each other (Dearing & Rogers, 1996, p. 5). Agenda-setting scholars ask how important an issue is in comparison to others within and amongst the individual agendas; this is referred to as salience (Dearing & Rogers, 1996, p. 8). This thesis concentrates on the directional impact of the media agenda on the public agenda, bearing in mind that while the policy agenda is conceptually important, it is responsive to the media and public agendas (Dearing & Rogers, 1996, p.22). Therefore, determining how the media affects the policy agenda is out of the realm of this study.
The amount that people learn about an issue from the media is not conclusive, but it is found that they learn in direct proportion to the emphasis placed on issues by the mass media (Lazarsfeld, Berelson, & Gaudet, 1993). In regards to the agenda-setting function Lang and Lang show that the mass media forces attention to certain issues by presenting and suggesting what
viewers should think about (Lang & Lang, 1966). Today’s partisan media, for example, is shown to engage in biased story selection by reporting more heavily on topics that favor their sides and downplaying stories that harm their point of view (Baum and Groeling, 2008). The landmark study by McCombs and Shaw (1972) tried to show the agenda-setting capacity of the mass media by matching what voters were saying were key issues in presidential campaign with the actual content of the mass media. Their findings showed a strong relationship between the emphases placed on campaign issues by the media and the judgments of voters (ibid). Since, salience is the key to agenda-setting it follows that terrorism, especially since September 11, 2001, carries significant weight.
It is the very nature of terrorism to attract major news coverage; it is a tactic that wants to reach an audience beyond the one directly victimized. The media “inextricably and symbiotically” linked to the mass media, they do not merely provide information, but are increasingly seen as “active agents in the conceptualization of terror events themselves” (DeFoster, 2017, p. 32). Terrorism cannot be separated from the ways the media intends to portray it (Thussu, 1997). These sentiments, when applied to the media’s presentation of violent attacks and the impact it has on shaping its viewership’s definition of terrorism, are plausible. Any dispute that the correlations between media and a person’s perception of terrorism do not exist assumes that people have alternative means of observing violent attacks. The fact remains that the media is still a major primary source of information.
The issue of salience here then becomes about which kinds of violent attacks have the most salience. A 2006 study found that the characteristics that most often indicated which attacks were covered pre-September 11 included, interestingly, incidents that were linked to domestic, rather than international terrorist groups (Chermak & Gruenewald, 2006). After September 11, the round-the-clock coverage changed the fact that American news media and audiences thought about terrorist attacks. The news media was uncritical, formulating their analysis in narrow terms and manipulating preexisting Islamophobic stereotypes (DiMaggio, 2012). The landscape of the media coverage that followed those attacks, just from the amount of coverage informed how much Americans now think about terrorism.
With increased concerns over terror incidents, the American public uniquely conceptualizes mass shootings, which refers to “an event in which multiple people become victims of gun violence in a single incident” (DeFoster, 2017, p.46). There have been 85 mass
shootings in the United States in the past 30 years and 48 of those happed since 2006(Follman, Aronsen, & Pan, 2018). Media coverage of these events has only just begun to intensify in recent years, since between 2011 and 2014 (Peralta, 2016). These mass shootings have a new cultural quality as a result. Mass shootings today are becoming more salient in the mass media because there is the existing terror lens to view them through (DeFoster, 2017, p. 44). Yet the elements of an attack that prompts the public to think of terrorism, such as a bombing, a Muslim perpetrator, or ties to an organization, is not present in the media’s coverage of mass shootings (Huff and Kertzer, 2018). The media gives salience to terrorism and mass shootings in separate agendas and as a result only certain elements prompt what the American public thinks of as terrorism. 2.3 Framing
While agenda-setting focuses on the coverage of an issue, framing is the selection of certain attributes of an issue. The coverage of an issue influences the perceived importance, but it is the attributes linked to the issue in the news media that influences how the issue is thought about (Craft & Wanta, 2004, p. 456). This second level of agenda-setting, so to speak, is regarded as having substantive and affective attributes on issue evaluations (McCombs et al., 1997). Frames give a specific meaning to an issue, take for example a study done in Germany which investigated the positive and negative tone of mass media coverage of the German national election on voters and found that the media’s framing of an issue as positive or negative influenced the public’s perceptions of issue salience (Schoenbach & Semetko, 1992). In this and in other cases, phrasing an issue can alter the meaning that it will have for people.
Knowing now that framing shapes perspectives the implications can be notable. Framing is firmly situated in the relationship between the public and mass media because how the media decides to use it shapes they way the public conceptualizes an issue and can reorient their thinking about an issue as well (Chong & Druckman, 2007, p. 104). The conventional model of framing assumes that an individual can place different emphases on various considerations about a subject (Ajzen & Fishbein 1980, Nelson et al. 1997). In practice, Enelow and Hinich (1984) demonstrated that a voter’s outlook toward a candidate may depend on whether the voter favors the candidate on aspects that are of varying importance. It is then a set of dimensions that affects an individual’s evaluation of an issue.
The examples of research on frames in communication include media biases (Tankard 2001), public understanding, and opinion formation (Berinsky & Kinder, 2006; Huff & Kertzer,
2018). What these works also demonstrate is that framing is understood as a process that evolves. New issues are often just variants of other issues in the news, but they are distinguished by a lack of general agreement about how to construe them, where as older issues have a defined structure (Chong & Druckman, 2007, p. 108). “Traditional” issues, such as violent attacks, can therefore be transformed into “new” issues, like mass shootings and terrorism by reframing. Today the media discourse has shifted in regards to violent attacks. The discussion that comes after a violent attacks often involves a typology for classification.
Huff and Kertzer (2018) explain that they integrate the contentious components that emerge in the wake of violent incidents in order to systemize classification. The first consists of the relatively objective facts on the ground, information about the type and severity of the violence employed, and the incident’s target and location. The second consists of information pertaining to “whom” and “why” questions, concerning identity of the perpetrators and motivation attributed to them. This information is relatively subjective and is unavailable until days or weeks after an incident. They demonstrate the importance of type of tactic and number of casualties and political purposiveness of the actor in shaping the perception of a violent incident. By testing these elements, they find a set of frames that affect debates about violent incidents, but they concede it is not complete and other elements can also play a role. Introducing alternative frames can thus gauge new variances in opinion.
2.4 Motivated Reasoning
Motivated reasoning is now considered to play an important role in shaping opinions in conjunction with agenda-setting and framing. Lodge and Taber (2000, p. 184) define motivated reasoning as “the tendency to evaluate incoming information to support preconception and to devalue contrary evidence.” Individuals who already have established attitudes are more likely to recognize which side of an issue is consistent with their values. Individuals operating in a partisan mode are therefore even more likely to recognize the media agenda and frames consistent with their prior beliefs (ibid). To evaluate agenda-setting and framing effects without a connection to motivated reasoning is to view the process in a vacuum.
Motivated reasoning takes into account the influence of framing is contingent on an individual’s predispositions. Environmental factors, such as the current media agenda’s motivations can produce different responses but audience’s predisposed beliefs cannot be separated. My research will seek to demonstrate that public opinion in the United Sates about
violent attacks is often polarized based on type of information provided to them and that the media's messages contribute to that polarization, something that has yet to be studied.
The literature presented above on agenda-setting and framing had yet to combine those elements of media effects with motivated reasoning. Recently though scholars have begun to link all three, but lack information on how they are delivered. Slothuus and de Vreese (2010) look at how citizens respond to frames when sponsored by political parties. They argue that not only will individuals be more likely to adhere to a frame if it is endorsed by “their” party but their preferences are more noticeable on issues at the center of party conflicts and among the more politically aware. The previous research neglected the impact of the source of issue frames in the real world.
Gollust and Cappella (2014) find that frames, which are compatible with partisan stances, can make politically driven tendencies salient. So, it would stand to say that a Republican audience will engage in a media source that aligns with their views and the same applies to a Democratic audience. Since these audiences are motivated to engage in self-serving practices they will accept favorable arguments with little examination and refute worldview-challenging frames (Taber and Lodge, 2006). Roh et al. (2015) applied how partisan framing when coupled with motivated reasoning affects communication efforts for delivering environmental messages. They found that audience’s predisposed political orientation in relation to environmental beliefs might yield a backlash to such messages. This work, however, does not take into account salience and source of these frames. They also do not consider that those that choose to engage with sources that do not align with their political stances and that are covering issues that would not have salience within another source and still accept those arguments.
2.5 Gap in the Literature
The context that is lacking in this work and in many others is the consideration that in today’s environment the source for the most salient information, especially along partisan lines, is television news media. Further, the framing that most issues receive is given a political treatment, which is easily sought out through motivated reasoning because of the amount of choice in media sources. My research will add to motivated reasoning studies by looking at the impact of politicized views via media sources and how this situational factor determines the kind and strength of effects that this agenda and frames might have on opinion formation.Finally, to my knowledge no research has been done that looks to examine if a particular television news
audience is more likely to agree label acts of violence as terrorism as a result of media effects and motivated reasoning. This thesis will work to add to the research done on how the public understands terrorism but will do so specifically by employing a study of media effects and by doing so additionally contributing to media studies research as well.
3. Theoretical Framework
This chapter introduces the main argument of this thesis. First, this section will give a brief overview of the competing theories within the literature on media effects. It is important to explore the ongoing debate that is the “minimal effects” theory as it actually touches on the importance of developing a theoretical framework for my own argument.
3.1 Competing Theories
Within the literature on media effects and the HCME, two mainstream schools have emerged in the last decades: the minimal effects and oppositional media hostility (Bennett & Iyengar, 2008; Arceneaux, Johnson, & Murphy, 2012). These two schools have fundamentally countered the discourse on both media effects and HCME. This section gives a short summary on these two perspectives.
The earlier era of minimal effects emerged in the 1940s and early 1950s (Klapper, 1960), which was before the context of a mass media system. Scholars theorized that media messages were filtered through a social process that consisted of anything from political parties to churches, unions and service organizations (Putnam, 2000). Bennett and Iyengar find that individuals have only continued to disengage with group-based society and consumer’s ability to choose from a variety of media channels has increased. The end result of this is that individuals can choose the source of their news and the message they hear (2008).
The authors suggest that new research is needed to establish new categories of study like agenda-setting and framing. They propose that existing theories need to be adapted and even thrown out in order to reflect the characteristics of modern audiences and types of subject matter that is produced and information distribution. The lack of innovation when it comes to revising past institutions of research means that a lot of contemporary work has been guided by waning modernist tradition that does not explain the present political experience (ibid).
Media is almost unilaterally considered a convention of governance (Ryfe, 2006) but what minimal effects literature contends is that public confidence in media is at a new low. This is only compounded by the fact that younger generations are less engaged with current affairs and consuming news (Delli Carpini and Keeter, 1997). Yet scholars note that there has been no efforts made to look to the negative effects on viewers of increasingly managed messages and targeted audiences (Blumler and Kavanagh, 1999).
Bennett and Iyengar do make a valid point that the central focus of this field of work is research on electoral communication. They find that evidence in those studies typically only find marginal effects and few scholars have asked if there are other under theorized effects like a lack of trust in politicians, the electoral process and even feeling as if targeted communications are exclusionary (2008). There is also the new dynamic of the competition to produce even minimal media effects in the United States. The new communication channels that are emerging constantly are argued to be less democratic (Howard, 2006). These and other inconsistencies have led the advocates of minimal effects to call for new elements to a new agenda for studying media effects.
In brief, scholars of minimal effects argue, this is a new era, in which media, social structures, and identity formations that affect behaviors are all changing. The fundamental conclusion seems to be: “The increasing level of selective exposure based on partisan preference thus presages a new era of minimal consequences, at least insofar as persuasive effects is concerned” (Bennett and Iyengar, 2008, p. 725). Bennett and Iyengar in particular emphasize the likelihood of attitude change lessens due to the fragmentation in audiences (2008, p.724). Also, they point to future survey-based analyses needing to disentangle the effects of media exposure and political attitudes because in their view the media plays a role in fostering polarization (2008). Ultimately, minimal effects theory posits that media use is nothing more than the sum of pre-existing political attitudes and any impact that can be credited to it directly would be minor.
Another more recent theory is called oppositional media hostility. This theory proposed by Arceneaux, Johnson, and Murphy (2012) explains that in “the contemporary fragmented media environment, partisan news outlets intentionally report political news from ideological perspectives, raising the possibility that ideologically biased news may cause viewers to become increasingly suspicious of and antagonistic toward news media” (2012, p. 174). However, what they suggest is that the fragmented media environment also gives television viewers more
opportunities to disregard news outlets they disagree with as well as just ignoring news altogether and this actually moderates oppositional media hostility.
Whereas media hostility research suggests that people will find bias against their worldview in seemingly neutral reporting, oppositional media hostility finds that people form conclusions about the level of apparent hostility in political reporting. Despite knowing that ideological news coverage is now prolific, people think that ideologically congenial news is reported fairly. Additionally, because of the availability of choices fewer Americans are watching the news and thus direct effects of the news media are weakened. This suggests that media hostility is unlikely a direct result of partisan news media (ibid).
The study conducted by these scholars introduced the element of choice in their laboratory experiment, which was the very factor that allowed their results to demonstrate the overall effect of selective exposure. They proved with their work that even with constrained choices, selective exposure mutes the effects of partisan media. This research establishes that in today’s media environment oppositional information from ideologically identifiable news creates a strong negative reaction (ibid). They use these findings to concede that there should be concern over these effects causing greater division.
This thesis does not engage in the debate between media hostility and minimal news effects theories, as the overall aim of this research is to understand better the direct impact of media on people’s perceptions regarding terrorism. However, it does include the possibility that as shown by the Arceneaux, Johnson, and Murphy study that media fragmentation does have a direct effect on its viewership. And it accepts there is selective exposure and it is based on partisan attitudes, but unlike Bennett and Iyengar this thesis does not assume those factors moderate the effects of the media on its viewers.
3.2 Argument
Due to decades of debate, many contradictions exist regarding the effects of news media exposure and more recently partisan news media exposure in particular. As noted in the previous section, this is partly due to the contradictory nature of media effects, and the debate whether people even choose to watch the news when presented with as many options as there are today (Arceneaux, Johnson, Murphy, 2012). Another reason is that media effects can differ from situation to situation. Scholars focus on different causal mechanisms concerning media effects from, for example, viewer’s predispositions before choosing their news media source, the topic
being covered by the media, or even the ever-evolving media environment. Similarities may emerge in media effects theories, but given the evolving nature of media effects, it is necessary to revisit the topic once more.
The key to this research will be to view the mechanisms that are often seen as the drivers of media influence (agenda-setting, framing, motivated reasoning), through a certain lens of this time. The rise of new media has brought about a debate on fragmentation and selective exposure. Sunstein’s work (2001) is one of the most commonly discussed contributions to this discussion. His work proposes that people’s increasing ability to tailor their political information would have a polarizing influence on democracy because media consumers become less likely to come across information that opposes their partisan attitudes. In this high-choice environment, the strength of the message and media’s verdict, not lack of resources, poses the main obstacle to an objectively informed audience.
I will employ a theoretical perspective established by scholars such as Matthew Levendusky. It is that “unbalanced presentation of the facts on partisan outlets will generate even more marked levels of attitudinal polarization” (2013, p. 1). However, I will apply this specifically to coverage of violent attacks in the U.S. and audience’s perceptions of those incidents as terrorist attacks or something else. With these parameters, this thesis argues that agenda-setting, framing, and motivated reasoning work in conjunction to enhance the television news media’s ability to influence their viewer’s opinions. However, it is the HCME that shapes these mechanisms in a unique way that has not been explored before. The aim of this work is to systematically examine the role of each of these variables regarding media influence on the public’s perception of what constitutes an act of terrorism.
The influence of media is now contingent upon participants’ primary preference of media in what is now a high choice environment. Fifty years ago, U.S. audiences primarily depended on the evening newscasts broadcast by the four networks (Bennett and Iyengar, 2008). Audiences were getting news that was intended to attract the widest possible audience. Prior (2007) explains that the news was mostly non-partisan and because it was on all channels at the same time, anyone who wanted to watch television during the news had no choice but to watch it. News media was not tailored for Republicans for Democrats because it was trying to reach as many people as possible, which meant it stayed separate from partisan disputes.
Something that is true of today’s media environment is the increase in the number of cable channels available to Americans. According to Nielsen (2014) data the average American has access to 189 channels. Included in the average cable package are several 24-hour cable news channels, including Fox News and MSNBC. There is now a proliferation of choice that has fragmented a once uniform broadcast television audience into small narrow audiences. The choices that are available to audiences are now described as partisan outlets, which is to say “opinionated media” (Levendusky, 2013, p. 7).
Since television news became the main source of information, television news media have become responsible for the type of news that is reported (Gurevitch et al., 2009). Stories on these outlets are framed so that a certain agenda is advanced because they are delivered with a point of view that is digestible and consistent with audience predispositions (Jamieson, Hardy, Romer 2007, p. 26). The opinionated media is not only reporting the news but also creating a coherent one-sided version that helps people make sense of the events that are being reported on (Jamieson and Cappella, 2008). When the “other side” is presented it is done so in a manner that makes it easy for viewers to dismiss. Partisan news programs have thus strayed from reporting on facts.
Audiences for these partisan shows have grown, for example Fox News was not widely available as recently as the year 2000, but today even on a day when there is not a huge story to cover, during primetime Fox News can expect more than 2 million viewers (Cassino, 2016, p.2). So dismissing these programs, as having an overall small audience is ill advised because ultimately this is an intensely politically involved audience with many active viewers that will make their opinions heard and shape the discourse surrounding a topic (Jamieson and Cappella, 2008). Partisan media chooses which stories to present to audiences, they then choose how they want to talk about those issues, and finally, they try to attract people who have now been conditioned to go to a particular source to hear the story and the take they want to hear.
Influence can be conceptualized in this work as not only changing attitudes but also formation and attitude reinforcement (Miller 2002). It is important to not just view persuasion on broader terms because for research to accurately measure the effects of influence there needs to be a complete spectrum to measure against. The other goal is to disentangle reciprocal effects of media exposure and political attitudes and behaviors, as suggested by Bennett and Iyengar. This
is because the way that people interact with television news needs to be recognized in order to recognize that media influence is part of a dynamic process.
Polarization is a reflection of this process. Greater polarization is a manifestation of attitude strengthening and reinforcement and this product is a persuasive effect that originates in part from viewing of political media. It is true that motivated reasoning and political ideology lead viewers to watch specific types of media, but it should be recognized that the media plays a role in promoting polarization by being as fragmented and slanted as it is today. While I do recognize that party identification is an important variable for predicting the type of media exposure, I fundamentally argue that engaging with a partisan media outlet, such as MSNBC or Fox News, is more a reflection of the kind of news a person wants to hear and the kind of dialogue surrounding it.
Engaging with a partisan media outlet is a product of motivated reasoning, which in this research is not solely linked to partisan selective exposure. As explained previously, the belief that individuals desire an information environment that is entirely consistent with their political preferences does not always reflect the way people behave. There is evidence that there is an important difference between seeking consistent information and avoiding attitude-discrepant information (Garrett, 2009b). This is because attitude-attitude-discrepant information does not produce a significant decrease in the likelihood of examining a news item (Garrett, 2009a). This distinction is important because although participants of the survey conducted in this study were asked which news outlet they prefer to get their news from, the participants were not divided along party lines, which is consistent with these previous studies.
Agenda-setting and framing models shape the attitude consistent and discrepant information that is available for consumption. Agenda-setting refers to the correlation between the importance that mass media places on certain issues and the importance attributed to those issues by audiences (McCombs and Shaw, 1972). Framing is the assumption that how an issue is characterized in a news report can influence they way it is understood by the audience. By making violent attacks more salient in people’s mind, television news can then shape what people take into account when making judgments about whether an attack is a terrorist attack.
Over the last decade scholars have tracked the frames used to identify trends in issue definitions; thus the second level of agenda-setting, which sets the issue within a broader discussion (Semetko & Valkenburg, 2000). Once the frame is identified, the goal is to
understand how frames in communication affect public opinion, and a specific attitude needs to be isolated. Gamson and Modigliani (1987, p. 144) suggest examining frames produced by various actors on both sides of the issue to set them within a cultural context, which can be complemented by records of individual considerations. It is also necessary to select the sources for content analysis, usually these are mass media sources like, web sites and television broadcasts. The intent of the research determines the specific news outlets, such as comparing specific types of coverage across media.
This thesis claims that the selective news coverage of violent attacks in the partisan media affects the way viewers understand terrorism. This is because channels like Fox and MSNBC will choose to only cover certain attacks and report on them in different ways. According to their agenda they will decide which stories deserve more discussion, especially in cases that do not get around the clock coverage from all television news media. From there they can decide how they want to talk about the attack. Their ultimate agency is in their decision to take a stance themselves and label an attack as terrorism or to label it a hate crime, a murder, a mass shooting, or even just opening the discussion to calling an attack terrorism. This will affect the motivational factors that lead viewers to seek out a certain channel, because they will come to know where they can go to get coverage of a certain type of story and the kind of discussion that will be had. From there, consistent exposure to these sources will begin to influence the way people label violent attacks because as the research shows the public is responsive to a dominant message.
4. Content Analysis
This study looks at the effects of the U.S. media on the American public’s definition of terrorism. The theoretical framework makes the claim that there is a possible relation between the type of media consumed and the definition of terrorism that a viewer has. Therefore, a content analysis was utilized to examine the elements of the television news media that would most likely shape opinions. First, the design and findings of the content analysis are explored, as they directly inform the survey design. The following section will then discuss the design and finding of the survey experiment.
4.1 Sampling Procedure and Data Overview
The content analysis used in this thesis uses human coded content analysis. This was done by watching archival television news coverage on Fox News and MSNBC’s websites and coding each individual statement. Specifically this was qualitative content analysis, so the focus was on the characteristics of the language used and the contextual meaning (Tesch, 1990; Lindkvist, 1981). This approach allows inductive category development and new insights to emerge (Hsieh and Shannon, 2005). Furthermore, this approach builds upon established findings from other studies.
Previous content analyses of the media used transcripts of television programs provided by databases like LexisNexis. With this method researchers can easily find how often a certain term (“terrorism” or “terrorist”) is mentioned by various news sources. This process is automated, so it is quick and for large scale studies can be conducted fairly cheaply, but it is far from faultless. The main problem with this process is it is not adept at determining the context of a statement; the tone of the content is lost when put solely into text. So, while this sort of analysis is able to measure the frequency of certain topics being mentioned by different sources, it is not able to determine the impact of media content on how viewers perceive a story about a violent attack (Cassino, 2016).
With the use of transcript based analysis it has been found that some attacks receive more media coverage than others and have a greater impact on audiences based on the type of information that is relayed. Primarily, the information that has the greatest effect is the type of violence, the severity, and descriptors of the perpetrators, such as religion, an individual versus an organization, and mental illness (Huff and Kertzer, 2018; Kearns et al., 2018). Those same categories were consistent with my content analysis but what emerged that was different from previous works is the media’s choice to take a stance in labeling an attack as terrorism. With no real consensus on a single definition of terrorism, the media then is choosing their own and in doing so is deliberately choosing to cover only certain attacks and framing those attacks in ways that are consistent with that agenda.
The content analysis was conducted with the use of the GTD, an open-source database including information on terrorist events from around the world between 1970 and 2016. The GTD’s definition of terrorism is broken down into three criteria and in order for an act to be considered terrorism is must meet at least two: (1) the act aids a social, political religious, or
economic goal, not including the acquisition of profit; (2) is intentionally trying to coerce, intimidate, or convey a message to an audience that is broader than the immediate victims; and/or (3) is outside the context of legitimate warfare, specifically the parameters of international humanitarian law barring the targeting of civilians and other noncombatants (START, n.d.). The GTD is therefore a useful resource as their definition allows for a wider range of events to be examined.
While the GTD’s definition of terrorism is comprehensive it is not conclusive. This is an important point because the “terrorist” attacks gathered from this database are not unilaterally considered terrorism. The U.S. government agencies may not have found credible evidence to identify them as terrorism and they may not have been prosecuted as terrorism. It is important to note that the majority of the incidents, which were examined, took place after 2008. This is significant because the majority of the incidents that were left to possibly examine occurred during a single administration and as such when the media was reporting on the White House’s assessment of the attack, whether it aligned itself with the administration was a stance in itself on how they view terrorism. Some of these events also may not meet scholarly definitions of terrorism. Lastly, and to the point of this research, the public and media may have also felt they were not acts of terrorism. However, for the purposes of gathering a comprehensive list of events the GTD is a useful resource for gathering examples of varying types of “terrorist” attacks from which to code and examine in the media.
For the purposes of this research, “terrorist” attacks were restricted to events occurring in the U.S. between September 12, 2001 and December 31, 2016, which is the most recent year in the database. This was done in order to narrow down the events that would be looked at and because the media coverage of terrorism post-September 11 reflects a turning point in the way terrorism was represented in the media; terrorism coverage after September 11 develop into spectacle that became accessible for consumption immediately (DeFoster, 2017). The coverage of the attacks was also narrowed to only the first five videos within the first two weeks of the attack happening on each network’s website. This was done because of the time constraints and scope of this study. Additionally, only attacks with fatalities were examined, as it has been shown that attacks with fatalities attract more media coverage (Kearns et al., 2018). This was necessary in order to guarantee a higher chance of finding archival footage on the websites of Fox News and MSNBC. Linked attacks, meaning they were closely related and coverage of them
could not be separated, were also merged into a single incident. 58 incidents were left after merging.
From there each event was assigned a random number and the attacks that were assigned the numbers one through ten were to be examined. This process shed light on the media’s coverage of terrorism because the search had to be expanded beyond the first ten. Several of the attacks from the GTD did not have any coverage on either Fox or MSNBC’s websites. As such, the attacks from the list were examined in numeric order, until ten with at least five videos on the respective websites were found. Trying to find attacks that had coverage on Fox News and MSNBC revealed that, although the networks have distinct differences in how they cover violent attacks, they were similar in the attacks they did choose to report on.
Both networks covered the same attacks, so there was never a situation in which one network presented an attack that the other did not. They either both covered the same attack equally or did not report on it at all. This shows that both networks cover the same types of attacks. In line with studies done on media coverage of terrorism, the attacks that received more extensive coverage were ones that resulted in fatalities and had Muslim perpetrators (Huff and Kertzer, 2018; Kearns et al., 2018). This also revealed that both networks equally disregarded covering attacks that by the GTD’s definition could be terrorist attacks possibly because attacks that are committed by a different type of perpetrator or have different motivations do not qualify as terrorism to them. This is notable because the GTD includes attacks by perpetrators that are anti-Muslim extremists, anti-police, right wing extremists, perpetrators of unknown origins and other categories, yet the media almost uniformly does not give these attacks any or as much news coverage. These attacks consequently become less salient to viewers of both channels because there is a lesser likelihood that audiences will see these attacks covered in the media, therefore they are not as aware that attacks by different types of perpetrators are happening.
It is then necessary to understand how the type of attack that does get covered by the television news media is talked about. Within the parameters of this search, the two-week window played an important role in illuminating a crucial factor of the news media’s coverage of violent attacks. In some of the cases the coverage evolved over time, as more information was uncovered, meaning Fox News and MSNBC’s decision to label an attack as terrorism within the two-week window is telling of how they choose to frame the attacks they choose to cover. Knowing that these events evolve and more information is released over time shows the media’s
conscious decision to label an attack as terrorism within a relatively short period of time. The content analysis reveals that Fox and MSNBC differ in this respect.
Through the content analysis it became apparent that Fox News and MSNBC both equally covered the same types of attacks that they considered terrorism but MSNBC was more likely to mention their considerations of the event not being terrorism until more information became available. Fox News was more likely to label an attack as terrorism more often within the two-week window than MSNBC. Additionally, both networks tended to focus on the particular elements of an attack that the public associates with terrorism, such as an attack being carried out by an organization, even if such information was not confirmed at the time of the reporting. In addition to this, attacks carried out by Muslim perpetrators were often tied to networks, either through online radicalization or through ties to foreign countries thus lessening the likelihood that the incidents would be interpreted as anything other than terrorism by their audience. Attacks by Muslim perpetrators, that could not be tied to a larger organization were overwhelmingly labeled as lone wolves unlike those carried out by white perpetrators that often prompted talk of mental illness. While MSNBC was just as likely to report on the same facts as Fox News, the difference was their intent to not take a stance as often in labeling the attacks. This was measured through the fact that MSNBC was more likely to mention and align itself with the White House or authority’s stance on labeling an attack as terrorism, which often relied on more information becoming available. This method is more likely to instruct viewers to also wait in their labeling of attacks.
For example, with the attack in Garland, Texas, in which two Muslim gunmen opened fire on a Mohammed cartoon contest, MSNBC mentioned more frequently than Fox News that authorities were not confirming ties to a foreign network. In contrast Fox News, mentioned several times that any hesitation to label this attack as terrorism was a failure to “connect the dots.” There was an overall different tone between the two networks. Fox News labeled more incidents as terrorism, irrespective of how recently an attack had occurred, while MSNBC labeled fewer attacks as terrorism in the same time frame. The political motivations of each network are not being debated, it is more examining how often a network labels attacks as terrorism, how soon they make that decision and the factors they use to make those decisions, and what role this plays in shaping the public’s opinion of terrorism.
The more recent attacks, that is to say those since, 2012, that were carried out by a non-Muslim perpetrator did receive greater amount of news coverage. By the GTD’s standards these attacks qualify as terrorist attacks, yet the media is portraying them as either attacks carried out by lone gunmen or are qualified as hate crimes. In the case of the shooting in Wisconsin at a Sikh Temple, in which the gunman was a self-proclaimed white supremacist, MSNBC and Fox News differed in their labeling of the attack. Fox News focused more on descriptors of the gunman’s military background and never took a stance on calling the attack an act of terror because concrete motives had not been identified. MSNBC however, engaged in several conversations about the perpetrator’s motivations, including the fact that he had a history of extremism by attending several “hate events” and having connections to a network of other “skin heads.” MSNBC drew attention to politicians that were calling the attack an act of terrorism and a hate crime. This could also be considered to indicate that Fox News and MSNBC are portraying violent attacks in ways that are distinctly different and thus shaping what audiences take into account when forming their definitions of a terrorist attack.
These are all findings that are consistent with previous studies that index the media’s portrayal of terrorism, but it is the intent of using this kind of information for the purposes of being able to label an attack as an act of terrorism that will be studied through a survey of Fox News and MSNBC viewers. Viewers of Fox News and MSNBC are hearing different takes of the same event and these channels’ choice to label an event as terrorism, when they could or could not be terrorism is worth examining. The media’s choice to only cover certain attacks and to only label certain attacks as terrorism so soon after they occur has proven to be a combination of bias and conjecture and as a large number of people receive information on these events from these sources it will be informative to understand how audiences react to a news source’s discernable judgment on the topic of terrorism.
5. Analysis
5.1 Research Design
This study aims to examine the role of media effects on people’s perceptions of terrorism. This has been done by conducting content analysis of media coverage from a targeted list of violent attacks from the Global Terrorism Database (GTD). With the knowledge gained from
the content analysis a survey was designed and distributed. This study’s purpose is then to contribute to the field of terrorism research and media effects research. Therefore, the objective was to measure how a person’s level of exposure to a certain news channel (MSNBC or Fox News), which has a partisan stance, affects how they view violent attacks and whether they qualify them as an act of terrorism.
5.2 Population and Sample
This study had 200 participants, participants were asked to identify with a range of ages. The female respondents for the Fox News survey were 28% 21-29, 20% 30-39, 15% 40-49, 23% 50-59, and 15% 60 and older. The male respondents were similarly distributed with 22% 21-29, 38% 30-39, 13% 40-49, 15% 50-50, and 12% 60 and older. The female respondents for the MSNBC survey were 38% 21-29, 33% 30-39, 14% 40-49, 12% 50-59, and 2% 60 and older. The male respondents were 2% 17 or younger, 2% 18-20, 25% 21-29, 42% 30-39, 13% 40-49, 13% 50-59, and 3% 60 and older. Among the participants for the Fox News survey 40% were female and 60% male and for the MSNBC survey 40% were female and 60% male.
5.3 Methods 5.3.1 Materials
The survey was fielded using Amazon’s Mechanical Turk (MTurk) in May 2018. Participants were asked to sign a consent form that described the purpose of the study and guaranteed confidentiality and anonymity. The consent also included an email address the participants could write to if they had questions.
Participants were paid $0.50 for their participation. I also limited participation in the study to MTurk workers located in the United sates, who met Master qualifications, meaning they were workers who had demonstrated excellence across a wide range of tasks through statistical monitoring (FAQs, n.d.). In order to control the composition of the sample there were two surveys. I employed a sampling strategy in which I fielded a Fox News version of the survey that was only available to those who self-identified as Fox News being their main news source, and then produced an MSNBC survey for those who self-identified as mainly watching MSNBC as their news source.
The survey was designed via Qualtrics and the data gathered through the surveys was analyzed with the use of the program SPSS. To assemble the graphic representation of the data the program Tableau was also used. No other computer programs were used to analyze the data. 5.3.2. Procedure
Participants were asked to participate in a study on the effects of the media on the public’s definition of terrorism, without emphasizing the focus on the framing and labeling themes. Participants were asked seven standard demographic questions such as, gender, age range, ethnicity, education, employment status, and state they reside in. Afterward, participants were asked to identify how much of their preferred channel they consumed per day, within a range of “a little” to a “great deal.” Then the participants were asked to read four scenarios of terrorist attacks as portrayed by their preferred channel.
After collecting all 200 participants, I converted the surveys to a dataset in SPSS. The dataset was used to analyze the data. Firstly, the internal consistency of the variables was measured, followed by a T-Test analysis.
5.4 Survey Design
This study conducted a survey experiment with a 4x2 design (Figure 1), in which four scenarios were presented to participants. Each scenario was based off of one of the attacks that were examined in the content analysis. The scenarios were selected because they best represented the majority of the types of attacks that were covered by the media and because they included the most often used descriptors of a terrorist attack, this will be elaborated further in the following section. Only minor details were changed or left out so participants would not be able to identify the real event it was representing. The details that were left out include, names of the perpetrators and the state or city the attack occurred in. The numbers of fatalities were changed within a small number, as Huff and Kertzer (2018) show that the effect of increasing casualties on whether an incident is considered to be terrorism is small. We know the use of violence increases the likelihood of an incident being classified as terrorism and bombings increase the likelihood of an attack being classified as terrorism (ibid). Therefore, the tactic of each attack was a changed to a shooting, in order to keep the tactic as a more neutral element for consideration.
The elements that were mentioned with the most frequency during the content analysis were then included in the description of the attack, along with the way the news channel labeled the attack. The survey for Fox News and MSNBC included the same scenarios, as it was shown that the two networks covered the same attacks. It was because of this fact that I could not test different scenarios for viewers of each channel, as that would not be based in reality. The two surveys differed in the way each network labeled the attack, which was consistent with the content analysis findings. After reading each scenario, participants were asked if they “agreed” or “disagreed” with the way Fox News or MSNBC labeled the attack. If the participant agreed they were asked why and given the three options: “I rely on Fox/MSNBC’s verdict,” “I accept Fox/MSNBC’s assessment,” or “Other” which required an explanation of their answer. Choosing “accept” signifies the labeling aligns with the verdict they are most used to hearing. Preferring “rely” signifies a stronger dependence on their media source; this is a clear signifier of counting on on their news source in order to draw a personal conclusion. If the participant disagreed they were asked why and given three options: “Being cautious of making a hasty decision,” “Coverage of the event feels like it is lacking detail,” or “Other” which required an explanation of their answer. Choosing “cautious” shows a process that relies more on other factors than just the news media. Selecting “coverage” means there is an awareness of the media’s choice to portray key elements of an attack and wanting more facts to make a decision. Based on the proposed conceptual model, the expectation is that each scenario would lead to different levels of agreement with the news media source of choice in the context of violent attack coverage.
5.5 Independent Variables 5.5.1 Scenarios
As mentioned, each participant was assigned to one of two surveys, based on their self-identified news channel preference. The Fox News and MSNBC surveys had four scenarios, and each aimed to gauge the level of agreement with their preferred news source based on how the news sources labeled the attack. Below are the four scenarios. The only difference between the surveys is the way the news networks labeled the attack, as such; the way this was presented in both surveys is included. However, only the way the news source for the survey the participants
