‘Rival visions of reality’: An analysis of the framing of Boko Haram in Nigerian newspapers and Twitter

Abstract

This article focuses on the framing of Boko Haram, a transnational terrorist group, in legacy and social media platforms. The discussion is predicated on the understanding that in spite of its popularity as a research tool, the concept of framing is still problematic. One area of contention has been the reliability and validity of framing analysis. Drawing on Robert Entman’s seminal definition, this study investigates the viability of two innovative framing approaches and explores the intersection of the framing of Boko Haram in four Nigerian newspapers and Twitter. The authors argue that, while newspapers continue to dominate the media space, it is important to acknowledge the growing relevance of social media in shaping and influencing the opinion of their users. The study’s findings support the viability of these approaches and come to the conclusion that exploring the differences between the platforms can unearth different versions of reality.

Keywords

Boko Haram framing analysis newspapers Nigeria terrorism Twitter

Framing analysis has been on the communication research agenda for several decades. Scholars have explored how communicators, especially journalists, use news frames to influence our perception and knowledge of political issues and public events. But, despite its popularity as a research tool, framing remains problematic at several levels due to the abstract nature of the concept. As Maher (2001: 84) observed, framing ‘has proved to be an elusive concept to measure’. Other pitfalls in the application of the multipronged concept have been identified by some scholars (e.g. Cacciatore et al., 2016; Carragee and Roefs, 2004; Gorp, 2007). In a 2009 study, for example, Jorg Matthes highlighted four problems of framing research, namely: ‘lack of operational precision, the descriptive focus of many analyses, neglect of visuals, and insufficient reliability reporting’ (p. 349). In addition, several studies have raised questions about reliability and validity of framing as a methodological approach for analysing media content (Hertog and McLeod, 2001; Scheufele, 1999; Tankard, 2001) and the propensity for ‘researcher bias’ (Gorp, 2005: 503).

To address the perceived weaknesses in framing analysis, there has been a surge in the application of statistical procedures and use of software in studies. For example, Matthes and Kohring (2008) employed hierarchical cluster analysis in the framing of biotechnology in the New York Times; Burscher et al. (2016) applied partitioning cluster analysis and automatic sentiment analysis in their examination of news about nuclear power; and Frederick et al. (2016) carried out thematic analysis of news articles and tweets about the Sochi Olympics using the Leximancer software. Against the backdrop of this expanded approach to framing, this study applies statistical approaches proposed by Matthes and Kohring (2008) and Frederick et al. (2016) in an examination of frames in the coverage of an archetypal political conflict – terrorism – in newspapers and Twitter. With the media framing of Boko Haram, a transnational terrorist group, as its focus, this study identifies frames in legacy and social media platforms’ coverage of the group in a selection of Nigerian newspapers and on Twitter. Comparing the two platforms is not empirically straightforward because these platforms operate on a significantly different ethos, with Twitter driven by subjectivity and newspapers normatively objective. However, a comparison is justified because it reflects the hybrid nature of the media environment and acknowledges how the participatory norms embedded in social media impact on public discourse. The comparison was also informed by the shift in news production and consumption processes in response to technological developments. All the newspapers used in the study enjoy online visibility and are therefore not spatially limited. From that standpoint, it could be argued that geographical location is not critical in terms of reach and readership as the papers are globally accessible, just like Twitter.

The main aims of the study are (1) to explore the viability of new approaches in the identification of frames in the coverage of Boko Haram, and (2) to illustrate how the interface of legacy and social media platforms can unearth divergences and reveal ‘rival visions of “reality”’ (Norris et al., 2003: 5).

Framing, media and terrorism

Framing analysis in the context of terrorism is a dynamic research process that investigates how frames about terrorist organizations, terrorist attacks, or individuals linked with terrorist activities, are produced and sustained over time. It is also concerned with ways in which the media shape social reality and/or influence public knowledge about terrorism (Norris et al., 2003). Interest in this area of research has grown following the events of 9 September 2001 and the ensuing ‘war on terror’. However, terrorism is still an ‘essentially contested concept’ (Gallie, 1956: 169), which makes it prone to ‘opportunistic appropriation’ (Saul, 2008: 3), resulting in quips such as one man’s terrorist being another man’s freedom fighter. Terrorism, in the context of this study, is conceptualized as a form of political violence that

… is premeditated and designed to create a climate of extreme fear; is directed at a wider target than the immediate victims; inherently involves attacks on random or symbolic targets, including civilians; is considered by the society in which it occurs as ‘extra-normal’, that is in the literal sense that it violates the norms regulating disputes, protest and dissent; and is used primarily, though not exclusively, to influence the political behaviour of governments, communities or specific social groups. (Wilkinson, 1977: 51)

It is also a tactic adopted by both state and non-state actors to intimidate and instigate anxiety (Morin, 2016). Even though terrorism has become an increasingly critical subject of public discourse, as well as an important area of research since the defining event of September 2001, it is not a new social problem. Some scholars contend that epochs of terrorism are interlinked and exist in a continuum (Copeland, 2001; Crenshaw, 2008; Tucker, 2001). The modern form of terrorism defies traditional rules of engagement. For instance, this method of political conflict can be undertaken by well-established command and control structured organizations or lone wolves:

Unlike other forms of violence, its target is not its immediate victims, but a larger audience. Its main goal is not to harm or punish the immediate victims but to send an intimidating message to a target population, state, or organization. (Morin, 2016: 3)

Thus, the coverage of this ‘new form of war’ (Margolis, 2004: 403) has attracted the attention of media scholars because it allows for an exploration of different perspectives presented by various political actors.

As information purveyors, the news media play a pivotal role during political conflicts by serving as a dais for contest over meaning. This is particularly critical given the media–terrorism symbiosis. Terrorists engage the media to enhance their tactical and strategic aims as well as for extensive coverage of their exploits, without which their efforts would be without meaning (Martin, 1985), and the media create a spectacle out of terrorist attacks because they meet newsworthiness criteria such as ‘novelty, conflict, social impact, importance, consequence and human interest’ (Zhang et al., 2013: 451). Yet, unlike legitimate state actors who are granted unrestricted access to the media, terrorist organizations can only attract coverage through the back door, depending on the intensity of their violence (Kampf, 2014). While unprecedented changes in the communication ecology have altered the media–terrorism interface, the news media are still major battlefields for the propagation of information and, through them, terrorists can achieve their goal of inflicting terror on an audience beyond the locations of their attacks.

The emergence of social media platforms such as YouTube, Facebook and Twitter, has enabled terrorist organizations to disseminate their messages without the mediating influence of legacy media platforms. The unique features of social media platforms – resistance to control, high interactivity, ease of use, affordability, suitability for visual and textual content – make them ideal for recruiting, fundraising, training, communication, strategizing and execution of attacks (Galily et al., 2015: 999; Weimann, 2006). These platforms allow for the presentation of competing, counter-establishment, or dissident frames and content produced and packaged by ‘produsers’ (Bruns and Schmidt, 2011: 4) and ‘terrorpreneurs’, who are not bound by journalistic standards. The growing importance of social media platforms in the dissemination of terrorist narratives points to a need to interrogate their influence on the framing of political conflict as these platforms tend to serve as soapboxes for people who congregate around homogeneous ideologies. The perceived success of these platforms as radicalization and recruitment centres is emblematic of their capacity to influence the perception of their audiences.

Context of study: Nigeria and Boko Haram

Nigeria, Africa’s most populous country, is a divided society that is polarized along geo-religious cleavages. The north is perceived to be predominantly Muslim and the south is home to mostly Christians. This polarization, which can be attributed to the country’s colonial history, fuels competition for ‘the control of state power, resource allocation, and citizenship’ (Osaghae and Suberu, 2005: 4). Political conflicts over the years have exacerbated the divides to the point of eruptions such as the 30-month civil war (1967–1970). However, even though the country had experienced different manifestations of political violence that could have been recognized as terrorist acts, terrorism was not acknowledged as a threat until the emergence of Boko Haram (Ette, 2016).

Officially known as Jama’atul Alhul Sunnah Lidda’wati wal Jihad (People Committed to the Propagation of the Prophet’s Teaching and Jihad), Boko Haram has engaged the Nigerian state in an asymmetrical warfare since the turn of the century. The group was initially regarded as a domestic threat, but it has since breached national boundaries into Cameroon, Niger and Chad, and has emerged as a transnational security threat. In 2011, the US House of Representatives Committee on Homeland Security noted an evolution in the group’s capabilities and asserted that its reach had gone beyond Nigeria (United States House of Representatives, 2011). Like other terrorist organizations associated with Islam, Boko Haram employs violence as a strategy to advance its anti-Western ideology and attempts to establish an Islamic caliphate in Nigeria (Adesoji, 2014).

The group has carried out several large-scale attacks in some parts of Nigeria, including the bombing in 2011 of the Nigerian police headquarters and the United Nations’ House, both in Abuja, and the Baga massacre in 2015, during which more than 2000 civilians were killed. The sect’s campaign of terror has led to more than 17,000 deaths (Obaji, 2016), the displacement of over two million people (Internal Displacement Monitoring Centre, 2016) and a host of other socio-political and economic problems in Nigeria. Some of its atrocities have made headlines around the world and the kidnapping of more than 200 Chibok secondary schoolgirls in April 2014 earned the group global denunciation and triggered the #bringbackourgirls Twitter campaign.

Boko Haram began to seize Nigerian territories in 2013 and, by January 2015, had gained control of about 14 local government areas in the north-eastern part of the country, which led to the declaration of a state of emergency in the main theatres of the insurgency – Borno, Yobe and Adamawa states (Bappah, 2016). However, it would seem that recent military-based counterinsurgency efforts have curtailed its spread. According to President Muhammadu Buhari, all Nigerian territories had been recovered from the insurgents by February 2016. However, the veracity of this claim has been challenged by David Rodriguez, the US Africa Command (USAFRICAOM) leader (Sahara Reporters, 2016). Still, to the President’s credit, some Internally Displaced Persons (IDPs) have returned to their homelands (Premium Times, 2017) and Sambisa forest, the former stronghold of the insurgents has been recaptured (Broomfield, 2016). Yet, following recent bombings in Maiduguri, the epicentre of the Boko Haram war, there are concerns that the threat posed by the group is far from over.

Like other international terrorist groups such as Daesh, Al Qaeda and Al Shabaab, Boko Haram employs suicide bombings, improvised explosive devices (IEDs), vehicle-borne improvised explosive devices (VBIEDs), kidnappings and beheadings (Okpaga et al., 2012) as weapons of attack. The nature and magnitude of these attacks result in extensive media coverage on legacy platforms. The group also uses social media platforms including YouTube and Twitter to publicize its successes.

As noted earlier, a great deal of framing research in communication has focused on analysis of media content and this is reflected in Boko Haram-related literature. Many studies have focused on the framing of the group in Nigerian newspapers (Ngwu et al., 2015; Obaje, 2017; Okoro and Odoemelam, 2013), British newspapers (Ette, 2016) and broadcast media (Ezeah and Emmanuel, 2016; Jibrin and Jimoh, 2017). On the whole, while some studies have targeted legacy media content, there is a dearth of social media analysis, application of computer-based approaches and inductive framing analysis. Here, we bridge this gap by examining content on both traditional and alternative media. We apply automated approaches to enhance the reliability and validity of our study and to allow frames to emerge from the text by adopting an inductive approach.

Methodological approach

The data for this study have been collected as part of a larger study on the framing of Boko Haram in the Nigerian press and on Twitter over a six-month period. Here, we present data for the month of January 2017 on Twitter and in four Nigerian newspapers – The Guardian, The Punch, Leadership and Daily Trust. Although there were no ‘what a story-moments’ in terms of terrorists attacks during this period, the socio-political and economic situation in the country kept Boko Haram in the news. For example, President Buhari’s New Year’s speech highlighted the group’s activities and government’s counterinsurgency programme, which, against the backdrop of the economic condition in the country, raised questions about the cost of combating Boko Haram. The military’s botched air strike of a refugee camp that resulted in the death of more than 200 IDPs ignited concerns about the safety of victims of terrorist attacks and the impact of the group’s activities on the country.

The newspapers selected for this study represent the two dominant publishing hubs in Nigeria. As outlined earlier, the country is polarized along ethno-regional and religious divides. The publishing hubs reflect these cleavages with Daily Trust and Leadership representing the northern hub while the southern hub is represented by The Guardian and The Punch. Portable document formats of The Punch, Leadership and Daily Trust were obtained from paid-for subscription services, while The Guardian’s content was accessed on the newspaper’s website – www.guardian.ng. A basic search using ‘Boko Haram’, its abbreviated version ‘B’Haram’ and the known names of its leaders, ‘Abubakar Shekau’, ‘Darul Tawheed’ and ‘Abu Musab al-Barnawi’, generated a total of 633 articles.

In order to identify frames in newspaper coverage of the group, we adopted Matthes and Kohring’s (2008:263) approach, which conceptualizes ‘a frame as a certain pattern in a given text that is composed of several elements’ that can be identified through a statistical technique – hierarchical cluster analysis (HCA). These elements can be pre-defined to serve as variables in a manual coding process. Frames are those clusters that emerge from the HCA. Thus, as part of the larger study, a coding frame was designed, undergirded by Entman’s (1993) definition. The frame elements – problem definition (PD), causal interpretation (CI), moral evaluation (ME) and treatment recommendation (TR) – were crucial to the effective implementation of this method because they served as the variables in the coding process.

The unit of analysis was the article in which at least one of Entman’s frame elements was visible. This was based on the realization that frame elements were not visible in all the articles. Thus, only pertinent articles – 393 (about 62% of the originally retrieved articles) – were appropriate for further analysis (see Figure 1 for details).

Figure 1.

Articles in the study classified by newspaper.

The problem definition was operationalized under three variables: ACTOR, IDENTITY and IMPACT. These described the main actors in the articles, their labelling of the sect and the effect of the activities of the sect on the Nigerian state. The causal interpretation element consisted of two variables: CAUSAL ATTRIBUTION, which described the parties responsible for the emergence or sustenance of Boko Haram, and BENEFIT ATTRIBUTION, the person/parties responsible for the gains against the insurgents. The moral evaluation variable described how different frame sponsors perceived the activities of Boko Haram. It was concerned with the perception of the activities of the group as expressions of the Islamic faith. The treatment recommendation variable spelled out what solutions were advocated for resolving the Boko Haram conflict. The coding categories applied were ‘mutually exclusive, exhaustive, and independent’ (Matthes and Kohring, 2008: 266), with a mean reliability of 0.94 and a range of 0.7 to 1.0 based on Holsti’s formula. These original variables (categoricals) were converted to dummy variables to allow for further statistical analysis. To also ensure proper results, only variables with a frequency of five percent or more, were included in the final stages of the data analysis (Matthes and Kohring, 2008: 268).

In line with Matthes and Kohring’s design, data from the coding process were analysed through a statistical procedure – cluster analysis. A cluster analysis presents one way of making sense of complex data and is especially effective if the objective is to break down the said data into smaller units that could facilitate meaning creation (Estivill-Castro and Yang, 2004). A cluster analysis, according to Rokack and Maimon (2005: 322), ‘groups data instances into subsets in such a manner that similar instances are grouped together, while different instances belong to different groups’. The goal of this procedure was therefore to locate smaller homogeneous groups and distinguish them from more heterogeneous subsets (Berry and Linoff, 1997). Here, the approach was applied to identify different groups of articles that shared similar frame elements. Since each group/cluster represented a frame, the prevalent frame for each of the newspapers was the cluster category that indicated the greatest difference between group and the least difference within group members.

Although there are assorted clustering methods, the hierarchical cluster analysis has been adopted because there are statistical measures available (the scree plot) to guide the suitable number of clusters. The agglomerative strand of the hierarchical cluster analysis, specifically Ward’s method, was also selected because it starts out by showing each article as being unique, while tracing its membership in larger groupings via a dendrogram – a tree-like diagram that indicates the ‘nested grouping of objects and similarity levels at which groupings change’ (Rokack and Maimon, 2005: 331). An examination of the dendrogram, the scree plot and the table of means and standard deviation facilitated the identification of frames. Variables with higher means indicated the frame elements that were given more salience, while lower mean values represented elements that were given lesser prominence.

Twitter

The dataset from Twitter was mined using Mozdeh’s Twitter Time Series Analysis software. The search terms ‘Boko Haram’, ‘B’Haram’, ‘Abubakar Shekau’, ‘Darul Tawheed’ and ‘Abu Musab al-Barnawi’ were used to harvest 127,180 tweets from 41,849 unique contributors in January 2017. All Boko Haram-related tweets were considered relevant as there was no attempt to apply location filters. This was an acknowledgement of the global nature of media platforms in a digital age. The tweets were exported into the Leximancer software because the text-mining tool has a capacity to analyse large quantities of data from social media sites (Bal et al., 2012; Sotiriadou et al., 2014). The Leximancer text analytics software is not only a veritable tool for the analysis of qualitative data (Cretchley et al., 2010), but it is also capable of ‘providing a basic content analysis with descriptive statistics … and able to discover the key discourses or concepts within a text, what they are made up of and how those discourses stand in relation to others within the text (Isakhan, 2005: 9).

The software produces a ‘concept map’ that indicates the dominant themes and their relationship within a data set. In addition, its application of statistics-based algorithms has been known to reduce the reliability and validity challenges of manual content analysis (Biroscak et al., 2017). As in most computer-based approaches, we interpreted and discussed the generated themes in a meaningful way by firstly exploring actual tweets to gain a robust understanding of what the themes – ‘reoccurring concepts within text that provide general meaning to a reader’ – represented (Frederick et al., 2016: 802). Secondly, we employed an axial coding strategy to reduce themes to frames, based on their homogeneity (Benaquisto, 2008). The labelling of the frames was carried out to reflect ‘the emergent context’ (Frederick et al., 2016: 802).

Results and discussion

This study set out to answer two key questions: ‘How was Boko Haram framed in legacy media represented by four Nigerian newspapers?’ and ‘How was the group framed in social media represented by Twitter?’ The idea was to test if an analysis of the two platforms would produce comparable frames.

The framing of Boko Haram in the Nigerian Press

Pulled together, the frame elements in newspaper articles delineated four clusters. The largest cluster, which comprises 207 articles, produced a frame that we termed social consequences. It was prevalent in three of the newspapers: Leadership, The Guardian and Daily Trust. Articles in this cluster focused on the vicious nature of Boko Haram attacks and the traumatic impact of their activities on citizens. The stories also captured the role of Boko Haram in Nigeria’s humanitarian crises and the harrowing experiences of child soldiers forcefully recruited by the sect. As one article reported, some of the child soldiers saw ‘their parents killed, most often before their very eyes, their sisters taken away and their homes set ablaze before they were forcefully co-opted by Boko Haram to fight alongside them’ (Abubakar, 2017, para 19). A report published by The Guardian on 6 January 2017, outlined how

the Northeast region of Nigeria has suffered a severe humanitarian crisis as a result of Boko Haram insurgency. This resulted in a drastic decline in socio-economic activities including farming, pastoralism, commerce and social interactions among the populace especially in Borno, Yobe and Adamawa states. (Datti, 2017, para 1).

The second frame, War against Boko Haram, consists of 93 articles in which all the newspapers framed Boko Haram insurgents as outliers engaged in a senseless conflict with the Nigerian state. The articles described the strategies of the Nigerian armed forces to contain the rampage of Boko Haram. For example, in an interview with the Daily Trust, Lieutenant General Tukur Y Buratai, the Chief Commandant in the war against Boko Haram, described how the Nigerian military ousted the insurgents from their strongholds, an event which the Daily Trust had described as ‘The Great Victory in Sambisa’ (1 January 2017). Similarly, in a Christmas message, President Buhari celebrated ‘the fall of “Camp Zairo”’, the last stronghold of the Boko Haram terrorists in Sambisa Forest, and commended the military for their exploits against the insurgents (The Guardian, 4 January 2017)

The third frame, which was labelled economic consequences, describes the economic toll of Boko Haram’s activities on the Nigerian state. It was prevalent in the Daily Trust and The Punch and detailed the economic cost of tackling Boko Haram. For example, the Daily Trust reported that the reconstruction and rehabilitation of the north-east where the impact of the conflict was most evident would cost the country about 121 billion naira (£27 billion). Some reports in The Punch quoted political actors who claimed that ‘the state suffered economically from the insurgency, causing the flow of investment to nosedive’ and that ‘setbacks in agricultural production caused by the Boko Haram crisis had cut the state’s monthly internally generated revenue from N1bn to less than N200m’ (Livinus, 2017: 42).

The smallest cluster consists of 28 articles that identified members of Boko Haram as Islamic fundamentalists, which meant their activities were construed as being expressions of their Islamic faith. This frame was labelled Morality and was used to explain the religious or moral context of the activities of Boko Haram. According to De Vreese (2005: 56), ‘the morality frame interprets an event or issue in the contexts of religious or moral prescriptions.’ This frame was visible in papers published in the southern hub and echoed the fact that ‘members of the group expound radical views and contempt for Western civilization [and] preach against the Western way of life and values and campaign to spread extremist Islamist ideas across Nigeria through violence’ (Ette, 2016: 457). The group was accused of carrying out ‘dastardly and ungodly acts’, which resulted in the ‘killing of thousands of Christians and destruction of hundreds of churches and over 50,000 homes’ (The Punch, 2 January 2017: 12).

The framing of Boko Haram on Twitter

The Leximancer concept map produced 17 dominant themes from 127,180 Boko Haram-related tweets, which morphed into three frames through an axial coding procedure. One of the frames was new and invisible in the newspapers. The three frames were: War against Boko Haram, Morality and War against Women. As noted already, two of these frames were similar to the frames identified in the newspapers. However, the focus of the clusters was different from narratives in the newspapers. For example, the War against Boko Haram frame on Twitter was concerned with winners and losers, and the role of President Buhari and the military as the ouster of the sect. Tweets in this category described the war as an internecine conflict, especially after a botched air strike by the Nigerian military on a refugee camp. To some commentators, the military was equally culpable of killing innocent civilians. A tweet claimed: ‘The Nigerian Military has killed as many innocent people as Boko Haram in 2 years – RT@Batarhe. Some tweets made the President and the military objects of ridicule by framing them as ‘losers’: RT @Amaka_Ewko in particular regularly mocked the President’s claim of victory over the insurgents, especially after a Boko Haram flag that the military said was captured from the insurgents was put on display. @Amak_Ewko responded: ‘Buhari’s Army can’t arrest Boko Haram members but their flag. Meanwhile the group is on the move’; ‘Biafra Today: Buhari Caught Lying Again After Raising Fake Boko Haram Flag in Aso Rock, Got exposed by IPOB.’

The Morality frame, which was also prevalent in newspapers published in the southern hub, interpreted Boko Haram’s activities in the context of Islam. Tweets associated Boko Haram with other terrorist organizations such as ISIS, which claim to be representatives of true Islam. The tweets highlighted the atrocities of Boko Haram as expressions of their religious beliefs: [RT @pmclauth: ISIS & Boko Haram taking sex-slaves, is no different from what goes on in Bradford & Luton. All in core Islamic texts], [Daily drawing. Details: DSS confirms arrest of Abu Musab Al-Barnawi ISIS Recognised Boko Haram leader], [@TheLoneVenda @zithabb Well, Boko Haram, Isis, Saudi Arabia, Yemen etc all exercise Sharia … it’s inhumane … can’t let that spread, can we?], [@pmggal Entire Muslims are terrible … no matter … its ISIS … Boko Haram just want to kill u.@iSlaveTrolls @finite_planet @SarishkaSingh].

The War against Women frame, which was only prevalent in Twitter, was used to describe how women and girls are the primary victims of the insurgents. Tweets focused on Boko Haram’s sexual and gender-based violence and how kidnapped women are turned into sexual slaves, raped and impregnated: [Boko Haram are nothing but sexual pests preying on young girls. A bunch of paedophiles], [Everyone suffered Boko Haram’s kidnapping and sexual slavery, including young Nigerian girls], [Must read: ‘Captivity bonding’ of women and young girls (youngest 11 yrs) under Boko Haram], [Heart Breaking Story: I Was Raped, I Have Three Kids for Three Terrorists – 15-year-old Boko Haram Escapee Tells Story …], [Boko Haram Jihadists Teach Boy Soldiers How to Rape Women]. The tweets also highlighted how women are deployed as weapons of war: [RT @UKInfidel: #IslamExposed: #BokoHaram Scum Use Little Girl as #SuicideBomber in #Maiduguri], [Army kills three female suicide bombers, arrests Boko Haram leader on wanted list], [RT @AshleyWarrior: 10 year old girl used as a suicide bomber in likely Boko Haram Nigeria NYE attack #IslamIsTheProblem #BanIslam].

Altogether, the frames that emerged from the analysis of tweets demonstrated how social media platforms such as Twitter empower individuals to express anti-establishment views in ways not possible on legacy media platforms.

The foregoing discussion was underpinned by one of the objectives of this study, which was to address the reliability and validity challenge of framing through the use of the approaches suggested by Matthes and Kohring (2008) and Frederick et al. (2016). These methods proved viable and led to the location of five divergent and convergent terrorism-related frames in newspaper and Twitter texts. Two frames, the War against Boko Haram and Morality, were prevalent on both platforms while the social consequences and economic consequences, and the War against Women were prevalent in newspapers and on Twitter, respectively. It is worth pointing out that there were subtle differences in the tone and direction of newspapers and tweets in the convergent frames. For example, in the War against Boko Haram frame, the President was singled out for positive coverage in the newspapers published in the northern hub. Although the military received accolades for their gallantry in the fight against Boko Haram, it was the President who was lavishly celebrated in reports of counterinsurgency programmes. He was portrayed as a warrior who defeated the insurgents in fulfilment of his campaign promise, while ex-President Goodluck Jonathan was vilified for allegedly facilitating the proliferation of the sect. In contrast to the northern press, the southern-based newspapers attributed victory to the military. The variance in direction could be explained by the ethno-religious cleavages that underpin Nigerian politics and journalism.

The Morality frame also conveyed different undertones. This frame was prevalent in The Guardian and The Punch, the southern newspapers. The publications described members of Boko Haram as Islamic fundamentalists. The northern papers, in contrast, labelled them as miscreants, criminals and troublemakers. The lexical choice of labels was instructive. As explained earlier, northern Nigeria is home to Islam. For Muslims in the north, attributing the group’s acts of violence to their Islamic faith would have raised questions about the assertion that Islam is a religion of peace. Thus, describing the conflict as a war between criminal elements and law-abiding citizens could have been an attempt to distance their activities from Islamic ideologies and practices. For the southern press, the labelling of members of Boko Haram as Islamists, in addition to associating the group with Islam, could also have been an attempt to present it as a northern problem. In fact, an army spokesperson, in an interview with The Punch, attributed the emergence and sustenance of the sect to the northeast (Akinloye, 2017).

The emergence of convergent frames, War against Boko Haram and Morality, might have been shaped by the context of our study and the Nigerian media ecology. Firstly, our study covered the month of January when news about recent counterinsurgency success was still making headlines and Twitter represented an alternative platform where ‘produsers’ could express their opinion about the issue. While most stories in the press were news or interview-based, thus requiring a degree of protocol, tweets presented a personal perspective. Secondly, as location is one of the numerous factors that influence editorial content, it could be argued that this contributed to the emergence of the Morality frame in the two southern newspapers and on Twitter.

The divergent frames, social consequences and economic consequences can be categorized as generic frames (Matthes, 2009: 350) and indicative of news values and journalistic practices. In addition, they might have been used to draw attention to Nigeria’s humanitarian crises, which could attract global attention and financial support, considering the country’s economic situation – Nigeria had been in recession since August 2016 (Martin, 2016).

As noted earlier, one of the frames identified on Twitter, War against Women, was not visible in the newspapers. The emergence of this frame signifies the difference between legacy and social media. Unlike social media, the legacy media are subject to institutional positions. Moreover, they have established identities and are restricted by internal and external factors such as ownership and operational environment. As a global platform, Twitter is not constrained by the limitations that are embedded in the content production process of newspapers. ‘Produsers’ on Twitter have no institutional positions to protect. They also have fluid and fuzzy identities and even journalists who use Twitter often do that in their personal capacity and not as a voice of their media organizations.

Conclusion

This study set out to explore the framing of Boko Haram, a transnational terrorist group, in a selection of newspapers and Twitter through the use of innovative approaches to content analysis. Matthes and Kohring’s (2008) hierarchical cluster analysis approach aided in the identification of frames in newspapers while the Leximancer software expedited the analysis of a larger set of texts, a feat that would have been difficult to execute using the first approach due to the volume of texts involved. For example, there would have been a need to apply sampling techniques, which may have increased the likelihood of bias. Although both approaches allow for some elements of subjectivity, they, however, contribute to the transparent means of identifying frames within text, thus minimizing the reliability and validity challenges of framing research. It is worth noting that we acknowledge that quantitative methods do incorporate some measures of subjectivity – for instance, while the cluster analysis enables the grouping of different articles, the labelling and interpretation of frames are subjective decisions. This often entails reviewing articles to interrogate latent meanings in each frame. The qualitative analysis of the frames made it possible to go beyond the data and drill deeper into the text to understand embedded meanings in the frames and explain their context. Pulled together, the two approaches also made it possible to identify clusters of articles and tweets that had similar frame elements and consequently to detect emerging frames.

This study has highlighted the relevance of social media platforms in the framing of issues of public interest and, while newspapers have continued to dominate the media space, it is important to acknowledge the growing relevance of alternative platforms in shaping and influencing the opinions of their users. Although it is possible to exaggerate the impact of Twitter, the number of relevant tweets during the timeframe of the study suggests that the platform provides a forum for the politically engaged. The frames that were identified demonstrate the differences between the two platforms even in instances of convergence. While convergence between legacy and social media platforms does exist, unearthing divergences has the potential to reveal a different version of reality. This was apparent in the ways in which the two platforms presented Nigeria’s ongoing battle against the Boko Haram insurgency.

Footnotes

Acknowledgements

We are indebted to anonymous referees who provided insightful and helpful comments and feedback.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Mercy Ette is a Senior Lecturer in Media, Journalism and Film at the University of Huddersfield, UK. She holds a PhD in Communication Studies from the University of Leeds. Her research focuses on journalism and conflict, media and democracy, gendered mediation and political communication. Her research has been published in leading journals including International Journal of Press/Politics, Media, War & Conflict, International Communication Gazette, Gender, Place and Culture and Journal of African Media Studies.

Sarah Joe is a Doctoral Student in Media, Journalism and Film at the University of Huddersfield. Her research interests include framing analysis, social media and political communication.

References

Abubakar

(2017) Former child soldiers fighting bigger battle than Boko Haram. Daily Trust, 28 January. Available at: http://hud.ac/drd (accessed 19 July 2017).

Adesoji

(2014) Between Maitatsine and Boko Haram: Islamic fundamentalism and the response of the Nigerian state. Africa Today Magazine 57(4): 98–119.

Akinloye

(2017) Military must look beyond Sambisa Forest to defeat Boko Haram – Brig.-Gen. Ojo. The Punch, 1 January. Available at: http://hud.ac/dob (accessed 14 September 2017).

Bal

Campbell

Pitt

(2012) Viewer reactions to online political spoof videos and advertisements. In: Close

(ed.) Online Consumer Behavior: Theory and Research in Social Media, Advertising, and E-tail. New York: Routledge, 185–208.

Bappah

(2016) Nigeria’s military failure against the Boko Haram insurgency. African Security Review 25(2): 146–158.

Benaquisto

(2008) Axial coding. In: Given

(2008) The Sage Encyclopedia of Qualitative Research Methods. London: Sage, 51–52.

Berry

MJA

Linoff

(1997) Data Mining Techniques – for Marketing, Sales and Customer Support. New York: John Wiley.

Biroscak

et al . (2017) Leximancer software as a research tool for social marketers: Application to a content analysis. Social Marketing Quarterly 23(3): 223–231.

Broomfield

(2016) Boko Haram ‘crushed’ by Nigerian army in final stronghold. Independent, 24 December. Available at: http://hud.ac/doc (accessed 24 August 2017).

10.

Bruns

Schmidt

(2011) Produsage: A closer look at continuing developments. New Review of Hypermedia and Multimedia 17(1): 3–7.

11.

Burscher

Vliegenthart

Vreese

CHD

(2016) Frames beyond words: Applying cluster and sentiment analysis to news coverage of the nuclear power issue. Social Science Computer Review 34(5): 530–545.

12.

Cacciatore

Scheufele

Iyengar

(2016) The end of framing as we know it … and the future of media effects. Mass Communication and Society 19(1): 7–23.

13.

Carragee

Roefs

(2004) The neglect of power in recent framing research. Journal of Communication 54(2): 214–233.

14.

Copeland

(2001) Is the ‘new terrorism’ really new? An analysis of the new paradigm for terrorism. Journal of Conflict Studies XXI(2): 7–27.

15.

Crenshaw

(2008) The debate over ‘new’ vs. ‘old’ terrorism’. In: Karawan

et al . (eds) Values and Violence: Intangible Aspects of Terrorism. Dordrecht: Springer.

16.

Cretchley

et al . (2010) Conversations between carers and people with schizophrenia: A qualitative analysis using Leximancer. Qualitative Health Research 20(12): 1611–1628.

17.

Datti

(2017) NEMA’s efforts in managing crisis in the Northeast. The Guardian, 6 January. Available at: http://hud.ac/d4g (accessed 6 October 2017).

18.

De Vreese

(2005) News framing: Theory and typology. Information Design Journal 13(1): 51–62.

19.

Entman

(1993) Framing: Towards clarification of a fractured paradigm. Journal of Communication 43(4): 51–58.

20.

Estivill-Castro

Yang

(2004) Fast and robust general purpose clustering algorithms. Data Mining and Knowledge Discovery 8(2): 127–150.

21.

Ette

(2016) Condensational symbols in British press coverage of Boko Haram. International Communication Gazette 78(5): 451–468.

22.

Ezeah

Emmanuel

(2016) Foreign media framing of Boko Haram insurgency: A critical analysis of BBC and Aljazeera online coverage. International Journal of Advanced Scientific Research and Management 1(10): 30–44.

23.

Frederick

Pegoraro

Burch

(2016) Echo or organic: Framing the 2014 Sochi games. Online Information Review 40(6): 798–813.

24.

Galily

Yarchi

Tamir

(2015) From Munich to Boston, and from theater to social media: The evolutionary landscape of world sporting terror. Studies in Conflict & Terrorism 38(12): 998–1007.

25.

Gallie

(1956) Essentially contested concepts. Proceedings of the Aristotelian Society 56: 167–198. Available at: https://www-jstor-org-443.web.bisu.edu.cn/stable/pdf/4544562.pdf (accessed 10 October 2016).

26.

Gorp

(2005) Where is the frame? Victims and intruders in the Belgian press coverage of the asylum issue. European Journal of Communication 20(4): 484–507.

27.

Gorp

(2007) The constructionist approach to framing: Bringing culture back in. Journal of Communication 57(1): 60–78.

28.

‘The Great Victory in Sambisa’ (2017) Daily Trust, 1 January.

29.

Hertog

McLeod

(2001) A multiperspectival approach to framing analysis: A field guide. In: Reese

et al . (eds) Framing Public Life: Perspectives of Media and Our Understanding of the Social World. Mahwah, NJ: Erlbaum.

30.

Internal Displacement Monitoring Centre (IDMC) (2016) Nigeria IDP figures analysis. Available at: http://hud.ac/dod (accessed 5 September 2016).

31.

Isakhan

(2005) From despotism to democracy: Reporting Iraq’s January 2005 election in the Australian and Middle Eastern print media. Griffith Research Online, 29 November–2 December. Available at: http://hud.ac/doe (accessed 28 August 2017).

32.

Jibrin

Jimoh

(2017) CNN and Aljazeera framing of the Boko Haram conflicts in Nigeria (January–December 2014). International Journal of Communication: 119–135.

33.

Kampf

(2014) News-media and terrorism: Changing relationship, changing definitions: Media and terrorism. Sociology Compass 8(1): 1–9.

34.

Livinus

(2017) FG begins Yola inland cargo port construction. The Punch, 7 January.

35.

Maher

(2001) Framing: An emerging paradigm or a phase of agenda setting? In: Reese

et al . (eds) Framing Public Life: Perspectives of Media and Our Understanding of the Social World. Mahwah, NJ: Erlbaum.

36.

Margolis

(2004) Terrorism and the new forms of war. Metaphilosophy 35(3): 402–413.

37.

Martin

(1985) The media’s role in international terrorism. Terrorism: An International Journal 8(2): 127–146.

38.

Martin

(2016) Africa’s biggest economy has crashed into recession. Business Insider, 31 August. Available at: http://hud.ac/doh (accessed 5 September 2017).

39.

Matthes

(2009) What’s in a frame? A content analysis of media framing studies in the world’s leading communication journals, 1990–2005. Journalism & Mass Communication Quarterly 86(2): 349–367.

40.

Matthes

Kohring

(2008) The content analysis of media frames: Toward improving reliability and validity. Journal of Communication 58(2): 258–279.

41.

Morin

(2016) Framing terror: The strategies newspapers use to frame an act as terror or crime. Journalism & Mass Communication Quarterly 93(4): 986–1005.

42.

Ngwu

Ekwe

Chiaha

(2015) Nigerian newspapers framing of the Chibok school girls abduction (a study of The Guardian, Daily Sun, The Daily Trust, and Leadership newspapers). Global Journal of Arts, Humanities and Social Sciences 3(8): 78–93.

43.

Norris

Kern

Just

(2003) Framing terrorism. In: Norris et al . (eds) Framing Terrorism: The News Media, the Government, and the Public. New York: Routledge, 3–23.

44.

Obaje

(2017) News frame patterns: An evaluation of newspaper coverage of Boko Haram attacks in Nigeria. Covenant Journal of Communication (CJOC) 4(1): 1–25.

45.

Obaji

(2016) The Nigerian war that’s slaughtered more people than Boko Haram. The Daily Beast, 11 March. Available at: http://hud.ac/dok (accessed 4 September 2016).

46.

Okoro

Odoemelam

(2013) Print media framing of Boko Haram insurgency in Nigeria: A content analytical study of The Guardian, Daily Sun, Vanguard and Thisday newspapers. Research on Humanities and Social Sciences 3(11): 86–94.

47.

Okpaga

Chijioke

Eme

(2012) Activities of Boko Haram and insecurity question in Nigeria. Arabian Journal of Business and Management Review 1(9): 77–99.

48.

Osaghae

Suberu

(2005) A history of identities, violence, and stability in Nigeria. Centre for Research on Inequality, Human Security, and Ethnicity (CRISE), Working Paper No. 6, University of Oxford.

49.

Premium Times (2017) 85 percent of IDPs return home in Yobe – Official, 27 June. Available at: http://hud.ac/dol (accessed 24 August 2017).

50.

Rokack

Maimon

(2005) Clustering methods. In: Maimon

Rokack

(eds) Data Mining and Knowledge Discovery Handbook. New York: Springer, 321–352.

51.

Sahara Reporters (2016) US General says Boko Haram still holds territory in Nigeria, 11 March. Available at: http://hud.ac/dom (accessed 24 July 2017).

52.

Saul

(2008) Defining Terrorism in International Law. Oxford: Oxford University Press.

53.

Scheufele

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

54.

Sotiriadou

Brouwers

(2014) Choosing a qualitative data analysis tool: A comparison of NVivo and Leximancer. Annals of Leisure Research 17(2): 218–234.

55.

Tankard

(2001) An empirical approach to the study of media framing. In: Reese

et al . (eds) Framing Public Life: Perspectives of Media and Our Understanding of the Social World. Mahwah, NJ: Erlbaum.

56.

Tucker

(2001) What is new about the new terrorism and how dangerous is it? Terrorism and Political Violence 14(3): 1–14.

57.

United States House of Representatives Committee on Homeland Security (2011) Available at: https://www.scribd.com/doc/74286783/Homeland-Security-Committee-Report-Boko-Haram-Emerging-Threat-to-the-U-S-Homeland?secret_password=1sq5dweo8gt9mkb9t5fj (accessed 30 April 2018).

58.

Weimann

(2006) Terror on the Internet: The New Arena, the New Challenges. Washington, DC: United States Institute for Peace [USIP] Press.

59.

Wilkinson

(1997) The media and terrorism: A reassessment. Terrorism and Political Violence 9(2): 51–64.

60.

Zhang

Shoemaker

Wang

(2013) Reality and newsworthiness: Press coverage of international terrorism by China and the United States. Asian Journal of Communication 23(5): 449–471.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB