Framing Ferguson: The interplay of advocacy and journalistic frames in local and national newspaper coverage of Michael Brown

Abstract

The shooting death of Michael Brown by a police officer in August 2014 served as a pivotal case that pushed excessive use of police force against minority groups to the national spotlight. Guided by the scholarship on protest coverage, this article investigates the interplay between advocacy and journalistic framing in the coverage of the Ferguson protests by national and local news. A content analysis of five newspapers during the first cycle of protests identified how journalistic frames of presentation derived from the ‘protest paradigm’ literature related to the content frames pushed forward by the Black Lives Matter movement. Results reveal that initial stories were predominantly episodic and focused on violence to the detriment of demands and grievances. However, episodic stories were also critical of the police response and the use of military-grade weapons to contain the demonstrations. As the weeks progressed, journalists gave space to the ideas of protestors in a more thematic way, especially on issues related to race beyond the topic of police brutality. Taken together, findings suggest small but significant progress as time continued during the first month of demonstrations after Brown’s shooting. Results presented here challenge the paradigmatic nature of protest coverage but reinforce that more space should be given to contextual narratives behind social movements’ actions in addition to coverage that is critical of police and protestor tactics.

Keywords

Content analysis journalism political communication race/ethnicity social change

Introduction

On 9 August 2014, Ferguson Police Officer Darren Wilson approached 18-year-old unarmed Michael Brown jaywalking. In a matter of 61 seconds, Brown and Wilson had a verbal and physical altercation, and Wilson shot Brown 12 times (CBS News, 2014). Within hours, spectators, mourners, and protestors demanded action against the police officer. Community unrest and peaceful protests grew alongside several bouts of rioting and looting. The subsequent release of autopsy reports, deployment of the National Guard to Ferguson, and non-indictment of Officer Wilson prompted unrest to escalate and continue through 2014.

Brown’s death received extensive national media coverage, as did a number of cases that followed, including the deaths of Eric Garner, Tamir Rice, and Freddie Gray. These events contributed to a renewed conversation about police brutality in the United States. Under these circumstances, Black Lives Matter (BLM), which began after the death of Trayvon Martin in 2012, emerged as an organized nationwide movement protesting police use of force against Blacks. BLM ‘now serves as shorthand for diverse organizing efforts – both sporadic and sustained – across the country’ (Rickford, 2015).

Ever since the Kerner Commission report (National Advisory Commission on Civil Disorders, 1968), American news media have been encouraged to fairly and accurately portray minority communities and their concerns. Nearing the Commission’s 50 year anniversary and in the wake of what some have called a new civil rights movement, the monitoring of media portrayals of Blacks remains an important focus of research.

As such, this research examines the media coverage of the first cycle of protests in Ferguson, assessing the relationship between journalistic and advocacy frames in local and national coverage. Based on previous studies on the ‘protest paradigm’ – a coverage pattern portraying collective action as spectacle, highlighting violence and deviant behavior (McLeod, 2007) – this approach draws from political communication, social movement, and mass communication research. We assess reporters’ adherence to expected journalistic norms and the penetration of prominent frames built by the movement. These findings can help answer critical questions about journalists’ abilities to represent the actuality of civil and social unrest. Even in more balanced coverage where news includes protestors’ demands and agendas amid official narratives, adherence to delegitimizing presentation frames could trivialize or even criminalize advocates in subtle ways, and accordingly help explain why national sentiments about a contested issue become polarized. In addition, by understanding how presentation and advocacy frames work together in mainstream coverage, we can start unpacking why certain types of movements are presented with an emphasis on violent tactics, while others are presented in a more legitimizing way focusing on demands.

Media frames and protest coverage

In this project, we employ framing as a theoretical toolkit for examining how issues are organized and understood by message producers within the context of social movements. Frames are persistent patterns of information presentation based on selection, emphasis, or exclusion by which symbol-handlers routinely organize discourse (Gitlin, 1980). This analysis examines how journalistic frames (established frames that frequently appear in news coverage) can serve as a conduit for advocacy frames (content choices associated with the narratives put forth by advocacy movements).

Here, it is important to make a clear conceptual distinction between the two different types of frames examined in this study: journalistic frames and advocacy frames. Both frames are a result of communication manipulation, but refer to two separate phenomena. McCombs and Ghanem (2001) refer to this distinction as ‘frames that are attributes of communication presentations and frames that are attributes of the objects being presented in the content’ (p. 79).

Journalistic frames stem from the routine, patterned presentations mainstream journalists have regularly used to style narratives. In this study, journalistic frames refer to different arrangements of information about the protests in the news (Scheufele and Iyengar, 2012). These frames originate from routines of reporting, such as the beat system, news values, pack mentality, and reliance on official sources. Typically, journalistic frames are explored in communication research that analyzes news coverage of protests and are applicable across studies of different social movement organizations (e.g. McLeod and Hertog, 1999; Shahin et al. 2016).

But journalists’ decisions on what to include and how to present the information are also bound by socio-structural influences that are external to the newsroom (Scheufele, 1999; Shoemaker and Reese, 2014). These influences include the content frames promoted by outside forces. In this research, we’re particularly interested in the advocacy frames pushed forward by BLM. Advocacy frames, predominant in the social movement framing tradition (Benford and Snow, 2000), are source-sponsored narratives pulled directly from the protest movement. These frames are developed by protestors to ‘inspire and legitimize the activities and campaigns of a social movement organization’ (Benford and Snow, 2000, p. 164). Advocacy frames can be ignored or reproduced by media organizations in stories that will organize them using different presentation patterns.

In this study, a story about police brutality against Blacks (advocacy frame) in Ferguson can be structured around different journalistic configurations. It can be framed as violent, riotous, or unruly, or could conversely emphasize legitimate protest activities and demands in a more thematic way. It could also focus on clashes with the police or the peaceful aspects of the demonstrations. Similarly, two stories using the riot journalistic frame could bring different advocacy frames, focusing on either police militarization or institutionalized racism, for example.

This project aims to identify which journalistic frames in national and local news correlate with the unique advocacy frames promoted by BLM. More specifically, we analyze how journalists used the following frames previously identified by the literature on protest coverage: thematic/episodic and frames of protest, which include confrontation, riot, and legitimizing. These frames can be used to portray specific messages pushed forward by advocacy groups (advocacy frames). In the context of BLM, we analyze the following advocacy frames: race, police militarization, police brutality, economic inequality, and new media activism. Table 1 outlines the conceptual framework used in this study.

Table 1.

Conceptual framework for presentation and content frames.

Frames	Conceptualization	Level of abstraction
Journalistic frames	Stylistic-based frames describing the structural features of a story; transferable to other contexts
Generic frames		High; applicable to all journalistic coverage
Episodic	Individual behaviors, episodes, and singular events
Thematic	Relates individual circumstances to a larger context
Frames of protest		Moderate; applicable to all protest movements
Riot	Highlights ideological deviance and criminal behaviors
Confrontation	Emphasizes protestor–police clashes
Legitimizing	Focuses on protestors’ grievances and demands
Advocacy frames	Content frames emerging from social movements’ agendas (BLM)	Low; applicable specifically to BLM
Race	A call to action and response to anti-Black racism in society
Police militarization	Militarized nature of equipment and tactics used by the riot police when containing protestors
Police brutality	Individual encounters between citizens and police or vigilantes, often leading to extrajudicial deaths or excessive use of force
Inequality	History of economic inequality in African American communities
New media activism	Online conversations surrounding hashtags and calls to action on social media

BLM, Black Lives Matter.

Examining this relationship between journalistic presentation and advocacy frames aids in the evaluation of frame building and how the overall framing of protest movements takes shape (Benford and Snow, 2000; Scheufele, 1999). Ultimately, news coverage can influence individuals’ opinions and public policy on an issue (Tewksbury et al., 2000). As advocates seek to control how their issue is framed through their own public relations strategies, this research can elucidate how these persuasive efforts are presented to audiences and policymakers by the media.

Journalistic frames

Episodic and thematic frames

Iyengar (1994) suggested that general journalistic framing could be separated almost exclusively into episodic and thematic frames. Episodic frames focus on individual behaviors, episodes, and singular events, and is often quicker to produce, drawing less attention to systematic or societal problems. Conversely, thematic frames relate individual circumstances to larger contextual issues. The overreliance on episodic frames commonly marginalizes protestors by omitting context that puts chaotic events into perspective (McLeod and Hertog, 1999).

Protest frames

Journalists have developed standard reporting routines during social unrest, typically in ways that delegitimize social movements challenging the status quo – a pattern called the ‘protest paradigm’ (Gitlin, 1980; McLeod, 2007; McLeod and Hertog, 1992). This article uses three frames found by Hertog and McLeod (2003) in the coverage of protests by US mainstream media: confrontation, riot, and legitimizing. The confrontation frame emphasizes protestor–police clashes, with coverage centered on violent battles. This frame highlights protestors as combatants, not as political actors with legitimate demands. The riot frame stresses protestors’ ideological deviance, accentuating criminal behaviors, such as looting, trespassing, and destruction of public property. Together, these two frames more strongly adhere to the protest paradigm, which deemphasizes protestors’ demands in favor of violence and spectacle (McLeod and Hertog, 1992). During the Civil Rights era, for example, news coverage exaggerated Black involvement in rioting and crime, and emphasized official voices (National Advisory Commission on Civil Disorders, 1968).

Conversely, some frames legitimize protestors’ activities, with social movements able to pose a social critique and voice opinions worthy of consideration. Hertog and McLeod (2003) identify one legitimizing frame that can appear frequently in the mainstream press, which they have called the protest frame. This frame focuses on protestors’ grievances and demands, providing space for their social critique in simple terms, and highlighting the validity of the ‘conflict between the anarchists and powerful institutions within society’ (Hertog and McLeod, 2003: 156). As such, the legitimizing protest frame departs significantly from the protest paradigm and gives social movements media space to broadcast their criticisms. In this article, we will use the term ‘legitimizing’ to describe this frame to avoid confusion with the general term ‘frames of protest’.

Previous studies on frames of protest and race show patterns that are consistent with research in other movements. For example, when civil rights protests are present in coverage at all, news generally frames protestors as deviants (Thomas, 2004; Watkins, 2001). Contrarily, a recent exploration of BLM revealed that The New York Times and St. Louis Post-Dispatch provided coverage that was seemingly sympathetic to demonstrators (Elmasry and El-Nawawy, 2016). However, instead of focusing on journalistic and advocacy frames, this research collapsed framing into positive and negative valence. Coupled with relatively low Scott Pi’s values, their inquiry provides ample opportunity to re-examine the patterns of coverage within more refined operationalizations and time periods.

Consistent with frames in the literature, we anticipate mainstream coverage will be marginalizing to the protestors:

H1. Coverage is more likely to use delegitimizing frames (riot and confrontation) than legitimizing frames.

This article also assesses if coverage by a local newspaper is different from the news provided by the mainstream national publications. Elmasry and El-Nawawy (2016) found that the mere percentage of sympathetic local coverage (79.5%) was greater than coverage from The New York Times (65%) during peak times of conflict. However, previous research on the Rodney King case revealed that the mainstream media provides more legitimizing coverage than local media (Smith, 1994). Because our time frame specifications are similar to those of Smith (1994), we predict the following:

H2. Coverage in national newspapers is more likely to be legitimizing than local newspapers.

Moreover, this research uses weekly increments to further nuance the progression of frames within an individual cycle of protest (Tarrow, 1994). Boyle et al. (2005) found that coverage of protests became less delegitimizing over time. As time progresses, we also expect journalists to become more familiar with the underlying issues and start providing a more contextual coverage. Therefore, we pose the following hypotheses:

H3. Coverage of Ferguson will progressively become more legitimizing over time in (a) local and (b) national newspapers.

H4. Coverage of Ferguson will progressively become more thematic over time in (a) local and (b) national newspapers.

In addition to thematic framing, this article also addresses the progression of frames of protest related to presentation and emphasis on violence. Because studies have not explored how the riot and confrontation frames change over time, we ask the following RQs:

RQ1. How did the use of the riot frame progress over time during the first month of the protests in Ferguson in (a) local and (b) national newspapers?

RQ2. How did the use of the confrontation frame progress over time during the first month of coverage of the protests in Ferguson in (a) local and (b) national newspapers?

Advocacy frames in the media

In addition to journalistic frames, this research investigates how news media organizations included key advocacy content frames emerging from social movements’ agendas. These frames are typically source-sponsored and can be strategically used in moments of political and social unrest to entice public debate (Tewksbury et al., 2000). Advocacy frames are intrinsically linked to the legitimizing frame of presentation, as they emerge from demands and grievances of the social movement being portrayed. However, in media content, advocacy frames are not always related to protestors themselves and are sometimes included as the journalist’s own observations or information from other sources outside the social movement. Similarly, the mere presence of an advocacy frame does not guarantee that the totality of the story will be thematic. In this study, advocacy frames are treated as attributes of the object being presented and can be accompanied by different journalistic frames of presentation.

Before we proceed, a clarification about our object of study must be made. As Freelon, McIlwain, and Clark (2016) explain, there are important distinctions between the hashtag #Blacklivesmatter, the formal chapter-based organization BLM, and the overall movement beyond the centralized organization. In this article, we discuss BLM as a general social movement rather than referring to it as a singular protest act, individual centralized groups, or a hashtag. As such, we concur with Tarrow (1994) that an isolated incident of contention is usually not a social movement unless its participants share more than temporary solidarity and sustain their challenges against opponents. Here, BLM includes the breadth of demands that appeared in social media, in online and offline mobilization forums, and those associated with the formal BLM organization website, which, at the time of Brown’s death, was in its infancy. We understand BLM as a collectivity ‘acting with some degree of organization and continuity, partly outside institutional or organizational channels, for the purpose of challenging extant systems of authority’ (Snow et al., 2004: 11).

This project addresses a series of advocacy frames that emerged – and later intensified and expanded – during the first month of coverage of the BLM efforts. Several key issues were on advocacy agendas, starting with police brutality and institutional racism. But as events unfolded – department statistics and procedural documents were made public, police began using military-like tactics and excessive force to control protesting and rioting alike – other frames emerged. We selected the following advocacy frames pushed forward by BLM on their official communication (BLM, 2015) and social media conversations (e.g. Bonilla and Rosa, 2015; Freelon et al., 2016): race relations, police militarization, police brutality, economic inequality, and new media activism.

Race relations

At the core of many BLM protests and advocacy groups’ message is the attention to race and the problematic realities Blacks endure at societal levels. In fact, the main principle of the BLM movement is a ‘call to action and a response to the virulent anti-Black racism that permeates our society’ (BLM, 2015).

Police militarization and police brutality

Two branches of calls for police accountability can be found in the advocacy frames of police militarization and police brutality. The extrajudicial deaths of Black people by police and vigilantes, beginning with the death of Trayvon Martin, were a rallying cry for BLM. The inclusion of the police brutality refers to these individual encounters between police officers and citizens, often overlooked and sometimes protected by institutional and judicial practices (Lawrence, 2000). Police brutality can be presented in association with race, but this is not always the case. In the case of police militarization, after the initial rioting in Ferguson, multiple agencies including BLM noted the militarized nature of equipment used by police, from tank-like vehicles to aircrafts. President Barack Obama responded to concerns by issuing an order to better track the equipment used by local police departments.

Economic inequality

A discussion of economic inequality within Ferguson as well as the history of economic inequality in Black communities was another prominent advocacy frame. Ferguson, a suburban community north of the central city of St. Louis has seen dramatic economic changes since 2000. In 2012, one in four residents lived below the federal poverty line.

New media activism

Social media were vital to the BLM movement as a means to promote collective action (Freelon et al., 2016). Several hashtags surfaced as important discourse organizers and incorporating these online conversations was critical mobilization.

In this article, we analyze how these five advocacy frames were selected and presented by journalists as protests unfolded in Ferguson. Thus, we ask,

RQ3. How did the advocacy frames presented progress over time in (a) local and (b) national newspapers?

In addition, even when certain ideas are covered, journalists can stylistically control presentation in an article via frames of presentation. The final research question focuses on understanding the interaction between frames of presentation and advocacy frames:

RQ4. What is the relationship between frames of journalistic presentation and advocacy frames?

Methods

This study uses a content analysis of articles published by national and local newspapers during the first cycle of protests in Ferguson. Our sample included articles from four national newspapers appearing in Cision’s¹ ranking of the top 10 most-circulated print outlets: The New York Times, The Wall Street Journal, USA TODAY, and The Washington Post (Cision, 2014). We also included stories from The St. Louis Post-Dispatch because of its proximity to Ferguson.

A census of articles was collected for 4 weeks, beginning 8 August 2014, the day before Brown’s death, and ending 4 September 2014. For our data analysis, this period of time was divided into 4 weeks of coverage. Week 1 (8–14 August) included the death of Michael Brown, and was marked by intense clashes with police, rioting and occasional looting. Week 2 (15–21 August) was also marked by violent and peaceful protests, although there were fewer arrests. On 15 August, Officer Darren Wilson’s name was released for the first time increasing tensions in the area. On 18 August, Ferguson was declared a state of emergency and the National Guard was deployed. During week 3 (22–28 August), protests continued as The National Guard left Ferguson. On 25 August, Brown’s funeral was held; Reverend Al Sharpton delivered the eulogy, emphasizing the need to continue non-violent protests of police force practices and The White House issued its intentions to review federal funding of military equipment to local police. Finally, during week 4 (29 August–4 September), police in Ferguson began wearing body cameras and reports of protest activity on all accounts declined.

Articles were retrieved via Factiva (2015) news archives, with 867 articles mentioning the word ‘Ferguson’ collected. Of those articles, we removed 244 articles that were obituaries, duplicates, letters to the editor, and advertisements. The final sample had 623 articles: 43.5 percent from The St. Louis Post-Dispatch (n = 271), 22.6 percent from The New York Times (n = 141), 19.3 percent from The Washington Post (n = 120), 7.4 percent from The Wall Street Journal (n = 46), and 7.2 percent from USA TODAY (n = 45).

Coding was divided into two teams: Two coders per team and each team coding a different set of variables. One team coded for the thematic/episodic frames. Another coder team identified the presence or absence of three protest frames – riot, confrontation, and legitimizing – and advocacy frames. Reliability tests were performed on a random sample of 52 stories, following the recommendation of Riffe et al. (2014). Krippendorff’s alpha scores were used to calculate inter-coder reliability (Hayes and Krippendorff, 2007), using ReCal (Freelon, 2013). Alphas for each variable are included with variable operationalizations.

Variables

Journalistic frames

Coders identified whether the article, in its entirety, was either mostly episodic or mostly thematic. Episodic articles focused on singular or individual events, describing a specific moment in time. Thematic articles, however, looked at the ‘big picture’, placing political issues and events in a more general context (α = .82). Thematic and episodic frames often coexist in the same story. For this study, we focus on identifying the predominant frame in each article. It is important to note that the mere presence of legitimizing frames or advocacy frames from social movement discourse does not make the article predominantly thematic. In fact, it was common for stories to be mainly descriptive of singular events, mentioning demands of protestors in a positive way but without providing any contextualization beyond a laundry list of events and grievances (e.g. Davey, 2014).

Coders also measured stories for the presence of three journalistic frames: riot, confrontation, and legitimizing. Frames are not mutually exclusive; therefore, multiple frames might coexist in articles. Articles were coded as including the riot frame if they discussed riots, vandalism, or associated protestors with a threat to order. Articles that included information about destruction of property, interruption of traffic, disruption of business, and other ways in which protests can disturb the routine of the citizens were included in this frame (α = .78). Coders measured articles as having the confrontation frame if stories indicated conflict between protestors and the police, including mentions of arrests, protestors in prison, or clashes during demonstrations. This category also includes assertions referring to police arrests of journalists covering the events (α = .87). Finally, coders measured the presence of the legitimizing frame if the article included sentences legitimizing the protestors and their motives. Any mentions – which were later further coded into different advocacy frames – of the protestors’ demands, quotes from bystanders supporting those demands, and overall explanations of the social movement determined the presence of legitimizing frame (α = .78).

Advocacy content frames

Frames were identified by examining the content of each mention of various advocacy demands from the BLM social movement. Coders were instructed to identify if the story contained one or more of the following frames: race relations (α = .90), police militarization (α = .91), new media activism (α = .82), economic inequality (α = .82), press freedom (α = .87), and police brutality (α = .87). These frames are also not mutually exclusive.

Results

Data characteristics

Of the stories coded, 10.8 percent were published on week 1, 45.8 percent on week 2, 21.9 percent on week 3, and 21.5 percent on week 4. For presentation frames, 34 percent of stories coded had the riot frame, 28.1 percent had the confrontation frame, 8.2 percent had the legitimizing frame, and 35.5 percent were primarily thematic. When it comes to advocacy frames, 31.9 percent of the articles mentioned race, 13.5 percent mentioned police militarization, 8.3 percent mentioned economic inequality, 6.7 percent mentioned social media activism, and 10.3 percent mentioned police brutality. These overall findings were further disaggregated into separate weeks of coverage for local and national newspapers.

Journalistic frames

For protest frames, findings reveal that 34 percent of the overall content contained the riot frame, suggesting that items overall emphasized the threat that protestors posed to society, acts of vandalism, and damage to private property. One article described the scene on West Florissant Avenue as chaos created by protestors and other neighbors: ‘With more than 10 days and nights of protests, many of them violent, the neighborhood has been flooded by people fleeing chaos and creating some of their own’ (Bogan, 2014). Similarly, several items emphasized consequences of protests to children in Ferguson, highlighting feelings of insecurity. For example, The St. Louis Post-Dispatch authors depicted the effects of school cancelations in local districts, describing that the clash between police and protestors threatened nutritional needs of children ‘whom rely on school for something to eat’ (Crouch and Bock, 2014).

The confrontation frame appeared in 28.1 percent of the stories coded, stressing the protestor–police conflict, and often emphasizing police use of military gear. For example, several articles addressed arrests and mentioned protestors who were ‘refusing to disperse, unlawful use of a weapon and interfering with a police officer’ (Giegerich et al., 2014). Another article described confrontations arguing ‘the racial disparity between the public here and its protectors has come to define the violent aftermath of Brown’s death’ (Leonnig et al., 2014).

The legitimizing frame appeared in only 8.2 percent of items analyzed, suggesting that pieces containing either a neutral or positive reference to demands of protestors as legitimate were rarely published. Note that since the frames are not mutually exclusive, when the legitimizing frame was present, it was always used in conjunction with other delegitimizing frames, except on three occasions. This means not only that journalists reluctantly covered protestors’ demands but did so only in stories that also emphasized violence and threats to public order. For example, in the story ‘Not their grandfathers’ protest’ (Brown, 2014), The Washington Post provided a lengthy piece on the socioeconomic grievances that led to protests in Ferguson. The piece contained several quotes from critical nonofficial sources, including Bradley Rayford, chief executive of a student government association. Rayford explained, ‘It begins with getting a traffic ticket. You get pulled over and get this huge ticket. In some parts of the city, tickets actually double. Get a couple of those and soon most people can’t afford their bills’ (para. 15). Yet, the rest of the piece emphasized riot and confrontation, attributing the Ferguson chaos to protestors who are described as ‘fueled by rage, mobilized by social media and sometimes, or so it seems to the old guard, capable of a bit of disrespect’ (Brown, 2014: para. 3). Episodic stories were also more prevalent, with only 35.5 percent of the stories being predominantly thematic. H1 predicts that the general character of news coverage would be delegitimizing. Because frames emphasizing riots and confrontations were more than three times more common than those legitimizing grievances and demands, H1 was supported.

When it comes to the differences between local and national coverage (H2: Coverage in national newspapers is more likely to be legitimizing than local newspapers), results from Table 2 reveal that the riot frame appeared more frequently on The St. Louis Post-Dispatch (38.7%) when compared to the four national outlets (30.4%) (χ²(1) = 4.75, p < .05). No statistically significant difference was found on the other frames by scope of outlet (H2 was partially supported).

Table 2.

Cross-tabulation of advocacy frames and presentation frames by newspaper scope.

	Local	National	Total	χ²
Presentation frames
Riot	38.7%	30.4%	34%	4.75*
Confrontation	25.5	30.1	28.1	1.64
Legitimizing	8.1	8.2	8.2	0.01
Thematic	37.1	34.2	35.5	0.52
Advocacy frames
Race	29.9	33.5	31.9%	0.93
Police militarization	7.7	17.9	13.5	13.62***
Economic inequality	10.7	6.5	8.3	3.48*
New media	5.9	7.4	6.7	0.54
Police brutality	8.5	11.6	10.3	1.66

p < .05; **p < .01; ***p < .001.

For H3 (Mainstream coverage of Ferguson will progressively become more legitimizing over time in (a) local and (b) national newspaper), results on Table 3 show that local coverage slowly became more legitimizing toward the protestors, but results were only marginally significant (χ²(1) = 7.79, p < .06).

Table 3.

Cross-tabulations of journalistic frames over time – National and local newspapers.

	Week 1	Week 2	Week 3	Week 4	χ²(1, N = 623)
Local newspaper
Riot	71.1_a	44.5_b	25.9_b,c	21.7_c	30.41***
Confrontation	18.4_a,_b	39.1_b	20.4_a,b	11.6_a	19.49***
Legitimizing	5.3_a	3.6_a	11.1_a	14.5_a	7.79
Thematic	20_a	31.6_a,b	41.5_a,b	51.6_b	11.66**
National newspapers
Riot	55.2_a	37.1_a	17.1_b	18.5_b	23.40***
Confrontation	44.8_a	36.6_a	20.7_a,b	18.5_b	14.05**
Legitimizing	3.4_a,b	5.1_b	15.9_a	9.2_a,b	9.45*
Thematic	7.4_a	30.7_a,b	36.7_b,c	53.3_c	19.45***

p < .05; ** p < .01; ***p < .001.

a,b,c

Each subscript letter denotes a subset of week categories whose column proportions do not differ significantly from each other with p values adjusted using the Bonferroni method in SPSS (p < .008 for six tests). For example, on the riot frame (local), weeks 2 and 3 do not differ from each other as they bear the ‘b’ subscript, but differ from week 1 as indicated by the ‘a’.

For national news, legitimizing coverage peaked on week 3 (15.9%) and the difference between the weeks was also statistically significant (χ²(1) = 9.45, p < .05). Hypothesis 3a was partially supported and 3b was supported. Important to note, although this progression suggests increased balance in coverage, the legitimizing frame was still seen less than 10 percent of the time overall. Similarly, local coverage progressively became more thematic (H4a accepted), with an increase of about 10 percent points each week from week 1 (20%) to week 4 (51.6%) (χ²(1) = 11.66, p < .01), with weeks 1 and 4 being statistically different. For national news, thematic coverage also progressively increased from only 7.4 percent on week 1 to 53.3 percent on week 4 (χ²(1) = 19.45, p < .001) (H4b accepted), also with weeks 1 and 2 being statistically different than week 4.

For RQ1 (How did the use of the riot frame progress over time during the first month of the protests in Ferguson in (a) local and (b) national newspapers?), results presented on Table 3 reveal a statistically significant difference in the use of the riot frame between the 4 weeks studied for local papers (χ²(1) = 30.41, p < .001). Z-tests show weeks 1 and 2 are statistically different from each other and week 1 is different from weeks 3 and 4. There was no statistically significant difference between weeks 3 and 4, and 3 and 2. This means coverage focused progressively less on elements of deviance from week 1 (71.1%), to week 2 (44.5%) and into weeks 3 and 4 (around 20% for both). Similarly, national outlets (H3b) had a statistically significant decrease in the riot frame from weeks 1 (55.2%) and 2 (37.1%) to weeks 3 (17.1%) and 4 (18.5%) (χ²(1) = 23.40, p < .001). For the confrontation frame (RQ2) – which focuses on clashes between police and protestors – results reveal a statistically significant difference across the 4 weeks of coverage for the local (χ²(1) = 19.49, p < .001) and national outlets (χ²(1) = 14.05, p < .01). The analysis by week shows that the focus on confrontation peaks on week 2 for the local news, which is statistically different than week 4.

Advocacy frames

When it comes to the differences between local and national coverage, results on Table 2 reveal a statistically significant difference for police militarization, which was more prevalent on national news (17.9%) when compared to local stories (7.7%) (χ²(1) = 13.62, p < .001), and economic inequality, which was more common on local news (10.7%) than national news (6.5%) (χ²(1) = 3.48, p < .05).

RQ3a asks about the evolution of advocacy frames as the weeks progressed between 8 August and 4 September 2014 at The St. Louis Post-Dispatch. No statistically significant result was found at the local level (Table 4). For the national outlets (RQ3b) results show that police militarization, which includes the use of military-grade weapons in containing the demonstrations, peaked on week 2 (24.6%), but receded back to 13.6 percent on week 3 and 9.2 percent on week 4 (χ²(1) = 10.73, p < .05). Conversely, new media activism was mentioned the most in the first 2 weeks (20.7% on week 1, 9.7% on week 2) of protests and became less common in weeks 3 (1.2%) and 4 (3.1%) (χ²(1) = 15.17, p < .01).

Table 4.

Cross-tabulations of advocacy frames over time – National and local newspapers.

	Week 1	Week 2	Week 3	Week 4	χ²(1, N = 623)
Local newspaper
Race	34.2_a	24.5_a	35.2_a	31.9_a	2.69
Police militarization	5.3_a	8.2_a	5.6_a	10.1_a	1.28
Economic inequality	10.5_a	7.3_a	11.1_a	15.9_a	3.35
New media	5.3_a	6.4_a	5.6_a	5.8_a	0.08
Police brutality	10.5_a	6.4_a	9.3_a	10.1_a	1.13
National newspapers
Race	37.9_a	32_a	35.4_a	33.8_a	0.56
Police militarization	10.3_a	24.6_a	13.6_a	9.2_a	10.73*
Economic inequality	6.9_a	5.1_a	8.5_a	7.7_a	1.24
New media	20.7_a	9.7_{a, b}	1.2_b	3.1_b	15.17**
Police brutality	6.9_a	9.1_a	13.4_a	18.5_a	4.87

p < .05; **p < .01; ***p < .001.

a,b,c

Each subscript letter denotes a subset of categories whose column proportions do not differ significantly from each other with p values adjusted using the Bonferroni method in SPSS (p < .008 for six tests). For example, for new media (national), weeks 1 and 2 do not differ ( ‘a’ subscript), but week 1 differs from 3 and 4 as indicated by the lack of ‘a’, while week 2 does not differ from any other week (has both ‘a,b’ subscript).

Interplay between journalistic and advocacy frames

RQ4 asks about the relationship between advocacy and journalistic frames in the coverage of Ferguson. Results from Table 5 reveal that 39.2 percent of the stories containing race were presented with the riot frame, versus 31.6 percent of the nonracial-related stories with the riot frame (χ²(1) = 3.48, p < .05). Stories mentioning race were also presented thematically (41.1%), more often stories not containing race (32.7%) (χ²(1) = 3.86, p < .05). Stories on police militarization were associated with riot (χ²(1) = 14.47, p < .001), confrontation (χ²(1) = 22.96, p < .001), and episodic frames (χ²(1) = 4.36, p < .001), but also with the legitimizing frame (χ²(1) = 7.27, p < .05). That means that while stories on police militarization emphasized episodes of clashes between protestors and police, they also provided more space for demands. The frame of economic inequality was not statistically correlated with any presentation frame at the p < .05 level. When it comes to new media activism, stories were associated with the riot frame (χ²(1) = 8.62, p < .01), confrontation (χ²(1) = 13.16, p < .001), and legitimizing (χ²(1) = 4.31, p < .05). Finally, for police brutality stories, they were associated with confrontation (χ²(1) = 10.47, p < .01) and the legitimizing frame (χ²(1) = 50.48, p < .001), but those demands were less likely to be framed thematically (χ²(1) = 3.86, p < .05). Overall, findings suggest that coverage of race was more thematic and focusing on deviance, while the issue of police militarization and brutality, regardless of race, was more episodic, but sympathetic to protestors.

Table 5.

Cross-tabulation of advocacy and presentation frames (N = 623).

	Race			Police militarization			Economic inequality
	Absence	Presence	χ²(1)	Absence	Presence	χ²(1)	Absence	Presence	χ²(1)
Riot	31.60%	39.20%	3.48*	31.20%	52.40%	14.47***	33.50%	40.40%	1.02
Confrontation	28.8	26.6	0.31	24.7	50	22.96***	28.5	23.1	0.71
Legitimizing	6.8	11.1	3.20	6.9	15.5	7.27*	7.5	15.4	3.91
Thematic	32.7	41.1	3.86*	37	25	4.36*	65.4	55.8	1.92
	New media			Police brutality
	Absence	Presence	χ²(1)	Absence	Presence	χ²(1)
Riot	32.50%	54.80%	8.62**	32.90%	43.80%	3.00
Confrontation	26.3	52.4	13.16***	26.1	45.3	10.47**
Legitimizing	7.6	16.7	4.31*	5.5	31.3	50.48***
Thematic	63.9	72.5	1.19	36.8	24.2	3.86*

p < .05; **p < .01; ***p < .001.

Discussion

This project examined news coverage of the Ferguson protests during the first cycle of contention that characterized the month following Michael Brown’s death. We assessed the interplay between journalistic frames and frames that emerged from the content promoted by advocacy groups as protests unfolded and BLM entered mainstream news. Our findings suggest that emphasis on episodes of violence, riot, and clashes with the police characterized the first days of the coverage, but journalists gradually gave space to the demands and grievances of protestors in a more thematic way. Our results also reveal that while race was associated to presentations that focused on deviance, issues related to police accountability were accompanied by legitimizing presentation frames.

The overall coverage of the protests focused on tactics and violent confrontations with the police, confirming the continuation of the protest paradigm’s norms (Gitlin, 1980; McLeod, 2007; McLeod and Hertog, 1992). This is particularly true for initial coverage, and its episodic focus on violence and conflict has extremely important consequences for a social movement. As previous studies have found, this type of news can be used to enable the police to preemptively justify the use of disproportional force when containing demonstrations (Donson et al., 2004). Future studies should investigate if this was the case in Ferguson – especially from a qualitative perspective – as police and national guard intervention ultimately escalated violence in the second week of demonstrations. Focus on confrontations with the police may also serve as a deterrent for those considering joining the demonstrations.

Yet, the analysis by time periods shows that coverage progressively moved away from the emphasis on violence and toward a more thematic portrayal of demands and grievances. This finding suggests that protest paradigm patterns seem to be a product of journalistic norms and the characteristics of ‘breaking news’ coverage, rather than an explicit ideological resistance to protestors (as argued by Gitlin, 1980). In Ferguson, initial stories were episodic and journalists only documented the chaos that characterized street demonstrations. After week 2, coverage became more thematic and legitimizing frames increased while delegitimizing frames decreased. However, this increase in the legitimizing frame peaked at a mere 15 percent of the stories published. As such, this article does not make the case that coverage became predominantly legitimizing, but instead that the decontextualization that happens when focusing on negative activities slightly subsided over time, perhaps giving space to a more neutral narrative.

This shift may be a result of increased familiarity with reporting. The Washington Post reporter Wesley Lowery explained in an interview that prolonged exposure to the issue shifted reporting patterns. In the first 48 hours, he described the scene in Ferguson for journalists as one ‘caught in a journalistic pack mentality … Everyone is standing on the street corner waiting for all hell to break loose … and that limits their vision in so many ways’ (Longform, 2016). Lowery, who spent months in Ferguson, explained journalistic practice as widely experimental and unfamiliar – a ‘test tube of ways and things [he] would go back and do differently’ (Longform, 2016).

From a practical standpoint, it is unrealistic to expect that news coverage from day one would be predominantly thematic. Sporadic protests are often marked by confusion and it is understandable that journalists need time to make sense of the movement in a more comprehensive manner. But we argue that journalists covering demonstrations could approach their episodic coverage in more responsible ways, making sure that demands of protestors are also present in the initial articles they produce. When violence is present, reporters should address the underlying motives behind protestor and police actions and explore grievances. This is increasingly important when it comes to reporting riots and looting. In Ferguson, for example, many people claimed rioters weren’t protestors at all (e.g. Rosenbaum, 2014) – just unfortunate opportunists – and that distinction should consistently be made clear in news coverage. Reporters should strive to identify the actual individuals behind such actions, rather than implicitly connect them to protestors (e.g. stories that state that ‘protests were marked by rioting and looting’).

Findings also reveal that the riot frame prevailed more in the local St. Louis Post-Dispatch than in the other national newspapers. This may be a product of local journalists’ familiarity with officials and police officers, sources who often hold strategic official narratives of their own. The interplay between media frames and alternative advocacy frames resonating from government and police agencies may also be a useful inquiry that we believe will support this finding.

Our results also show that while journalists gave space to advocacy frames, they were less likely to clearly position those as part of protestors’ agendas – an observation supported by the lower frequency of the legitimizing frame. Race relations were consistently the most important frame used in stories, an emphasis that could lead to a generalization of advocacy frames into a binary ‘black and white’ issue. We argue that this simplification can be problematic because BLM demands are more complex, involving patterns of systematic oppression and economic inequality. The issue of police brutality is key to this wave of protests and was covered in a predominantly episodic manner.

Though other advocacy frames appeared less frequently, interesting patterns in the progression of these frames emerged during the 4 weeks analyzed. For example, a discussion of economic inequality is used sparingly but is stable throughout this period. Police militarization increases as a reflection of the reality of police escalation and National Guard deployment to the Ferguson area. Also, the mentions of social media conversations decreased over the 4 week period, despite continued conversations and mobilization in the online realm. And, perhaps most alarmingly, only after violent police–protestor encounters happened did discussions of police brutality increase – a central argument for an accurate understanding of the BLM movement. We ask future research to adopt a media sociology approach and explore the routines and influences that lead to the promotion of certain advocacy frames over others. In addition, it is crucial to understand how these patterns may impact audiences from a media effects perspective.

Most importantly, this project addressed the interplay between advocacy and journalistic frames, an important yet understudied line of communication research (De Vreese, 2012). We find that police militarization and police brutality were associated with both delegitimizing and legitimizing frames of protest. From this, we can conclude that not all episodic emphases on violence are equal and negative when it comes to media depictions of protests. News reports can also be critical of police responses and future studies should explore if and when episodic coverage focusing on violence can work in favor of protestors’ agendas, highlighting disproportional use of force and generating outrage among populations that are not directly affected by the grievance. Effects studies may find that, in the case of Ferguson, episodic depictions of the police arriving in tanks with military-grade weapons did more to generate outrage than thematic stories focusing on racial inequality, for example.

Ultimately, the findings of this study illustrate that while journalists have made significant progress in some elements of protest reporting since the Kerner Commission of 1964, they still marginalize protestors through presentation framing choices and advocacy framing omissions. In the case of Ferguson, that happened via an emphasis on episodic violence to the detriment of protestors’ ideas, which is in line with the literature on the ‘protest paradigm’. Further advancements in journalistic work must also come from reflective work, pattern assessment, and the use of frames that not only focus on tactics but also give space to grievances and demands of those marching the streets.

Footnotes

Acknowledgements

The authors thank the coders Stacie Richards and Nicole Rusli.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors thank the University of Texas at Austin for supporting this project via their student enrichment program.

Notes

ORCID iDs

Rachel R Mourão

Danielle K Brown

Author biographies

Rachel R Mourão is an assistant professor in the School of Journalism at Michigan State University. Her areas of interest include journalism studies, political communication, new media, and Latin American Studies. Her recent research focuses on how technology impacts journalists covering political events in the United States and Brazil. Mourão holds a PhD in Journalism from the University of Texas at Austin, an MA in Latin American Studies from the University of Florida, and a BA in Journalism from the Universidade Federal do Amazonas, Brazil.

Danielle K Brown is an assistant professor in The Media School at Indiana University. Her research interests include race, social movements, and digital communication. Her recent work appears in Journal of Communication, Journalism Studies, and Journalism. Prior to her academic career, Brown worked as a photojournalist, public relations specialist, and graphic designer. Brown received her PhD in Journalism from the University of Texas at Austin.

George Sylvie is an associate professor in the School of Journalism at The University of Texas at Austin. He studies newsroom management, diversity, ethics, and management styles. He coauthored the 5th edition of the Media Management: A Casebook Approach and has authored, coauthored, or edited seven other books on media management, reporting, and writing, with chapters in nine others. He has published in Journalism & Mass Communication Quarterly, Journal of Media Economics, International Journal on Media Management, Journal of Media Business Studies, Howard Journal of Communications, Newspaper Research Journal, and Mass Communication Review and has presented seven top three papers.

References

Benford

Snow

(2000) Framing processes and social movements: An overview and assessment. Annual Review of Sociology 26: 611–639.

Black Lives Matter (2015) Black Lives Matter. Available at: www.blacklivesmatter.org

Bogan

(2014) Living beside riot zone can be confusing for kids in Ferguson. The St. Louis Post-Dispatch, 21 August. Available at: http://www.stltoday.com/news/local/crime-and-courts/living-beside-riot-zone-can-be-confusing-for-kids-in/article_b9343fa1-f01f-55aa-84d3-fabbc9ee928c.html

Bonilla

Rosa

(2015) # Ferguson: Digital protest, hashtag ethnography, and the racial politics of social media in the United States. American Ethnologist 42: 4–17.

Boyle

McCluskey

McLeod

, et al. (2005) Newspapers and protest: An examination of protest coverage from 1960 to 1999. Journalism & Mass Communication Quarterly 82(3): 638–653.

Brown

(2014) Turmoil keeps customers away from some West Florissant businesses. The St. Louis Post-Dispatch, 18 August. Available at: http://www.stltoday.com/news/local/turmoil-keeps-customers-away-from-some-west-florissant-businesses/article_a9fb22ea-2d0e-534f-8c09-131a67c60da4.html

CBS News (2014) Ferguson police radio calls surface from Michael Brown shooting. CBS News, 14 November. Available at: https://www.cbsnews.com/news/ferguson-police-radio-calls-surface-from-michael-brown-shooting/

Cision (2014) Top 10 US daily newspapers. Available at: http://www.cision.com/us/2014/06/top-10-us-daily-newspapers/

Crouch

Bock

(2014) Ferguson unrest is taking its toll on education. The St. Louis Post-Dispatch, 10 August. Available at: http://www.stltoday.com/news/local/education/ferguson-unrest-is-taking-its-toll-on-education/article_36dc4c18-975b-5794-bbaf-3a9ebdef5e3c.html

10.

Davey

(2014) Amid mourning, call for change. The New York Times, 24 August. Available at: https://www.nytimes.com/2014/08/26/us/michael-brown-funeral.html

11.

De Vreese

(2012) New avenues for framing research. American Behavioral Scientist 56(3): 365–375.

12.

Donson

Chesters

Welsh

, et al. (2004) Rebels with a cause, folk devils without a panic: Press jingoism, policing tactics and anti-capitalist protest in London and Prague. Internet Journal of Criminology 2. Available at: http://orca.cf.ac.uk/60834/1/Rebel%20with%20a%20cause%20folk%20devils%20%20%20%20%20.pdf

13.

Elmasry

El-Nawawy

(2016) Do black lives matter? Journalism Practice 11: 857–875.

14.

Factiva (2015) Dow Jones Factiva. Available at: http://www.factiva.com/

15.

Freelon

(2013) ReCal OIR: Ordinal, interval, and ratio intercoder reliability as a web. International Journal of Internet Science 8: 10–16.

16.

Freelon

McIlwain

Clark

(2016) Beyond the hashtags: #Ferguson, #Blacklivesmatter, and the online struggle for offline justice. Available at: http://cmsimpact.org/resource/beyond-hashtags-ferguson-blacklivesmatter-online-struggle-offline-justice/

17.

Giegerich

Currier

Holleman

(2014) Are outside instigators to blame for Ferguson violence? It’s complicated. St. Louis Post-Dispatch, 20 August. Available at: http://www.stltoday.com/news/local/crime-and-courts/are-outside-instigators-to-blame-for-ferguson-violence-it-s/article_74c62482-d91e-552f-8c36-8fc98c99b0d2.html

18.

Gitlin

(1980) The Whole World Is Watching: Mass Media in the Making & Unmaking of the New Left. Berkeley, CA: University of California Press.

19.

Hayes

Krippendorff

(2007) Answering the call for a standard reliability measure for coding data. Communication Methods and Measures 1(1): 77–89.

20.

Hertog

McLeod

(2003) A multiperspectival approach to framing analysis: A field guide. In: Reese

Gandy

Grant

(eds) Framing Public Life: Perspectives on Media and Our Understanding of the Social World. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 139–161.

21.

Iyengar

(1994) Is Anyone Responsible? How Television Frames Political Issues. Chicago, IL: The University of Chicago Press.

22.

Lawrence

(2000) The politics of force: Media and the construction of police brutality. Jackson, TN: University of California Press.

23.

Leonnig

Lowery

Berman

(2014) Even before Michael Brown’s slaying in Ferguson, racial questions hung over police. The Washington Post, 13 August. Available at: https://www.washingtonpost.com/politics/even-before-teen-michael-browns-slaying-in-mo-racial-questions-have-hung-over-police/2014/08/13/78b3c5c6-2307-11e4-86ca-6f03cbd15c1a_story.html?noredirect=on&utm_term=.dd1e8a64824d

24.

Longform (2016) Longform Podcast #222: Wesley Lowery. Available at: https://longform.org/posts/longform-podcast-222-wesley-lowery (accessed 21 February 2017).

25.

McCombs

Ghanem

(2001) The convergence of agenda setting and framing. In: Reese

Gandy

Grant

(eds) Framing Public Life. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 67–82.

26.

McLeod

(2007) News coverage and social protest: How the media’s protect paradigm exacerbates social conflict. Journal of Dispute Resolution 185: 94.

27.

McLeod

Hertog

(1992) The manufacture of ‘public opinion’ by reporters: Informal cues for public perceptions of protest groups. Discourse & Society 3(3): 259–275.

28.

McLeod

Hertog

(1999) Social control, social change and the mass media’s role in the regulation of protest groups. In: Demers

Viswanath

(eds) Mass Media, Social Control and Social Change: A Macrosocial Perspective. Ames, IA: Iowa State University Press, pp. 305–330.

29.

National Advisory Commission on Civil Disorders (1968) The Report of the National Advisory Commission on Civil Disorders. New York: Bantam Books.

30.

Rickford

(2016) Black lives matter: Toward a modern practice of mass struggle. New Labor Forum 25 (1): 34–42.

31.

Riffe

Lacy

Fico

(2014) Analyzing Media Messages: Using Quantitative Content Analysis in Research, 3rd edn. New York: Routledge.

32.

Rosenbaum

(2014) Who are the protestors getting arrested in Ferguson? NPR. Available at: https://www.npr.org/2014/08/21/342207432/who-are-the-protesters-getting-arrested-in-ferguson

33.

Scheufele

(1999) Framing as a theory of media effects. Journal of Communication 49(1): 103–122.

34.

Scheufele

Iyengar

(2012) The state of framing research: A call for new directions. In Kenski

Jamieson

(eds) The Oxford handbook of political communication. New York, NY: Oxford University Press.

35.

Shahin

Zheng

Sturm

Fadnis

(2016) Protesting the paradigm: A comparative study of news coverage of protests in Brazil, China, and India. The International Journal of Press/Politics 21(2): 143–164.

36.

Shoemaker

Reese

(2014) Mediating the Message in the 21st Century: A Media Sociology Perspective. New York: Taylor & Francis Group.

37.

Smith

(1994) Transmitting Race: The Los Angeles Riot in Television News. Cambridge, MA: Joan Shorenstein Barone Center on the Press, Politics, and Public Policy, John F. Kennedy School of Government, Harvard University.

38.

Snow

Soule

Kriesi

(2004) The Blackwell Companion to Social Movements. Malden, MA: Blackwell.

39.

Tandoc

Jr (2015) Why web analytics click: Factors affecting the ways journalists use audience metrics. Journalism Studies 16(6): 782–799.

40.

Tarrow

(1994) Power in Movement: Social Movements, Collective Action and Politics. New York: Cambridge University Press.

41.

Tewksbury

Jones

Peske

, et al. (2000) The interaction of news and advocate frames: Manipulating audience perceptions of a local public policy issue. Journalism & Mass Communication Quarterly 77(4): 804–829.

42.

Thomas

III (2004) Television news and the civil rights struggle: The views in Virginia and Mississippi. Southern Spaces. Available at: https://southernspaces.org/2004/television-news-and-civil-rights-struggle-views-virginia-and-mississippi

43.

Watkins

(2001) Framing protest: News media frames of the million man march. Critical Studies in Media Communication 18(1): 83–101.