Notes on Why Some Terrorist Attacks Receive More Media Attention Than Others

Introduction

• President Trump stated that media neglect to report some terrorist attacks.
• His administration released a list of purportedly underreported attacks with predominantly Muslim perpetrators.
• Reporters and academics dismissed Trump’s claim, demonstrating that these attacks were often covered extensively.
• The study argues that media do not cover some terrorist attacks at all, while others receive disproportionate coverage.

Research Question

• Why do some terrorist attacks receive more media coverage than others?
• Media are naturally drawn to covering conflicts, especially shocking or sensational ones (Tuman, 2010).
• Terrorism is most effective at spreading fear when given widespread media coverage (Powell, 2011).
• Most research focuses on framing and its impact on public opinion (Norris, Kern, & Jost, 2003; Powell, 2011; Ruigrok & van Attevelt, 2007).
• A few studies have focused on the quantity of media coverage rather than the context.
• Incident-level factors can impact the amount of media coverage (Chermak & Gruenewald, 2006; Nacos, 2002; Persson, 2004).
• Perpetrator nationality impacts the amount of media coverage (Weimann and Brosius, 1991).
• These works are largely focused on the pre-9/11, pre-digital media age.
• These studies do not focus on perpetrator religion as a key predictor of coverage in the context of domestic terrorism.
• The amount of coverage increases public awareness and signifies that the event is worthy of public attention.
• Media frames matter but can only have influence if they reach an audience.
• The study addresses two gaps:
  • Factors that explain differences in the quantity of media coverage post-9/11 and in the digital media age.
  • How perpetrator religion impacts these coverage disparities.
• The paper is organized as follows:
  • Literature on media coverage of violence, crime, and terrorism.
  • Methodological approach, sample, and analyses.
  • Results, policy implications, public perception, and avenues for future research.

Media Coverage

• Most of the information we get about the world outside of our local context comes from media.
• Media play a vital role in how we form ideas about people, places, and things which we have not personally experienced (McCombs, 2003).
• Media attention lends legitimacy to the voices and frames that are chosen to be featured (Bekkers, Beunders, Edwards, & Moody, 2011).
• Media coverage also amplifies incidents and ideas by providing a platform to spread certain positions and perspectives to a broader audience (Bekkers, Beunders, Edwards, and Moody, 2011).
• This platform is further expanded by members of the public disseminating media amongst themselves (Nacos, 2002).
• Media coverage of subjects significantly increased online discussion of that topic immediately, and this effect persisted for nearly a week (King, Schneer, and White, 2017).
• People also discuss news media content in various forums, resulting in further – not necessarily accurate – analysis of the information provided.
• The rapid spread of information – regardless of its veracity – is especially common when focusing events occur.
• A focusing event is a sudden, attention-grabbing event that draws public awareness to an issue (Kingdon, 1995).
  • Attention-grabbing and easy to politicize.
  • Relatively uncommon.
  • Reveal a cause of harm or potential harm.
  • Depicted as being particular to certain areas or groups (Kingdon, 1995).
• When something becomes a focusing event, debates and discussions surrounding certain policy topics markedly increase and receive greater media attention (Kingdon, 1995).
• Media coverage does not necessarily determine how we feel about these issues, but it sets the tone for which issues we discuss and how we discuss them (McCombs, 2003).
• Media creates a perspective for viewers that may be incongruent with reality, particularly when discussing an issue that people do not directly experience (Gerbner, 1998).
• Media are primarily responsible for providing information, and thus frames, to the public in the aftermath of a terrorist attack (Altheide, 1987).
• There is clear evidence that media coverage impacts public perception across a host of topics including civic engagement (McCarthy, McPhail, & Smith, 1996), mental health issues (Stack, 2003), and national security threats (Slone, 2000).
• Both news media (Graziano, Schuck, & Martin, 2010; Miller & Davis, 2008; Weitzer & Tuch, 2005) and entertainment media (Callanan & Rosenberger, 2011; Donahue & Miller, 2006; Donovan & Klahm, 2015; Eschholz, Blackwell, Gertz, & Chiricos, 2002; Kearns & Young, 2017) impact the public’s views of crime and justice.
• Media depictions are especially impactful when people do not have direct experience with a topic (Adoni & Mane, 1984).
• Media are primarily responsible for providing information to the public, who use that information to contextualize and understand terrorism.
• When news media spend time on an issue, this suggests to the public that the topic is valid and important for understanding the world around them.
• The amount of attention that a story gets is an indicator of its importance (McCombs, 2003).
• The “CNN effect” – whereby media influence politics and government during conflict and natural disasters – suggests that media framing can impact public opinion and potentially sway policy decisions (Gilboa, 2005).
• Exposure to media coverage of terrorist attacks is positively correlated to perceived personal risk for being victimized, fear of others (Nellis & Savage, 2012), and short-term anxiety levels (Slone, 2000).
• Media are especially impactful at setting public discourse and, as a result, influencing public opinion in regard to limiting or protecting personal freedoms and civil liberties, as they feature and prioritize certain political viewpoints and narratives over others (Guasti & Mansfeldova, 2013; Hall, 2012; Norris, Kern, & Just, 2003).
• Political organizations use media to set the priorities of the public (Chermak, 2003), which means that biases in media reporting can have real-world consequences.
• In short, media coverage influences public opinion and perceptions of the world, which can, in turn, influence how the public perceives relevant people, policies, and groups.

Media Coverage of Violence

• Violent crime has been declining steadily in the United States for the past twenty years.
• Public perceptions of violent crime do not reflect this.
• People still perceive that it is increasing as the violent crime rate in the United States decreases (Gramlich, 2017).
• Media may influence this disparity in perceptions of violence.
• Homicides receive a disproportionate amount of news coverage relative to both the actual risk of being victimized and the frequency of the crime (Paulsen, 2003; Peelo, Francis, Soothill, Pearson, & Ackery, 2004; Sorenson, Manz, & Berk, 1998).
• Violence, broadly construed, is one of the most prominent topics in the news media, and enjoys something of a privileged position, yet it is rare in day-to-day life for much of the audience.
• Media influence increases as actual experience with a problem decreases, which could explain this discrepancy between real and perceived violent crime rates (Slone, 2000).
• Half of Americans are concerned that they or a family member will be the victim of a terrorist attack, despite the actual risk being minuscule (Jones & Cox, 2015).
• Coverage of violence fills that role, while also potentially providing useful information to the viewer.
• An event may be attention-grabbing but lose relevance quickly.
• For a topic to maintain relevance, it must receive ongoing coverage by the media for approximately one to eight weeks (Coleman, McCombs, Shaw, & Weaver, 2009).
• The perceived relevance of an incident fades as time passes without the media referring to it (Coleman et al., 2009).
• Stories are selected for coverage based on how much attention they can potentially attract (Xiang & Sarvary, 2007). This is due to the "infotainment" format of news media.

Media Coverage of Terrorism

• While some terrorist attacks are sensationalized and extensively covered, the majority receive little to no media attention (Chermak & Gruenewald, 2006).
• An issue’s relevance influences the amount of media coverage that it receives (McCarthy et al., 1996).
• Some terrorist attacks may be deemed more relevant than others due to their inherently political, attention-grabbing nature, and potential to be a focusing event.
• Terrorism lends itself to being used as a focusing event, as it is uncommon and can raise awareness of potential weak points in national security.
• Examples of terrorist attacks used as focusing events:
  • Dylann Roof’s terrorist attack sparked fierce debates about the Confederate Flag and gun control policy.
  • Robert Lewis Dear’s attack on a Planned Parenthood facility was used to argue that promoting misleading information could have deadly consequences.
• These attacks are used as focusing events, shifting the public discourse to political topics secondary to terrorism itself and often facilitating or inspiring new policy.
• Brian Jenkins (1974, p. 4) stated that “terrorism is theater,” a metaphor reflecting that perpetrators engage in violence to communicate with an audience.
• Media coverage of attacks amplifies a group’s messaging and sensationalizes the event (Picard, 1993).
• Media and terrorist groups have a mutually reinforcing relationship.
• Media do not cover all terrorism equally.
• Attacks received more coverage if there were casualties, if it was a hijacking, if an airline was targeted, or if domestic groups were involved (Chermak and Gruenewald, 2006).
• Perpetrator identity was not considered as a factor that would impact the amount of coverage an attack receives.
• Even minor attacks may receive coverage if the target, location, or groups involved are of high symbolic or political significance to the public (Nacos, 2002).
• A terrorist attack will receive less coverage if it is framed as a crime (Persson, 2004).
• Whether an attack is framed as terrorism or a crime is complicated by the fact that there is no one accepted definition of terrorism to rely on, even among experts (Schmid, 2013; Spaaij & Hamm, 2015).
• There are myriad potential factors that can impact why a particular terrorist attack receives more news coverage than others.
• The study examines how the following factors influence the amount of news coverage that a given terrorist attack will receive:
  • Who committed the attack
  • What the target was
  • How many people were killed
• Events are more newsworthy if they can be typified as reflecting current beliefs and social structure, and can be scripted in ways that reinforce stereotypes (Lundman, 2003).
• Consistent with the social identity perspective (Tajfel & Turner, 1986), media in the predominantly white, Christian United States may portray members of this in-group in a more favorable way than people who are not members of the majority race or religion.
• Actors portraying terrorists are generally Muslim or Arab, while white actors play the hero (Alsultany, 2012).
• Most Arab movie characters are portrayed as dangerous stereotypes (Shaheen, 2012).
• Perpetrators of terrorism are disproportionately non-white (Gilliam & Iyengar, 2000).
• Media coverage may explain public perceptions of terrorism and identity.
• To Americans, there is an implicit association between terrorism, people of Middle Eastern descent, and Islam (Alsultany, 2012; Gottschalk & Greenberg, 2008; Park, Felix, & Lee, 2007; Saleem & Anderson, 2013).
• Muslims are increasingly viewed as a national security threat (Allouche & Lind, 2010).
• Members of the public are more likely to consider an attack terrorism when the perpetrator is Muslim (Huff and Kertzer, 2017).
• Incidents committed by Muslims were more likely to be labeled as terrorism and were also judged more harshly (West & Lloyd, 2017).
• Muslims were vastly overrepresented in broadcast media coverage of terrorism (Dixon and Williams, 2015).
• News stories about Middle Eastern people often focused on terrorism, asylum seekers, and cultural practices that are alien to Western cultures (Akbarzadeh & Smith, 2005).
• Media may frame terrorism as a specifically Muslim problem because that is a dominant narrative (Sultan, 2016).
• Domestic terrorism is often portrayed as a minor threat committed by mentally ill perpetrators, whereas terrorism influenced by radical interpretation of Islam is framed as a hostile outside force (Powell, 2011).
• If the perpetrators were Muslim and the victims Christian, the innocence and goodness of the victims and their spirituality will often be presented in juxtaposition with Islam (Powell, 2011).
• When the perpetrator(s) of a terrorist attack are members of an out-group or “other,” we should expect to see more media coverage.

Hypothesis 1

• Terrorist attacks will receive more media coverage when the perpetrator is Muslim than when the perpetrator is not Muslim.

Hypothesis 2

• Terrorist attacks will receive more media coverage when the perpetrator is arrested than the perpetrator is not arrested.

Hypothesis 3

• Terrorist attacks will receive more media coverage when the target is a governmental facility or employee(s) than when the target is non-governmental.

Hypothesis 4

• Terrorist attacks will receive increased media coverage as the number of fatalities caused by the attack increases.

Alternative explanations

• White homicide victims receive more media coverage than minority victims (Gruenewald, Chermak, & Pizarro, 2013).
• Attacks against an out-group receive less media coverage.
• Certain dates, such as Hitler’s birthday and the anniversary of 9/11, attract more violence.
• When attacks occur within close proximity to these symbolic dates, they may receive more media coverage.
• We may expect to see less media coverage when responsibility for the attack is unknown (Weimann & Brosius, 1991; Weimann & Winn, 1994).
• We may expect to see more coverage when the individual(s) responsible are connected with a larger group that uses terrorism.
• Classification differences can explain variation in coverage, potentially resulting in ambiguous cases receiving less media attention.

Methods

• Data
  • The data for this study consisted of media coverage for terrorist attacks in the United States between 2006 and 2015.
  • Listed in the GTD.
  • The GTD lists 170 terrorist attacks during this ten-year span.
  • 136 terrorism episodes in the United States during this time.
  • Two sources: LexisNexis Academic and CNN.com.
• To measure media coverage.
  • LexisNexis Academic searches through the full text of thousands of news publications.
  • Limited the search results to newspaper coverage from US-based sources between the date of the attack and the end of 2016.
  • Searches news articles from national sources.
  • Searched CNN.com’s archives to obtain additional news coverage that is solely in digital format.
  • 3541 news articles in the dataset.
• Variables:
  • Dependent variable:
    • The outcome variable for all hypotheses was the number of news stories about the incident.
    • Added the number of relevant articles from LexisNexis Academic and CNN.com to yield the total number of articles for each terrorism series.
    • Estimate models with the total number of articles from major sources only (35.6% of the articles) and with the total number of articles from other sources only (64.4% of the articles).
  • Independent variables:
    • Three binary perpetrator-level variables were coded: perpetrator Muslim, perpetrator arrested, and unknown perpetrator.
    • Three binary target type variables were coded: law enforcement/governmental target, Muslim target, and minority target.
    • Fatalities as the number of people killed – excluding the perpetrator(s) – in each terrorism series.
    • Included a binary indicator to denote whether or not the attack occurred near a symbolically significant event in the United States as a control.
• Results:
  • Negative binomial regression models are most appropriate since the dependent variable is a non-negative count of news articles per attack.
  • Attacks by Muslims receive significantly more coverage than attacks by non-Muslims.
  • If the perpetrator is Muslim, we see 357% more news stories about the attack.
  • 357% more news stories.
  • 287% increase in coverage when the perpetrator is arrested.
  • 211% increase if the target is governmental.
  • 46% increase per fatality, on average.

Discussion

• The study examined quantitative differences in media coverage of terrorist attacks.
• The quantity of articles is important for public perception of terrorism.
• This study is the first post-9/11 and digital media age study focused on the quantity of coverage that terrorist attacks receive.
• This is the first study to explicitly examine how perpetrator religion impacts coverage across such a wide range of terrorism cases.
• By modeling coverage over all terrorist attacks in the United States during a ten-year period, we are able to identify trends in coverage.
• Perpetrator religion matters for the quantity of coverage that an attack receives.
• attacks perpetrated by Muslims receive drastically more media coverage than attacks by non-Muslims.
• This finding is consistent with the literature on social identity (Tajfel & Turner, 1986) that highlights in-group and out-group dynamics whereby people who are perceived as “others” are portrayed and perceived more negatively.
• Media find bias against Muslims and Arabs in the context of entertainment media (Shaheen, 2012).
• Similar biases against Muslims exist in media coverage of terrorism.
• This may explain why people implicitly connect terrorism and Islam and view Muslims as a threat to national security (Allouche & Lind, 2010).
• Each of the other hypotheses was supported.
• When a perpetrator of an attack is arrested we find significantly more coverage.
• Attacks against the government receive more coverage.
• The number of fatalities in a given attack has a significant impact on the extent of coverage.
• Models, the variables testing other counterarguments were not significant.
• Attacks that targeted either Muslims specifically or minorities in general did not receive less media coverage.
• Incidents that occurred near significant dates did not receive more coverage.
• There was no difference in the amount of coverage for attacks connected to a larger group versus those without this connection.
• The results and the robustness of the models demonstrate that media give disproportionate coverage to terrorism when the perpetrator is Muslim, though other factors also matter.
• The identity of a perpetrator as Muslim has primacy as the key driver of the amount of coverage, relative to each of the other factors.

Limitations and Future Directions

• Future research could replicate the project with broadcast coverage.
• Conduct similar analyses in other countries to address concerns with generalizability.
• Some media outlets may selectively cover certain attacks more than others in a way that reflects the ideological perspective of the news organization.

Policy Implications

• The media does not cover some terrorist attacks enough.
• Attacks by Muslim perpetrators received less coverage is unsubstantiated.
• Media do not cover these events equally.
• Attacks perpetrated by Muslim receive a disproportionate amount of media coverage.
• Whether the disproportionate coverage is a conscious decision on the part of journalists or not, this stereotyping reinforces cultural narratives about what and who should be feared.
• These findings help explain why half of Americans fear that they or someone they know will be a victim of terrorism and implicitly link terrorism and Islam (Saleem & Anderson, 2013).
• Reality demonstrates, however, that these fears are misplaced.
• One way to combat misplaced fears about terrorism is to change the public narrative on terrorism to cover attacks more evenly and based on consistently applied criteria.