Mylynn Felt, Ph.D. student: University of Calgary
Media coverage of high profile teen suicides linked to computer-mediated peer harassment in the early 2010’s has generated sufficient public discourse to establish cyberbullying as a Canadian social problem. This discourse led to provincial legislative changes and culminated with the 2014 federal Protecting Canadians from Online Crime Act. In order to analyze the mediated public discourse that led to these changes, I applied framing theory to a content analysis of print news coverage of the deaths of Jamie Hubley, Amanda Todd, Rehtaeh Parsons, and Todd Loik. Analysis reveals a dominant frame of cyberbullying as a social problem. This frame emphasizes female over male victims. Results also show that the teens were reified as victims to forward a cause. Policy changes reflect this social problem construction despite the reality that these deaths represent the most extreme cases and are inconsistent with what the literature describes as more common forms of cyberbullying.
Mylynn Felt, MA, M.Ed., Ph.D. student
Department of Communication, Media & Film
University of Calgary
SS 320, 2500 University Dr. NW
Calgary, AB T2N 1N4
The headlines are familiar. “Death by a thousand clicks” (Toronto Star 13 April 2013). “Teenager’s death is leading to an outpouring of concern about cyberbullying, and justice ministers are debating what to do” (Globe and Mail 24 April 2013). “Justice Ministers urge Ottawa to tackle cyberbullying” (Globe and Mail 19 April 2013). “Lessons from teen tragedy” (Calgary Herald 13 Oct. 2012). “The problem with ‘cyberbullying’” (Globe and Mail 19 Oct. 2013). “New ‘holistic’ legislation coming in the fall to stop cyberbullying” (CP Wire 26 Sept. 2013). The mediated discourse of several high-profile teen suicides have recently aggregated to form a new social problem on the heels of an established concern for schoolyards.
How does Canadian media frame the concept of “cyberbullying”? By examining national representations of suicides preceded by computer-mediated harassment, I examined the methods journalists use to construct cyberbullying as a social problem. Where do media representations tend to attribute blame? Do the reports trivialize criminal behaviour in youth by suggesting a technologically deterministic stance? Does representation of cyberbullying as a problem remain consistent across coverage of different victims? To answer these questions, I conducted a content analysis of main-stream public discourse on cyberbullying. My analysis reveals a ‘social problems frame’ in the news coverage of high profile teen suicides attributed to cyberbullying. The social problem frame reifies complex cases to simplistic representations. This aids in establishing a pattern aimed toward public attention and new legislation as solutions to the problem.
The term “cyberbullying” is increasingly used by politicians as well as reporters, educators, parents, and students. At least nine provinces in Canada have recently considered new legislation aimed at preventing cyberbullying (Panjvani, 2013). National legislation, in the form of Bill C-13: Protecting Canadians from Online Crime Act, recently received Royal Assent and became law in March 2015. Implied with the application of this term is the belief that the nature of harassment as mediated through technology differs significantly from non-mediated verbal or physical attacks.
Cyberbullying literature often focuses on measuring, diagnosing, and responding to the issue. What is under-represented in cyberbullying literature is an analysis of how disparate circumstances are grouped through public discourse to construct a commonly recognized problem. This process of construction directly impacts all praxis related to cyberbullying. The construction of the problem also directs the manner of research undertaken in the name of understanding and preventing the problem. Using framing theory and content analysis, this study examines public discourse of cyberbullying as represented in print news coverage of the deaths of four Canadian teens.
Cyberbullying Becomes a Canadian Problem
It was the suicides of Amanda Todd in 2012 and Rehtaeh Parsons in 2013 that made cyberbullying a Canadian social problem. In Todd’s case, an unknown perpetrator convinced her to lift her shirt for the webcam as he chatted with her through social media. He captured the image and threatened to expose the photo to her peers if she did not provide further explicit material, which she did. This online harassment continued for years. Ultimately, he started a Facebook page with that image as the profile photo. Todd moved several times but at each school, she found the perpetrator had contacted teens at the new school and forwarded the image as well as insulting comments. It wasn’t until 2014 that Dutch police identified her perpetrator, Aydin Coban, and arrested him for exploiting and extorting victims in the Netherlands, UK, and Canada. He reportedly had male and female child and adult victims. Part of what made Todd’s death an international news story was a nine minute video she posted to YouTube prior to her death. In the video, she holds written cards in front of the camera, detailing the harassment she experienced from her unknown cyberbully as well as other peer conflicts. As of December, 2015, , 2014, the video had been viewed more than eleven million times.
In Parsons’ case, it was the photo of her severely intoxicated, mostly unconscious and being sexually violated by a male while he gestured a thumbs up to the camera that provided the main source of her taunting. The image was sent to her peer group, leading to many derogatory comments online and in person. At least four males were involved in the incident. Although police decided to not make any initial arrests, public outcry after her death led to a reversal of that decision. Two teens were arrested. One pled guilty to producing child pornography, and the other pled guilty to distributing child pornography (CBC 15 Jan. 2015). Each was sentenced to a year of probation.
These cases led to local legislative changes such as the Nova Scotia Cybersafety Act enacted in May 2013 and the New Brunswick Education Act, which was amended by Bill 45 in 2012. The Department of Justice Canada provides an overview of cyberbullying legislative movements inspired by high profile cases. The footnote to this comment lists both Amanda Todd and Rehtaeh Parsons (Ha, 2014).
Female teens are not the only ones to experience what has been called cyberbullying. On 15 Oct. 2011, 15-year-old Jamie Hubley died by suicide. Initially, his death was covered by Ontario newspapers because of the political profile of his father, Kanata South Councillor Allan Hubley. Within days of the first simple announcements of his passing; however, Hubley’s death became an on-going story. As the only openly gay student of his high school, Hubley struggled to fit in and blogged about his experiences on Tumblr. He experienced ongoing harassment and, despite showing an awareness of the viral “It Gets Better” campaign, he posted that he struggled to see how he could survive the years ahead until his life would get better.
In the wake of Todd and Parson’s deaths, a mother in Saskatchewan approached reporters claiming that her 15-year-old son, Todd Loik, who also died by suicide, was driven to his death due to online peer harassment. She discovered the messages his peers sent after his death. The details of this harassment are limited in news coverage. His mother says that the comments were vile and ongoing. She shares little detail, noting that the statements are too cruel to repeat. By the time of Loik’s passing, public discourse was solidly framing what cyberbullying meant and focussing attention on how government policy should handle it.
Anti-cyberbullying attention moved from provincial to national responses. Shortly after Hubley’s death, the Standing Senate Committee on Human Rights received an assignment to research cyberbullying in November 2011. This assignment stemmed from Canada’s obligations under the United Nations Convention on the Rights of the Child (Ha, 2014). In December 2012, they issued a report titled “Cyberbullying Hurts: Respect for Rights in the Digital Age.” In June, 2013, the Coordinating Committee of Senior Officials Cybercrime Working Group issued a report to the FPT Ministers responsible for Justice and Public Safety. The report was titled “Cyberbullying and the Non-consensual Distribution of Intimate Images.” It focused on looking for gaps in the criminal code. In response to this report and public pressure regarding cyberbullying, legislators proposed Bill C-13, the Protecting Canadians from Online Crime Act, in the fall of 2013. This legislation received Royal Assent in November 2014.
The distinction between what is currently considered cyberbullying and what is often termed bullying is socially constructed. The meaning of each term is dynamic and evolves in a manner reflecting common use. While attention to the issue has increased internationally, this change has been largely driven by public discourse generated from news coverage of teen deaths. This is certainly the case in Canada. In order to analyze the construction of cyberbullying, I conducted a frame analysis of media coverage of four teen suicides associated with peer computer-mediated harassment. This research focuses on news coverage of the deaths of Jamie Hubley, Amanda Todd, Rehtaeh Parsons, and Todd Loik.
In order to apply a constructionist approach rather than the more common a priori stance taken in the literature, I use framing theory to position my analysis of news media treatment of cyberbullying. Framing theory is one of the more mature and fully developed concepts in communication studies. It works well with social problems research because of its emphasis on the style of presentation claims makers adopt. Frame analysis is important because it reveals the subtle beliefs that affect message delivery and prime an audience to endorse the ideology promoted by claims makers of social problems.
Applications of framing theory vary in their definition of the terms frame and framing. This is one of the criticisms of this theory. However, many researchers rely on Entman’s (1993) definition.
“Framing essentially involves selection and salience. To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (p. 52).
Framing is an interpretation and presentation of reality. For Goffman, to frame is to “locate, perceive, identify, and label” (1974, p. 21). Kitzinger (2007) compares framing to photography. To reconstruct a scene, a photographer makes choices about what details remain in focus, which are out of view, and which are left in the background, as well as how fuzzy those background details look. In this manner, a photographer “frames a particular view” (Kitzinger, 2007, p. 134). This metaphor supports Reese’s claim that “Frames organize and structure, and thus are bigger than topics” (2010, p. 18). In the photography metaphor, an analyst might describe the subject of the photo as the topic while the choices about focus, detail, and angle of the shot would all be worth exploring as aspects of the frame. As Kitzinger observes, the mere act of explaining framing theory involves crafting and applying a frame. It is about choices. Frame analysis, then, involves identifying the choices of claims makers as well as the effects of those choices. Entman describes framing as “the process of culling a few elements of perceived reality and assembling a narrative that highlights connections among them to promote a particular interpretation” (2010, p. 336 and 2007, p. 164). Applying framing theory to the analysis of news text, then, requires a close examination of the assembled narrative. Referring back to Entman’s earlier definition of framing, such analysis would consider how claims makers define problems, diagnose causes, make moral judgments, and suggest remedies (1993).
Baden and Lecheler (2012) developed a knowledge-based model to explain why some framing effects persisted longer than others. Their research is predicated on a concept they borrowed from Price and Tweksbury. “The persistence of framing effects depends on whether frames ‘can help determine what knowledge is activated – and once activated, presumably used – when people are called on to make politically relevant judgments” (Baden & Lecheler, 2012, p. 360). They determined that “durable faming effects” occur whenever the respondent encounters something novel or whenever some sort of learning takes place (p. 374). When respondents encounter frames for long-standing position issues, only temporary framing effects occur. Similarly, when people already have a dense knowledge base on an issue or when they are primarily ignorant on the topic, framing effects do not endure for long. This occurs because frames do not appear novel to the more knowledgeable recipients. For the ignorant ones, there is no similar knowledge base for the frames to attach to and become enduring. Therefore, Baden and Lecheler find that “Lasting framing effects should be most prominent among medium-knowledgeable people” (p. 375). The final tenet of their model claims that “successive effects of competing frames interact only to the degree that the first frame is stored and still sufficiently accessible at the time of the second exposure to be retrieved as context for the second frame” (p. 375). This finding recognizes some of the effects of competing frames.
Entman (2010) also suggests that exposure to competing frames does not reduce framing effects but complicates them (p. 332). Baden and Lecheler (2012) also note that frames not only “manipulate belief importance” but also convey new information (p. 376). Again, these findings rest on a psychological or cognitive approach to framing studies. Frames affect thinking most often when respondents have a certain base of knowledge to connect with. This knowledge base is culturally situated. Without an existing set of beliefs, what others might call schemata, frames have no foundation to form lasting impressions. To the other extreme, when frames characterize an issue people already know much about or have deep set opinions in place, a new frame is not necessary and so has little enduring effect.
This all ties back to phenomenology. Humans have limited sensory capacity for interpreting information. One must categorize and store experiences in frames that allow for quick judgments. Were this not done, it would be impossible to operationalize complex data. One must be able to form quick impressions of likely scenarios. When entering a new room, there is an abundance of sensory information to process. This limits the capacity to notice fine details. When one enters a room multiple times and is familiar with it, one is more primed to recognize small changes and details. Frames provide a sense of predictability for recurring situations. Once a first-impression frame is reinforced, it becomes a predictable set to rely on when facing a similar situation. Being able to call on the pre-formed frame frees one’s senses to attend to other processes. That is why new frames have little lasting effect for people with already formulated frames on those issues. The new frame is not necessary for making sense of the situation. It is also why new frames struggle to provide lasting effect for those with no existing frame to process a situation. A completely new issue or topic requires so much sensory processing, frames then need repetition in order to endure. People presented with a consistent frame applied to multiple early exposures of a new issue will likely adopt that frame as their own. Chong and Druckman (2007) note that “framing effects depend on a mix of factors including the strength and repetition of the frame, the competitive environment, and individual motivations” (p. 111). They also support the notion that when an issue is new to the social problems market place, people are unsure how competing frames might align with their existing values (p. 113).
Many scholars have noted the importance of culture when considering frame resonance. Reese claims that frames are embedded in a web of culture – a “historically rooted but dynamic cultural context” (2010, p. 18). Borah notes that “Individuals use a set of available beliefs stored in memory” (Borah, 2001, p. 252; see also Chong & Druckman, 2007, p. 111) and suggests, like Baden and Lecheler, that ambivalence is key. Bruggemann (2014), too recognizes that “the individual is always nested within different contexts” and that frames grow out of culture which “manifests itself at the individual, organizational, and social level” (p. 67). He claims that journalists draw from a frame repository in order to craft their stories.
Some might question why an analysis of news frames matters at all. Most social scientists reject the notion that news media providers possess a hypodermic needle to inject thoughts in readers or viewers—that audiences openly receive the messages fed to them by these providers. Nevertheless, this view shifted when Bernard Cohen famously claimed that the press “may not be very successful in telling its readers what to think, but it is stunningly successful in telling its readers what to think about” (1963, p. 13). Entman (2007) criticizes Cohen; “The distinction misleads because, short of physical coercion, all influence over ‘what people think’ derives from telling them ‘what to think about.’ If the media really are stunningly successful in telling people what to think about, they must also exert significant influence over what they think” (2007, p. 165). Kuypers (2002) adds that “the press not only tells us what to think about (agenda-setting), but it also tells us how to think about it” (p. 198, emphasis in original). He suggests that priming and framing constitute the component of “how.” Framing theory does not presuppose that the frames identified in a specific news piece directly translate to the views of consumers. Entman (2012), for example, applies what he calls a cascading model to his frame analysis of political scandal. This cascade model recognizes the influence of strategic actors (politicians and interest groups), journalists, and members of the public as well as the flow of influence between these actors (p. 7). Elsewhere, he calls it a cascading network activation model, saying that “perceived public opinion can and does feedback to influence the future framing behavior of elites and journalists” (Entman, 2010, p. 333). He notes that journalists, social scientists, film directors, and in fact, everyone, frames. Therefore, news frames are a fertile ground for analysis. While public opinion is not entirely shaped by journalists, they certainly exert influence.
The presence of frames in news writing does not necessarily suggest bias. One of the foundational tenets in journalistic writing is objectivism. Another is balance. Reporters write with the intention of fairly representing all sides of an issue. They also aim to remove any evidence of their own opinion in their writing. Instead, a journalist relies on the quotes of his or her sources in order to express any opinion or value statement present in his or her writing. A news article may contain all these elements and still convey certain frames. That is due to the essential nature of frames. “Even the most ‘objective’ or ‘neutral’ journalism will inevitably contribute to the social construction of reality” (Bruggemann, 2014, p. 65). Journalists must make decisions about which facts are more relevant and which sources are more valid.
I am interested in examining the dominant frames present in media representations of the deaths attributed to computer-mediated harassment. I examine the implications of the “cyberbullying” frames and consider the cultural relevance of those frames. Once salient, frames are resistant to change. Nevertheless, even a position issue as entrenched as global warming can experience a frame change. Many social actors now apply the competing frame of climate change. Since, as Van Gorp (2007) stated, frames are socially constructed in such a subtle manner that most members of the public never question a frame or consider its nature, a careful examination of the relatively new collective action frame of cyberbullying is warranted.
As I prepared my data set, it was important to balance my case studies between males and females. I selected news coverage of two male and two female Canadian teens whose suicides were associated with computer mediated harassment. Using Factiva, I ran a key word search applying both the first and last name of the victim to a date range framed between the date the teen died and the date when government officials announced proposed legislation or policy changes as a result of the suicide. I also limited the search to Canadian publications.
Jamie Hubley, a 15-year-old from Ontario, died 15 Oct. 2011. On 30 Nov. 2011, officials announced the Ottawa Accepting Schools Act. This date range produced 154 unique articles, once duplications were eliminated from the set. Amanda Todd, a 15-year-old from British Columbia, died 10 Oct. 2012. Just five days later, on 15 Oct. 2012, officials announced a House of Commons Motion to study cyberbullying. While the motion was already in process, media coverage following her death purportedly caused an earlier release for this motion. Five days of news coverage for Amanda Todd produced 192 unique articles. Rehtaeh Parsons, a 17-year-old from Nova Scotia, died 7 April 2013. A few weeks later, officials announced the Nova Scotia Cyber Safety Act on 25 April 2013. This date range produces a set of 407 unique news articles. Finally, Todd Loik, a 15-year-old from Saskatchewan, died 9 Sept. 2013. Though Loik’s province already had cyberbullying legislation in place following the trend of past deaths, officials announced the Saskatchewan Anti-bullying Action Plan 14 Nov. 2013. This date range produced a data set of 48 unique articles once duplicates were eliminated.
I began with a quantitative content analysis of a data set of 815 articles from the Factiva key word search. These articles were saved as separate documents and given a name using the teen’s last name followed by the newspaper name and the date of publication. If more than one article appeared per publisher on a given date, I added a number after the title of the newspaper. It should be noted that while most of this data set focused on newspaper publications, the Factiva search also returned a handful of radio announcements and printed CTV interviews.
I analyzed these articles using the mixed method research tool Dedoose. Each article was coded with the following identifiers: date of publication, victim of focus, genre, publication title, and gender of victim of focus. During this initial process of coding, I reduced the data set to 801 articles due to duplications. For example, newspapers in two cities, both of which are affiliated with Post Media, sometimes published identical articles with slightly different titles. In this case, Factiva did not recognize the duplication, but as I coded identifiers, I discovered the duplication.
Next, I focused on a reduced portion of my data set to conduct mixed method qualitative content analysis and quantitative content analysis. In order to code the data, it was necessary to bring the data sample for each victim to a more evenly distributed set. In order to not skew the results with a disproportionate sample, I sought to evenly represent the news coverage over time and publications and across subjects. This meant I could not arbitrarily choose exactly 50 articles each. Given that the smallest sample, the Loik articles, initially had 50 articles (before further duplications were identified), I aimed to select at least 50 results for each victim. Using what Deacon et al. (1999, p. 46) call the systematic sampling method and Budd et al. (1967, p. 22) call the interval method, I randomly selected articles in each set following a chronological order. The Hubley set began with 154 articles, so in practice, I coded one article then skipped the next two as they were arranged in Dedoose (according to the title I had assigned moving from last name, publication title, and date). I coded a total of 51 Hubley articles. There were 192 Amanda Todd articles, but coding one in four of them produced less than 50, so I also coded one in three articles from her set. This produced 64 articles. The Parsons set included 407 articles, so I coded one in seven of that set and created 58 qualitative content analysis coded documents. Finally, the Loik set produced only 48 unique articles, so I coded each of them.
My qualitative content analysis of the reduced data set involved interpreting the content of each article for both latent and manifest meaning. As Zhang and Wildemuth (2009) recommend, I defined my unit of analysis as a theme. In application, this varied from a single word such as “depression” or “suicide” as examples of the category of effects to the several sentences it might take for a claimant to blame the death of the victim on cyberbullying. During my analysis, I refer to these units as excerpts. Once each excerpt was coded, I used Dedoose and Excel to quantitatively examine the qualitatively produced results.
I coded headlines and photo captions as well as the primary text. I included eight categories in my analysis. My categories for analysis began with Entman’s definition of what frames do: definition, cause, moral judgment, and remedy. To these four, I added four additional categories—effect, blame, establishing a pattern, and defensiveness. This created a total of 4,249 excerpts with 18,251 codes attached. Each excerpt, even if it overlapped with other excerpts, was assigned to a single category, one claimant tag, one specific or general tag, and as many primary and secondary descriptors as necessary to convey the full manifest and latent meaning of the excerpt.
Cyberbullying as a Social Problem
The dominant frame present in these news articles presents cyberbullying as a social problem. A social problem begins with public awareness of recurring circumstances perceived as detrimental to society. This recognition becomes a social problem when it has gained suffient saturation of awareness that the common public is able to recognize the issue as a set of social conditions that generate concern. Mass media is a common venue for facilitating the public discourse necessary to achieve this process. A social problem is essentially a collective subjective awareness of a set of distressing social conditions.
Joel Best recently generated a framework for social problem construction (2013). He calls it the basic natural history model of the social problems process. The six stages of this model are claims making, media coverage, public reaction, policymaking, social problems work, and, finally, policy outcomes. He acknowledges that not every social problem will fit the model but asserts that this is a typical pattern of emergence. He sees social problems not as conditions but as agreed upon concerns (p. 319). While some social problems constructs might seem more valid, more urgent, or more critical than others, each must compete for attention in what Best calls the “social problems marketplace” (p. 46). As an example, he imagines a claims maker who might have scheduled a press conference to present some aspect of a social problem on September 12, 2001. No matter how worthy that cause might have been, the message would be lost altogether in the news generated from the events of the day before. Which social problems gain attention has little to do with merit and more to do with a complex market competing for attention to claims. It is worth examining which circumstances establish enough resonance to gain traction as social problems.
The most prevalent category present in news coverage of these deaths is Remedies of Cyberbullying (35%), followed by Effects of Cyberbullying on the Victim (22%) and Definition of Cyberbullying (13%). Figure 2 shows that only thirty percent of the coded excerpts fit the remaining five categories. The following sections summarize noteworthy numeric results from the qualitative content analysis that focused on the data set reduced to more evenly represent each teen. These results are organized according to the categories identified. However, since nearly all of the instances of defensive excerpts appear in the Parsons data set, results for that category are not discussed again.
As claims makers utilize the social problem frame for cyberbullying, the emphasis is on reform. Much of the public discourse debates solutions to this emerging problem. Figure 3 lists all primary descriptors that appeared in this category along with the number of instances in which that descriptor appeared. Half the excerpts coded as Remedies for Cyberbullying call for public attention and legislative change. This is followed closely by calls for school responses and education, then by criminal investigations. The emphasis of public awareness and responsibility falling on public institutions such as legislative bodies, the school system, and the judicial system demonstrate that treatment of cyberbullying in these articles focuses more on a social problem than on an individual one. Remedies aimed more at teens and their families, such as parental intervention, resolving conflict with the bully, seeking counseling, and seeking retribution all appear less frequently than calls for a more systemic reform.
Furthermore, although two-thirds (66%) of the excerpts across all eight categories focus on the specific victim instead of cyberbullying in general, excerpts in the remedy category emphasize the general issue over specific victims. Remedies is the only category in which excerpts are tagged more frequently as covering Cyberbullying in General (898 excerpts tagged) over Comments Specific to the Victim (596). Even though the majority of the articles in this data set are published in reaction to the death of a specific teen, remedy discourse focuses not on what would remedy his or her individual case but what would improve societal conditions on the general problem of cyberbullying. Their deaths are utilized to provide a forum for discussion of a larger issue. Details of their life and their death are glossed over or else emphasized only to support claims in a larger discourse. Remedies in this social problem frame emphasize a cure for cyberbullying in general rather than redress for specific events.
The social problem frame of cyberbullying implies a high mortality rate. Taken together, suicide (44% of primary descriptors for effects) and death (25%) account for more than two-thirds of the descriptors of effects for the four teens. This is unsurprising given that a typical cyberbullying case would not make the news, and these cases became newsworthy only after their suicides. However, it is noteworthy that writers presented fourteen other effects from cyberbullying in addition to suicide and death.
One negative consequence of the dominant frame of cyberbullying as a social problem is that it establishes a strong link between the problems of online peer harassment and suicide. It is problematic that cyberbullying is only considered news worthy after a death. This creates a false correlation between suicide and cyberbullying. Most forms of bullying and of cyberbullying do not result in death. The cases covered in the news stories represent the most extreme cyberbullying circumstances.
Teen suicide, even when preceded by peer harassment, is rarely covered by main-stream news media, primarily because it is believed that such news coverage may lead to “copycat” incidents (Russell, 2006, p. 117, 167). However, when the harassment is technologically mediated, reporters frame the term “cyberbullying” and act as agenda setters and gatekeepers in prioritizing these cases to publicize. The Canadian Press Stylebook cautions reporters on coverage of suicide:
To be sure, care is always called for when covering stories that involve suicide. Media outlets have long been mindful of ‘suicide contagion’ – a phenomenon in which coverage of a news story that involves someone taking their own life can heighten the risk of others trying to follow suit (McCarten, 2013, p. 26).
According to the stylebook, an exception to this rule is made when there is “a compelling public interest” such as “the teen targeted by online bullying” (p. 26). The specifically identified exception of teen cyberbullying to the restriction against suicide stories in the most recent edition of the CP stylebook is evidence that cyberbullying is coming into its own as an established social problem.
Further evidence of the cyberbullying as a social problem frame comes from the category Placing Blame for the Death of the Victim. While many of the categories of analysis I selected deliberately examined the manner in which cyberbullying was addressed in these articles, this category was open. If the discourse focused on other contributors to the death of the teen, the blame category is where such arguments would appear. However, a full seventy percent of the blame descriptors credit cyberbullying or bullying for the teen’s death.
Hierarchy of victimization
My analysis considers news coverage of two male and two female victims. Results show that news coverage emphasizes females more than males. It is, after all, a common primary framework to consider female youths as requiring a greater degree of protection than males. Christie (1986) claimed that not all people who suffer injustices receive the label of victim equally. For him, the ‘ideal victim’ resembles Little Red Riding Hood. She is female; she is young; she is naïve; she is focused on performing good deeds; however, a stranger attacks her (Christie, 1986; also Walklate, 2008). Such victimization activates a strong sense of injustice. This characterization closely fits Amanda Todd. Part of her tragedy is the unidentified stranger who preyed on a naïve young girl. Of the four teens in this case study, her death generated the most news articles per day and inspired policy change at the most rapid rate.
Carrabine et al. (2004) claim that there is a hierarchy of victimization (p. 117). At the top of this hierarchy are those who deserve to be victimized the least, someone like Christie’s Little Red Riding Hood archetype. At the bottom of the hierarchy are those who are already marginalized by society but are perceived as persons who made choices to put themselves at risk for victimization. This includes groups such as prostitutes and the homeless. Somewhere in the middle of this hierarchy are those presumed to have enough power to prevent victimization. According to this view of victimhood, Todd, as a female teen, ranks near the top, likely followed by Parsons who is also a female teen but who made a decision to attend an event with alcohol. Moving down the hierarchy, Hubley would come next because he is male but is also in a minority as the only openly gay student at his school. Last would be Loik who is vulnerable because he is a teen but is perceived as someone who should be able to defend himself; he is a white male whose sexual orientation is not addressed in news coverage of his death.
This hierarchy of victimization fits media attention paid to the death of these teens. Loik only had an average of one news article per day from his death until an announced policy response. His window of time is also the longest. Hubley follows with more coverage and a slightly shorter time until officials announced policy changes. Parsons had the most articles in her data window; however, her death took much longer to inspire policy changes than Todd. Looking at the number of articles per day between the death of the teen and proposed legislation, Parsons received less attention than Todd who sits at the top of the hierarchy as the ideal victim. The time it took for public discourse following their death to generate policy change and the amount of daily mainstream coverage on their deaths demonstrate a clear hierarchy among these teens. Claims makers emphasize the more sympathetic victims such as Todd over the less ideal victims such as Loik.
Articles appearing early in the data set for each individual convey more depth to his or her story in the form of more details and a broader discussion. As the news coverage continues, however, it simplifies to focus less on the teen and more on cyberbullying as a social problem. As this shift occurs, specifics regarding the teen solidify to a reduced chunk of details repeated so frequently that the teen becomes little more than a caricature utilized to generate certain emotions such as sympathy and outrage in conjunction with discourse on cyberbullying. This discourse emphasizes a general problem affecting society as a whole, which holds the death of the teen up as a key example but limits full discussion on the individual in order to focus on the larger social problem.
Policies written in response to the mediated public discourse framing cyberbullying as a social problem risk addressing a skewed representation of this issue. If reporters removed the cyberbullying as a social problem frame from news coverage of the death of these teens, public discourse on their deaths would look very different. Without a pattern connecting these events to one another and an emphasis on remedying the problem, these deaths reduce to tragic individual incidents. Given that teen suicide rarely makes the news without some other factor of interest, their deaths would likely characterize some of their suffering but would not snowball to generate future discussion or spur policy changes. In order to effect social change in response to a constructed problem, claims makers must rely on recognizable patterns. These teens are abstracted to promote social change and prevent cyberbullying tragedies. However, legislation drafted in response to overly simplified characterizations of the most extreme cases will likely not address more prevalent, yet less extreme, cyberbullying incidents.
The teens in the news are portrayed as clear victims of an injustice. Rodkin and Fischer (2012), however, contest the clear lines between victims and perpetrators. Kowalski et al. (2008) see cyberbullying as a group phenomenon. The cyberbullying that concerns school officials is more indicative of group dynamic peer conflict than what Hubley, Todd, or Parsons faced. The cyberbullying Loik experienced is likely more reflective of the instances measured by cyberbullying studies. However, the details of his case are largely ignored in media portrayals of his harassment. Of the four teens in this study, he is also the only one whose parents did not testify in federal Bill C-13 hearings.
The pattern presented in these cases represents only the most extreme circumstances. In opinion articles, general public claims makers comment on the circumstances of these teens by relating to something they experienced. However, each correlation reveals a far less extreme circumstance of harassment. The four deaths in this case study also involve more complexity than just cyberbullying. Conditions of peer-on-peer physical aggression, depression, rape, international sexual extortion, and homophobia all contribute to the despair each teen faced before choosing suicide. While some of these details appear in the news coverage of their deaths, the details are often swallowed in the larger narrative of cyberbullying as the center of discourse. Their lives are reified to create a clearer and more consistent frame of cyberbullying as a social problem, despite the fact that this frame is likely inconsistent with the more common forms of electronic-mediated harassment youths experience on a daily basis.
Conventional bullying is more prevalent than cyberbullying, yet the two are often correlated. The role of bullies and victims also commonly overlaps. Neither the greater prevalence of conventional bullying nor the overlap between bullies and victims serves the cause of a collective action frame, however. Advocates for change need simplicity. This requires reification of cases as well as sympathetic victims.
The literature shows that marginalized groups are more likely to be cyberbullied (Rodkin & Fischer, 2012; Shariff, 2009). In the four cases I examined, this was most emphasized with Jamie Hubley. His cyberbullying was generally attributed to his sexual orientation. The cyberbullying of Rehtaeh Parsons was also somewhat associated with ‘slut shaming’ typical of rape discourse; however, this characterization was more common from the discourse of members of the general public rather than reporters. Amanda Todd was generally portrayed as a complete victim. Though she willingly lifted her shirt to provide the image used to harass her, little public discourse vilifies this choice or portrays her as an otherwise marginalized teen. By openly admitting and owning her mistake through her online video confession, Todd gains complete public support. She becomes the beautiful teen who made a grave mistake and is not conveyed as a member of a typically marginalized group. Todd Loik, a white male whose sexual orientation is not openly identified or discussed in public discourse, is also not portrayed as someone others would expect to be harassed. However, he also receives the least news coverage. Vulnerable groups make better victims for a cause.
Sympathetic victims and simplified patterns are effective for spurring legislative responses to issues on the public agenda, especially when they are framed as a valence social problem. The power of enduring words from victims’ social media accounts, YouTube videos, and the sympathetic figures of their parents as advocates for the cause demonstrate what Best (2013) claimed about the power victims have to own and define social problems. In mass media, that ownership is framed in a way to create a story worth publishing. While victims make only six percent of the claims in this data set and their family members make 13 percent, reporters make 44 percent of the claims. This is interesting given that two thirds of the excerpts coded refer specifically to a victim and not cyberbullying as a general issue. Though victims may have the power to own social problems, in print news, it is the writers who frame their story. This often involves simplification in order to resonate with what slowly becomes a familiar pattern leading to political response. However, quick fix legislation reflecting this valence social problem frame of cyberbullying will likely not change much in the daily lived experiences of the one third of teens who regularly face cyberbullying in its varied forms.
The window of time represented in this data set saw cyberbullying move from a new term to a fully developed social problem in Canada. This development includes local policies, provincial legislation, and finally federal cyberbullying legislation. Much of these policy changes reflect the dominant frame of cyberbullying as a valence issue and a social problem. This frame reflects the high profile, ideal victim cases. Once established, a dominant frame is resistant to challenge. For Canada, that dominant frame reflects Todd and Parsons which leads to a remedy that emphasizes the prohibition of non-consensual distribution of intimate images, whether that is the more common form of cyberbullying teens experience or not.
 This Cohen quote is included in many agenda setting and/or framing theory analyses. See, for example, the following: Entman, 2007; Kitzinger, 2004; and Kuypers, 2002.
Baden, C. & Lecheler, S. (2012). Fleeting, fading, or far-reaching? A knowledge-based model of the persistence of framing effects. Communication Theory, 22, 359-382. doi: 10.1111/j.1468-2885.2012.01413.x.
Best, J. (2013). Social problems (2nd ed.). New York: W.W. Norton & Company.
Borah, P. (2011). Conceptual issues in framing theory: A systematic examination of a decade’s literature. Journal of Communication, 61, 246-263. doi: 10.1111/j.1460-2466.2011.01539.x.
Bruggemann, M. (2014). Between frame setting and frame sending: How journalists contribute to news frames. Communication Theory 24(1), 61-82.
Budd, R. W., Thorp, R. K., & Donohew, L. (1967). Content analysis of communications. Toronto: The Macmillan Company.
Carrabine, E., Iganski, P., Lee, M., Plummer, K. and South, N. (2004) Criminology: A Sociological Introduction. London: Routledge.
CBC News. (15 Jan. 2015). Rehtaeh Parsons case: Teen in explicit photo gets probation. URL: http://www.cbc.ca/m/touch/news/story/1.2901502 [Dec. 5, 2015].
Chong, D. & Druckman, J. N. (2007) A Theory of Framing and Opinion Formation in Competitive Elite Environments. Journal of Communication 57, 99-118.
Christie, N. (1986) ‘The ideal victim’, in Fattah, E.A. (ed.) From Crime Policy to Victim Policy. London: Macmillan.
Cohen, B. C. (1963). The press and foreign policy. Princeton: Princeton University Press.
Deacon, Pickering, Golding, & Murdock, (1999). Researching communications: A practical guide to methods in media and cultural analysis. New York: Oxford University Press.
Entman, R. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication 43(4), 51- 58.
Entman, R. (2007). Framing bias: Media in the distribution of power. Journal of Communication, 57, 163-173.
Entman, R. (2010). Framing media power. In P. D’Angelo & J. A. Kuypers (Eds.), Doing news frame analysis (pp. 331-355). New York: Routledge.
Ha, L. (2014). A snapshot of cyberbullying. Victims of Crime Research Digest 7. Department of Justice Canada. Retrieved from: www.justice.gc.ca/eng/rp-pr/cj-jp/victim/rd7-rr7p2.html [Jan. 10, 2015].
Kitzinger, J. (2007). Framing and frame analysis. In E. Devereux (Ed.), Media Studies (pp. 134-161). Thousand Oaks, CA: Sage Publications.
Kowalski, R. M., Limber, S. P., & Agatston, P. W. (2008). Cyber bullying: Bullying in the digital age. Malden, MA: Blackwell.
Kuypers, J. A. (2002). Press bias and politics: How the media frame controversial issues. Westport, CT: Praeger.
McCarten, J. (Ed.). (2013). The Canadian press stylebook: A guide for writers and editors, 17th ed. Toronto: The Canadian Press.
Panjvani, L. (2013). An overview of anti-bullying legislation and alternatives in Canada. Lawnow.org. URL: http://www.lawnow.org/an-overview-of-anti-bullying-legislation-and-alternatives-in-canada/ [Jan. 10, 2015].
Reese, S. D. (2010). Finding frames in a web of culture. In P. D’Angelo & J. A. Kuypers (Eds.), Doing news frame analysis (pp. 17-42). New York: Routledge.
Rodkin, P. C. & Fischer, K. (2012). Cyberbullying from psychological and legal perspectives. Missouri Law Review 77, 619-640.
Russell, N. (2006). Morals and the media: Ethics in Canadian journalism, 2nd ed. Vancouver: UBC Press.
Shariff, S. (2009). Confronting cyber-bullying: What schools NEED TO KNOW to control misconduct and avoid legal consequences. New York: Cambridge University Press.
Van Gorp, B. (2007). The constructionist approach to framing: Bringing culture back in. Journal of Communication, 57, 60-78. doi: 10.1111/j.1460.2006.00329.x.
Walklate, S. (2008). Victims. In Dictionary of youth justice. Devon, United Kingdom: Willan Publishing.
Zhang, Y., & Wildemuth, B. M. (2009). Qualitative analysis of content. Applications of social research methods to questions in information and library science, 308-319.