Abstract
This research content analyzes 2,794 news articles in four leading newspapers to compare variables that predicted the coverage of Kenya governors and senators between 2013 and 2017—within the theoretical perspectives of coverage based on “news values” and “mirror of political reality.” Findings show the strongest predictors for governors’ coverage are commenting on devolution and corruption topics, while senators’ visibility is determined by criticizing national government—all under “mirror of political reality” reporting.
Introduction
News coverage of politicians remains a dominant field of political communication research with most studies focusing on parliamentarians, senators, congresspersons, and governors (Fico, 1985; Ireri & Ochieng, 2020; Morris & Clawson, 2005; Sellers & Schaffner, 2007; Tresch, 2009; Vos, 2016). The belief that you cannot understand politics without gaining knowledge of how the media operates has informed studies that desire to interrogate factors that give politicians limelight in the news (Amsalem et al., 2018). For political actors, frequent appearance in the news is critical in many ways—from building name recognition, gaining electoral advantage, influencing standing within party hierarchy, promoting policy goals, to mitigating negative publicity (Amsalem et al., 2018; Arnold, 2004; Van Aelst & Walgrave, 2016).
Thus, Van Aelst and Walgrave (2016) point out that the media function as an important arena where politicians continuously compete to grab the news media attention. When evaluating candidates for political offices, the media also provide a platform to analyze politician’s image, self-presentation, emotional displays, and personal attributes (Coleman & Wu, 2015; Olivola & Todorov, 2010). As such, among other reasons, whether in democratic or authoritarian democracies, politicians do not hesitate to exploit the power of the news media to its fullest to gain public support. These arguments are valid because as Lazarsfeld and Merton (2002) have explained, media confer status on public issues, persons, organizations, and social movements. Relatedly, while McNair (2005) says news media provide “oxygen of publicity” to politicians to enable their causes to be noticed and addressed, Ireri (2012) explains that it is rare to come across elected leaders oblivious about the importance of registering their views, thoughts, or activities in the news media.
With news media playing a central role in politics of any nation, Ireri (2012) says Kenyan politicians (the focus of this study) have not hesitated to exploit the power of the mass media to its fullest in their re-election bids and other political agendas central to them. This research, therefore, examines four leading English newspapers coverage of 47 governors and 67 senators in Kenya—first elected during the 2013 watershed general elections under the 2010 constitution. Within the theoretical context of media coverage based on “news values,” the research compares the predictors of governors and senators’ coverage depending on a noninstitutional characteristic (ethnic group size), and institutional characteristics (county size, party size, and political leadership). From the coverage as a “mirror of political reality” standpoint, the study compares whether commenting on the topics related to devolution and corruption, and criticizing the national government predicts the coverage of the governors and senators in print news.
When Kenya’s 2010 constitution was promulgated, two important political offices were created—governor and senate—devolving political power to 47 newly created county governments with governors as the chief executives. With this shift in political leadership, governors became “the new sheriffs in town” (Ochieng & Ireri, 2022), enjoying substantial powers in Kenya’s dynamic political landscape. On the contrary, at national level (national parliament), senators now play a central oversight role in the management of the devolved governments—thus putting them in a political collision course with governors. Thus, this power struggle in Kenya’s high-octane politics of devolved governance—pitting governors against senators (often sucking players in the national government), has attracted huge media coverage (Obala, 2018; Odiwuor, 2019; Wainaina, 2019) because as Cook (1986) has argued, journalists tend to cover more those in positions of authority. Similarly, with the coming of devolution, Kenya national newspapers have set aside pages that carry county affairs news—where the political differences between governors and senators are laid bare. This is why it is important to compare and establish whether variables that predict the coverage of governors and senators in national newspapers are similar or dissimilar. Also, most of studies on politicians’ visibility in the news media have mostly focused on one category of political actors (see Ireri & Ochieng, 2020; Midtbø, 2011; Sellers & Schaffner, 2007; Squire, 1988; Veblen, 1981; Vos, 2016; Weaver & Wilhoit, 1974), yet the various groups of political offices do not operate in isolation as they share common national goals, aspirations, cultures, and political environments—hence why this study comparing two political offices is significant.
It is a well-documented fact that the vast majority of studies of this nature have studied political actors in Western nations (Hooghe et al., 2015; Midtbø, 2011; Sellers & Schaffner, 2007; Tresch, 2009; Veblen, 1981), with very little evidence in young developing democracies like Kenya. Thus, one goal of the present research is to bridge the aforementioned research gaps. Importantly, this study’s findings could help draw important comparisons on whether factors that determine visibility of political actors in the West are similar or dissimilar to those in African countries. In African setting, the current investigation advances the work of Ireri (2012), Ireri and Ochieng (2020), and Ochieng and Ireri (2022) which have explored the visibility of Kenyan politicians in the national press. This study will also provide a useful foundation for further studies in African countries which share common characteristics with Kenya—in terms of nation traits (trade, Gross National Product, and economic ties), interaction and relatedness (colonial ties, regionalism, national interests, language and cultural ties), and the three traditions which characterize the Kenya media landscape (liberal–commercial, authoritarian–development, and advocacy-protest).
The New Devolved Governance in Kenya
From the way political governance systems are structured across the globe, countries either adopt centralized, decentralized, or a hybrid (a mixture of centralized and decentralized) political system. In a centralized political system, there is a hierarchical structure of authority, with all units and levels of government accountable to one center of power, while decentralization involves the ceding of political and economic powers to constituent political units (Wanyande, 2021). Wanyande (2021) explains that within the broad rubric of decentralization, there are different ways through which governance power is shared. These include deconcentration, delegation, privatization, and devolution. The most comprehensive form of decentralization is devolution as it involves genuine power-sharing between central and subnational governments—with the United States’ federal system of government being the widely known example (Wanyande, 2021).
Therefore, after gaining independence from Britain in 1963, Kenya had a brief stint with a devolved system of governance (two levels of governance) under a semifederal constitution—central government and regional governments—popularly known as majimbo in Swahili language (Kenya’s national language). It provided for autonomous regions with the power and resources to govern and initiate development projects in various local communities under a jimbo (semiautonomous region). This did not last long, however, as the national government circumvented the constitution, through starving the regional administrations with the necessary resources to execute their mandate, before ultimately dismantling the system a year after independence (Getzel, 1970; Ghai & McAuslan, 1970; Oyugi, 2005). Thus, from that period until in 2010 when the current constitution came into play, Kenyans have only known the national government with an elected president.
So the 2010 constitution saw a return to devolved leadership—which draws heavily from the US governance model—transferring some political powers and resources from the central government to the newly created 47 county administrations, democratically elected and largely independent (Kanyinga, 2016). Each of the 47 devolved units comprise of a County Executive and a County Assembly. The county governments promote socio-economic development and support-effective participation of the people in decision-making (Republic of Kenya, 2010). The governor is the Chief Executive Officer of the county and serves for two 5-year terms (Republic of Kenya, 2010). The devolved functions include planning and development, trade development and regulation, agriculture, health services, preprimary education, public works and services, control of pollution, cultural activities and public entertainment, transport, animal control, and welfare (Republic of Kenya, 2010).
The elections of 2013 were the first under the new constitution, heralding a major shift of Kenya’s political ecology since 1963. In county governments, governor appoints an Executive Committee, which supports him or her in budget and development planning. The County Assembly, which comprises elected Members of County Assembly (MCAs) at a ward level is tasked with approving the governor’s legislative agenda, among other duties (Republic of Kenya, 2010). Similarly, each county has an elected Senator who represents it in the Senate. So, Kenya has 67 senators—47 elected (for each county) and 20 nominated to represent various demographic groupings such as the youth, women, and people living with disabilities. The Senators’ responsibilities include protecting the county’s interests, crafting bills touching on counties, and determining the allocation of national revenue among counties as provided in Article 217 of the constitutions (Republic of Kenya, 2010). The constitution provides that 15% of the revenue raised by the national government is allocated to the 47 counties (Republic of Kenya, 2010). Furthermore, the Senate plays a powerful oversight role of impeaching the President, Deputy President, governors, and their deputies (Republic of Kenya, 2010). Finally, while the governors work from their respective county headquarters, senators are stationed in Nairobi (the nation’s capital)—as members of the Senate—the Upper House of Kenya’s national parliament.
The Kenya Newspapers Industry
Kenya’s press development is classified into four phases—the colonial era (1895–1962), the post-independence era (1963–1990), the multiparty era (1991–2000), and the digital media era (2002–present) (Ireri et al., 2019). The last phase has revolutionized the practice of journalism in Kenya with emerging of a plethora of online news outlets, both professional and semiprofessional as avenues for information dissemination in a country where, according to Ismail and Deane (2008), citizens consume “news and information voraciously” (p. 320). In pre-independence Kenya, Faringer (1991) groups the press into three categories: the African press, the Indian press, and European press. Both African and Indian press had similar objectives. The former was politically motivated, mostly obsessed with the agenda of African independence (Ali, 2009), while the Indian press content reflected anti-colonial resistance throughout the British Empire, which worried the colonial government (Frederiksen, 2011).
On the contrary, the main purpose of the European press was to provide information for the missionaries and settlers from England, to legitimize the rights of the colonial masters, and to provide a channel for social communication for the settlers (Ochilo, 1993). With growing African and Indian dissatisfaction triggered by the discriminative policies of the colonial regime, African leaders established newspapers to propagate their views and concerns (Karanja, 2000). Muigwithania, African Chronicle, Tangazo, Luo Magazine, Pamoja, Sauti Ya Mwafrika, Habari Za Dunia, and Afrika Mpya are some examples of the African publications (Makali, 2003). When Kenya gained independence in 1963, Ireri (2012) says the press remained mostly docile until the return of pluralism democracy in early 1990s, because as Aling’o (2007) explains, the press operated in a harsh environment—political and legal.
Newspaper circulation that reaches the adult population stands at 22.3% (Media Policy Research Center, 2015), and the industry is dominated by the Nation Media Group and the Standard Group—controlling over 80% of the national readership (Media Council of Kenya, 2020; Media Policy Research Center, 2015). A Media Council of Kenya (2020) national survey indicates newspaper readership stands at 25% from 21% in 2019, while trust in newspapers is 41% behind television (70%) and radio (67%). Daily Nation is the most-favorite newspaper at 61%—mostly preferred by those aged between 45 and 49 years (78%). The Standard at 23% is the second-most preferred newspaper—mostly favored by readers aged between 18 and 24 years (29%). The preference for the People Daily and The Star stand at 4% apiece—mostly read by those in the age bracket of 56 to 59 years (Media Council of Kenya, 2020). Of those who access their preferred publications, 51% purchase their own copy on a daily basis, 21% read from social places (like restaurants), and 11% via the internet (Media Council of Kenya, 2020).
Daily Nation, Saturday Nation, Sunday Nation, Business Daily, The East African, and Taifa Leo are all products of the Nation Media Group. The Standard Group publishes The Standard, Sunday Standard, and The Nairobian. Daily Nation and The Standard are the standard-bearer of print media journalism in the country. The Standard is the oldest mainstream newspaper, established in 1902. Daily Nation, established in 1960, is the most influential newspaper in the Eastern Africa region. Mediamax Group and Radio Africa Group are also engaged in newspaper business—publishing People Daily and The Star respectively. People Daily was started by businessman cum-politician Kenneth Matiba in 1993, as a weekly publication, but transformed into a daily paper in 1998. On July 1, 2014, it became Kenya’s only freely distributed newspaper. Obonyo (2003) says that during President Daniel Moi’s regime, People Daily was the voice of the political opposition and reported materials that Daily Nation and The Standard dared not touch. The Star was started in 2007 and positions itself as an authoritative political publication.
Three traditions make-up Kenya newspaper industry: the liberal-commercial, authoritarian-development, and advocacy-protest. Heath (1997) explains that the main characteristics of the liberal-commercial tradition are privately owned media that are competitive, commercial, and independent from official censorship. Under this tradition, the news media act as watchdogs of the government and provide citizens with information they need to make informed decisions about public affairs (Heath, 1997). In this context, Kenyan newspapers provide an important arena for robust public debates on democratic reforms, government accountability in public affairs, observance of human rights, and the respect for the rule of law (Kagwanja, 2019; Onyango-Obbo, 2019).
The main assumptions of the authoritarian tradition are mass media as powerful forces of influence and persuasion (Ellul, 1965). This model was prominent during the 24-year reign of former President Daniel Moi—between 1978 and 2002. Like other African authoritarian regimes, Moi used the media as a weapon to propagate his political mission—creating an unfavorable environment that constrained the media freedoms (Kalyango, 2011). The advocacy-protest tradition is grounded in enlightenment notions of human rationality, popular sovereignty, and free marketplace of ideas (Heath, 1997). Although this model became visible when Kenya ushered in pluralistic politics in 1991, it is still very active today as newspapers remain powerful in agitating for democratic reforms (Choto, 2019; Human Rights Watch, 2002; Ireri, 2015).
Last, in ownership, majority of Kenyan newspapers are owned by the political and economic elites. The interests of the two parties are mostly intertwined—and use the publications as tools to further their interests (Nyanjom, 2012). For example, the Aga Khan, the owner of the Daily Nation has traditionally maintained close connections with the government of the day (Cheeseman et al., 2019; Makokha, 2010). The Standard is owned by former President Daniel Moi family, while the People Daily is associated with former President Uhuru Kenyatta. Kiprono Kittony, the chairperson of the Radio Africa Group which publishes The Star, is from an influential political family. Thus, politicians’ involvement in newspaper ownership puts into question the norm of objectivity in news reporting. In fact, media owners rank third as the major forces that influence news decisions in Kenyan newsrooms (Ireri, 2015, 2017);
The impact of the media ownership and the country’s media political economy is felt in the direction in which newspapers cover politicians. Makokha (2010) points out that the politico-economic elite use these platforms to control the overall editorial, strategic, and political direction of the newspapers. In this vein, Cheeseman et al. (2019) argue that the political networks and leaning of media owners, which follow both ethnic and personal network and political calculations, have had significant influence in the reporting of political affairs. Thus, the nature of print media ownership explains why there is no media differentiation in Kenya’s newspaper industry. Instead, their political leanings are influenced by ownership, ethnic considerations and business interests (Esipisu & Khaguli, 2009; Galava, 2020).
Media Visibility of Politicians and Literature Review
Media visibility is defined as “media appearances” measured in political reporting by frequency with which a political actor appears in news (Tresch, 2009). Andrews and Caren (2010) describe media visibility as a matter of gaining coverage or not. From political communication literature, there are five theoretical perspectives which provide a robust foundation to study visibility of politicians in print and broadcast news. These perspectives are media coverage based on news values, media coverage as a mirror of political reality, political parallelism, media logic, and media economy. Vos (2014) has further simplified the five perspectives into “typology of micro-level of politicians” (p. 2447)—comprising two divisions: who a politician is and what a politician does—or what Sheafer and Wolfsfeld (2004) categorize as individual-level and institutional-level factors. Although there exists no singular theory on which studies of media visibility of political actors are founded, the above five theoretical perspectives are meaningful steps toward the formulation of an overarching theoretical framework. For sure, Kuhn (1962) argues that a theoretical framework evolves overtime as scientists continue to elaborate and extend a paradigm.
Therefore, this theorization and literature review focuses on media coverage based on “news values” and media coverage as a “mirror of political reality” because they are the only ones being tested in this study. “Who a politician is” constitutes of noninstitutional and institutional characteristics associated with a political actor. On the contrary, “what a politician does” involves political work and media work. Thus, media coverage based on news values (Galtung & Ruge, 1965; Shoemaker & Reese, 1996) governs the importance of an event or an individual to be included in news. Here, journalists apply their own standards of “newsworthiness” based on audience-centered routines (Shoemaker & Reese, 1996). In a nutshell, important news values such as “power elite,” “relevance,” and “impact on the country” raise the newsworthiness of political news sources (Vos & Van Aelst, 2018).
Within the coverage based on news values, noninstitutional characteristics have proved to be determinants of politicians’ appearance in news. Examples of noninstitutional attributes include age, gender, seniority, ethnic group, and attractiveness, among others. However, this study, only examines ethnic identity (ethnic group size). However, an assessment of ethnicity as a predictor of Kenyan politicians in newspaper news has so far received mixed findings. A 1-year study of politicians’ coverage in newspapers found that MPs affiliated with big ethnic groups received more coverage than those from smaller ones (Ireri, 2012). But later, a study spanning a 4-year coverage shows that ethnic group size didn’t predict parliamentarians’ visibility in the national press (Ireri & Ochieng, 2020).
Therefore, though the Ireri and Ochieng (2020) study found no support for ethnicity as a predictor of politicians’ media coverage, it’s important to interrogate it further because of three solid reasons. First, since Kenya’s independence ethnicity remains the fulcrum around which politics revolve (Oloo, 2015) – largely determining who gets elected. Second, ethnicity is often linked to political views and plays a big role in Kenyan journalism (Reporters Without Borders, 2022). Third, like Popper (1965a) argues that a refutation of a theory takes us closer to the truth; similarly, the continued interrogation of ethnicity in studies of this nature takes us nearer to understand the variable’s place in predicting the media coverage of Kenyan politicians.
While ethnic group as a predictor of politicians in news has not yet gained traction among scholars, other noninstitutional attributes have received considerable attention in political communication research. For example, seniority has been studied as a frequent predictor of politicians’ media coverage (see Cook, 1986; Ireri & Ochieng, 2020; Sellers & Schaffner, 2007). Studying the media coverage of U.S. senators, Sellers and Schaffner (2007) found that seniority appeared to attract a greater interest and coverage. Still in the United States, Cook (1986) reported seniority as a predictor of visibility before or after the introduction of television floor proceedings in the congress. In Kenya, seniority predicted newspaper coverage of 212 MPs in 2009 (Ireri, 2012) and 349 MPs between 2013 and 2017 (Ireri & Ochieng, 2020). In Canada, gender did not influence media visibility of party leadership hopefuls (Wagner et al., 2017), while in Kenya the variable failed to predict the visibility of 349 MPs (Ireri & Ochieng, 2020). For physical attractiveness, Tsfati et al. (2010) and Waismel-Manor and Tsfati (2011) have examined the variable among Israeli and the US politicians, respectively. Using a multivariate analysis, Tsfati et al. (2010) found that physical attractiveness was associated with the coverage of Members of the 16th Israeli Knesset in television news. Generally, this literature demonstrates that noninstitutional attributes play a central role in the appearance of political actors in the news media.
When it comes to institutional characteristics, they are considered to be naturally important for a politician’s success in politics—and thus “who a politician is” is expected to have impact on the amount of coverage (Vos, 2014). For example, the elite status of a politician is a popular guide for the selection of political actors to gain media coverage. This is so because high-ranking politicians in a party leadership hierarchy are deemed to be knowledgeable in particular topics of great interest to journalists and the general public at large. Moreover, powerful positions have more means to create (pseudo) events that fit the way journalists work (Vos & Van Aelst, 2018). Some of the institutional characteristics that have been examined include party and committee leaderships, cabinet portfolios, major party affiliation, and state or constituency size in population rankings. This study only examines political leadership, major party affiliation and county size.
In Belgium, Hooge et al. (2015) 9-year study of broadcast news reported that occupying an elite position was the strongest predictor of news coverage. In Norway, party leadership was a central variable with party leaders mentioned 200% more than others (Midtbø, 2011). Cook’s (1989) investigation into how the media covered congress reported that committee chairs or party leaders attracted more than half of the coverage between 1969 and 1986 since they have a greater influence on the policymaking process. Relatedly, Kuklinski and Sigelman (1992) also reported that leaders in the Senate had the same dominance due to the media’s focus on the legislative process. Weaver and Wilhoit (1974) findings reported a significant relationship between state size and the coverage of senators in news magazines, because large states provide a wide audience for the news, thus the reporters’ tendency to cover senators from those areas of representation. Fico (1985) confirmed Weaver and Wilhoit (1974) findings where governors representing larger constituencies received more coverage from reporters. Similarly, Kenyan governors representing the large counties in terms of population received more media coverage (Ochieng & Ireri, 2022). In Europe, Gattermann and Vasilopoulou (2015) found that political office and party leadership had a positive effect on the news visibility of MPs, while in the U.S. Senate, committee, and party leaders won more coverage compared with nonleaders (Sellers & Schaffner, 2007). However, Ochieng and Ireri (2022) found that in Kenya, political leadership was not a predictor of the county governors’ coverage in the national media. In political parties, Vos (2016) argues that the size of a party, measured by its electoral strength could influence the visibility of its members—an observation backed by Van Aelst et al. (2008), but not so in Kenya (Ochieng & Ireri, 2022). Thus, Vos (2016) found that in the coverage of Norwegian MPs, party attachment was key to their visibility—with politicians from big parties judged as being more newsworthy. Therefore, this review of literature paves the way for the study hypotheses, where hypothesis 1 focuses on noninstitutional characteristics, while hypotheses 2a, 2b, and 2c cover institutional attributes:
The second theoretical perspective is media coverage as a mirror of political reality. It postulates that media cover political actors proportionately to the degree of their political activities (McQuail, 1992; Tresch, 2009; Tuchman, 1978) guided by media routines as described in the Shoemaker and Reese (1996, 2014) hierarchy-of-influences model. Relatedly, this perspective could also be seen through the prism of Bennett’s (1990) indexing theory which postulates that the media rely on the political elite as the main sources of news—thus determining what dominates the news cycle. Thus, media coverage of news events mirrors the consensus or conflict among key government officials who manage the policy areas and the decision-making process that generates the news (Bennett, 1990; Zaller & Chiu, 1996). As mentioned above, the proportionate coverage of politicians’ activities is guided by media routines—which Lowrey (2008) says are repeated practices and forms that make it easier for journalists to accomplish tasks in an uncertain world while working within production constraints. That said, for media to cover politicians proportionately to their political events, their work is routinely monitored by journalists, because as Sheafer (2008) observes, news coverage should provide the electorate with essential information to hold elected politicians responsible for their actions or lack thereof.
Tresch (2009) explains that this perspective assumes a high degree of uniformity in news decisions and coverage. The “uniformity” element is what McQuail (1992) calls the “balance norm.” For example, in parliamentary proceedings, the media are expected to observe the balance norm as neutral chroniclers and impartial observers of the legislative process, and in doing so, the media provide a mirror image of the legislative process (Tresch, 2009). However, McQuail (1992) says the balance is rarely achieved because more attention goes to the sources who are deemed to be authoritative, or high political standing such as committee chairs.
Fogarty (2008), Gershon (2012), Ireri and Ochieng (2020), Ochieng and Ireri (2022), and Squire (1988) are some of the studies that have examined this theoretical perspective. Payne (1980) found that politically active politicians receive less media attention, while Fogarty (2008) reported no positive effect of politicians’ activities on their media coverage. Relatedly, Cook (1986) and Squire (1988) established that sponsoring bills also does not get a politician additional coverage. Reacting to issues that dominate the political agenda could pass as a mirror of political reality, thus becoming a sure ticket for coverage of politicians. Van Santen et al. (2013) research of France, Germany, and the Netherlands found that MPs’ reactions to ongoing debates or topical issues made them attract more media coverage. In Belgium, MPs who react to “hot topics” achieve significantly higher levels of visibility in media (Vos, 2016), while in Kenya, reacting to contentious issues of corruption and devolution influenced MPs’ and county governors’ coverage (Ireri & Ochieng, 2020; Ochieng & Ireri, 2022). Related to commenting on topical issues, criticizing the government also tends to attract media attention (Harcup & O’Neill, 2001; Ireri & Ochieng, 2020; Ochieng & Ireri, 2022). Consequently, this literature review on media coverage as a mirror of political reality leads to hypotheses 3a, 3b, and 3c:
Political parallelism, media logic, and media economy are the other three theoretical perspectives not examined in the present research. Hallin and Mancini (2004) explain that political parallelism is the extent to which the different media reflect distinct political orientations in their news and current affairs reporting. Hence, the media tend to give disproportionate attention to the positions, issues, and actors that correspond to their own preferences (Tresch, 2009). Media logic determines how material is organized, the style in which it is presented, and the grammar of media communication (Altheide & Snow, 1979). Thus, politicians who attend to media outlets’ production needs and requirements stand a better chance of earning coverage (Cohen et al., 2008; van Aelst et al., 2008). Under the media economy perspective, news outlets are seen as rational economic actors who balance cost of attaining information and the benefit of reporting the information to the public (Vos, 2014). This is because media organizations have limited time, space, and resources, and as such, select news to serve their own purposes and market interests (Vos, 2014).
Methods
To test the seven interrelated hypotheses, this study employed a positivist research design—specifically quantitative content analysis. Positivist data collection techniques such as quantitative content analysis use a deductive approach, starting with a theory and testing theoretical postulates using empirical data. Quantitative content analysis examines communication in a systematic, objective, and quantitative manner for the purpose of measuring variables (Kerlinger, 2000). Thus, the data-gathering technique was employed to establish a 4-year visibility of 47 governors and 67 senators in four leading English newspapers in Kenya—Daily Nation, The Star, People Daily, and The Standard.
In print journalism—the focus of the present research, quantitative content analysis has remained popular in measuring the media visibility of politicians using various units of analyzes—for example—a sentence, a paragraph, a whole story, or even a headline (Fico, 1985; Ireri & Ochieng, 2020; Tresch, 2009; Weaver & Wilhoit, 1974; Wolffeld & Sheafer, 2006). The four publications were selected for the present research because they are the leading English dailies in Kenya—and enjoy considerably high readership. Daily Nation is regarded as The New York Times of Kenya’s newspaper landscape, according to Onyebadi (2008). It commands over 55% of the newspaper circulation (Media Policy Research Center, 2015), and the most-preferred (Media Council of Kenya, 2020). Established in 1902, thus the oldest, The Standard is rated second in popularity after Daily Nation. People Daily is the country’s only daily free newspaper, a concept that ensures high readership, while The Star prides itself as an authoritative political newspaper in Kenyan journalism.
The 4-year sampling period is defined as April 9, 2013, to April 9, 2017, covering the first term of the devolution governance. Six constructed-weeks samples were drawn from each newspaper because as Luke et al. (2011) point out the approach is most efficient for time periods of 1-to-5 years. Six constructed weeks also increase the likelihood that the findings have external validity. Lacy et al. (2001) say the six constructed-weeks samples ensure that each source of cyclic variation—each day of the week—is represented equally.
Thus, the six constructed weeks produced 156 newspaper editions (Daily Nation = 42; The Standard = 42; People Daily = 36; and The Star = 36). People Daily and The Star do not publish on weekends, hence fewer editions. The 156 editions yielded 2,794 news stories that were content analyzed to determine visibility of the 47 and 67 governors and senators, respectively. The Star had a share of 33.4% of the stories (934), The Standard 25.2% (705), Daily Nation 25.1% (700), and People Daily 16.3% (455). As per the study’s purpose, a total of seven independent variables were analyzed—in relation to predicting media visibility (dependent variable) of the 114 politicians—big ethnic group, large county size, big party, political leadership, commenting on devolution, commenting on corruption, and criticizing the national government. The unit of analysis for visibility, big ethnic group, big party, county size, and committee or party leadership was a news article. The unit of analysis for commenting on corruption and devolution, and criticizing the government was a paragraph. Thus, any mention of devolution, corruption, or criticizing the national government only counted once in each paragraph as long as it is attributed to a governor or senator.
Before engaging in content analysis, a pilot study was undertaken with two coders trained in the coding procedures and the definition of the variables. Using 10% of Daily Nation articles, the coders conducted a pilot study of coding the seven independent variables, and dependent variable (the mention of a governor or senator) mentioned above. This was meant to ensure high intercoder consistency between the coders. Importantly, data from a pilot study helps strengthen the coding instrument through detection of poorly operationalized categories, chronically dissenting coders, and other flaws before the main coding. The pilot exercise yielded a mean Kappa’s inter-coder reliability coefficient of .93. The reliability coefficients for specific variables were as follows: mention of a governor or senator = 1.0; big tribe = 1.0; big county = 1.0; big party = 1.0; political leadership = 1.0; commenting on corruption = .80; commenting on devolution = 0.81; and criticizing the government = 0.81.
“Visibility” is a mention of a governor or senator in newspaper news. This operationalization is borrowed from Tresch (2009) who defines visibility as “media appearances” or “media prominence,” measured by the “frequency with which a political actor appears in the news (p. 79)” “Big ethnic group” is a politician who belongs to one of the five dominant ethnic communities in Kenya—Kikuyu, Kalenjin, Luo, Kamba, and Luhya. For context, an ethnic group postulates a language, an area, customs, values, a name, a shared lineage, and an awareness on the part of its members that they belong to the same group (Amselle & M’Bokolo, 1985). “Big party” means a governor or senator affiliated with one of the four major parties in Kenya at that time—The National Alliance (TNA), Orange Democratic Movement (ODM), United Republican Party (URP), and Wiper Democratic Party (WDP). “Political leadership” is a governor who chairs a committee in the Council of Governors or party official in his or her political party, or a senator who chairs a committee in the Senate or a leader in his or her political party hierarchy. “Big county” is a devolved unit with a population of over 600,000 people as per the Kenya National Bureau of Statistics (2015). Thus, under this definition, 29 counties qualify as “big counties.”
“Commenting on devolution” is where a governor or senator talks about devolution, a new system of governance which came into force after the promulgation of Kenya’s new constitution in 2010. To be specific, devolution focuses on three major things: (a) distribution of resources from the national government to the county governments; (b) distribution of resources within the country governments, and (c) formulation of new legislative bills by both national and county governments that touch on distribution of devolution resources. “Commenting on corruption” is where a governor or senator is talking about corruption, considering there are fears corruption is likely to be transferred to the county governments. “Criticizing national government” indicates negative criticism directed at the national government under the leadership of former President Uhuru Kenyatta. The operationalization for “big ethnic group,” “big party,” “big county,” “political leadership,” “commenting on devolution,” “commenting on corruption,” and “criticizing government,” were borrowed from Ireri and Ochieng (2020). To determine the predictors of governors or senators’ newspaper visibility, two ordinary least square regression analyses models were computed that regressed big ethnic group, big party, county size, committee leadership, commenting on devolution, commenting on corruption, and criticizing the government on the visibility of the two categories of politicians.
Findings
The 2,794 analyzed news stories resulted in 2,999 combined mentions for both governors and senators—with the former taking a share of 61.85% (1,852) and the latter 38.2% (1,147). This means that in the 4-year period across the four newspapers, governors received 6 to 10 of the coverage—despite the senators’ numerical advantage of 67 against 47. Therefore, hypothesis 1 states that ethnic group size predicts the coverage of governor and senators. From the least square regression analyzes, Table 1 below shows that ethnic group size (noninstitutional characteristic) neither predicted coverage of governors nor senators (b = .11, p = .170), and (b = .09, p = .109), respectively. When analyzed by institutional characteristics, county size determined the coverage of governors, but not senators—addressed by hypothesis 2a. Table 1 shows that county size was statistically significant in predicting the coverage of governors (b = .25, p <.05), but not for senators ((b = .08, p = .202). Hypothesis 2b tested whether party size predicted the visibility of the two categories of politicians in newspapers news. As Table 1 shows, party size was neither a predictor for governors nor senators—(b = −.01, p = .911), and (b = .07, p = .228). Similarly, political leadership (hypothesis 2c) did not predict the coverage of the politicians (b = −.11, p = .204), and (b = .14, p = .052) for governors and senators respectively.
Predictors of Governors and Senators in Newspapers Visibility
Turning to media coverage as mirror of political reality, hypothesis 3a examined whether commenting on devolution by senators and governors predicted their coverage in news. As Table 1 indicates, commenting on devolution was a determinant for both governors (b = .82, p < .05), and senators’ (b = .21, p < .05). However, devolution was stronger in predicting the coverage of governors than senators. Hypothesis 3b states: Commenting on corruption-related issues predicts the coverage of governors and senators in news. From Table 1, it is evident that corruption predicted coverage of governors only (b = .38, p < .05), and not the senators (b = .10, p = .367). Criticizing the national government predicted the coverage of both categories of politicians—governors (b = .33, p <.05), and senators (b = .59, p < .05), which means comparatively the variable (criticizing government) was a stronger predictor of the latter than the former. This was addressed by hypothesis 3c. Overall, devolution was the strongest predictor for governors’ visibility, while criticizing the national government was the determinant for the senators’ coverage. Yet another conclusion is that variables related to media coverage as a mirror of political reality are more influential in the visibility of Kenyan governors and senators in the national press.
Discussion
The major conclusion drawn from the present research findings is that factors related to media coverage as a mirror of political reality are more influential in the Kenyan newspapers’ coverage of governors and senators—two newly created political offices under the constitution of 2010. Both the gubernatorial and senatorial positions were first contested during the 2013 watershed general elections which ushered in the devolved system of governance—in addition to the existing national administration. Thus, it seems media coverage of political activities carries more weight in Kenyan political setting considering that Ireri and Ochieng (2020), and Ochieng and Ireri (2022) reported similar findings in their studies of 349 MPs, and 47 county governors respectively. Similar findings were reported by Van Santen et al. (2013), and Vos (2016)—both in Europe.
From a theoretical standpoint, it is safe to conclude that Kenyan journalists report politicians proportionately by routinely monitoring their political engagements. Two reasons could account for the tendency of Kenyan political journalists reporting to mirror a political reality. First, Kenya is always a politically charged nation where politics are the order of the day during campaign periods as well as in non-election cycles. Once an election cycle is over, almost immediately politicians engage in campaigns for the next election. Second, considering that this was the first term of devolution, the activities of governors and senators—both of whom played important roles in the implementation of the new system of governance—were always under the watchful eye of eager reporters.
Commenting on devolution was the strongest predictor of the governors’ newspapers visibility, while criticizing the national government had the highest influence in the appearance of senators in news. The executive and legislative functions of governors and senators, respectively, could explain this outcome. The governors, as the executive heads of a county, drive policy agendas. For governors, most of activities undertaken during the first term of devolution related to implementation of constitutional provisions to operationalize the 47 devolved units of governance. These included developing legislative instruments to set-up the county governments, putting in place governance structure, as well as accommodating functions transferred from the central government. As a result, these activities became relevant fodder for news to journalists. Today, whether on radio, television or newspapers, it’s rare to come across news coverage without reporting activities of a particular governor. For example, the four newspapers analyzed in this research have pages dedicated to news from counties, where Ochieng (2019) reported that county executives headed by governors were the main sources of devolution news (41.2%) in the same period under study—while national parliament (where senators are members) came third (13.1%) after the county assemblies (16.8%).
Turning to senators’ criticism of the national government, they are considered the vanguard of devolution concept at national level. One of their roles is to protect the devolved units from both internal and external interferences. For example, there were many attempts by the national government not to transfer some powers and resources to the county governments—a scenario that attracted a barrage of criticism from senators in their official oversight role to protect devolution. Consequently, the incessant disagreements between the Senate and the Executive over devolution affairs received wide media coverage (see Anami & Namuliro, 2013; Ontomwa, 2013)—knowing that conflict is a news value highly prevalent in journalistic practices. In fact, conflict was the most dominant media frame (40.2%) in the coverage of devolution during the same period under investigation (see Ochieng, 2019). This reflects Bennett et al.’s (2006) argument that the press corps has a calibration process based on official consensus and conflict that determines “what gets into the news, what prominence it receives, how long it gets covered, and who gets the voice in these stories” (p. 49).
Commenting on corruption also predicted the coverage of governors. Well, one of the major challenges facing devolution even today is corruption. And as the implementation of devolution gained momentum, graft became so pervasive in the counties to a point where the common street joke among concerned Kenyans was that corruption had also been devolved to the grassroots. This culture, which had bedeviled the national government over the years, became entrenched in the devolved structures (Cheeseman et al., 2016; D’Arcy & Cornell, 2016). Event in Mumo Nzau’s (2020) book that reviews the first five years of devolution, corruption is a recurring theme in the vast majority of the 47 counties. As such, the high graft in counties put the governors on the spotlight as several of them were accused of abetting the vice—something which in turn received heavy media coverage (see Standard Reporter, 2015). Largely, whether the governors or senators, commenting on devolution, corruption, and criticizing the national government could be seen from a political campaigns’ standpoint for their re-election. This is highly possible in a country which as mentioned above is always in a political mood from one election cycle to the next—hence the phrase “high-octane” Kenyan politics in this article’s title.
With regard to institutional characteristics, only county size predicted governors’ visibility—meaning the county chiefs representing big counties received more coverage. In a related study, Fico (1985) found that journalists were more likely to report on governors in charge of larger constituencies. The focus on the governors from big counties could be seen from a media economics perspective. This is because big counties have a large population audience, thus, newspapers would give their respective governors more coverage with a goal of increasing sales. As Picard (2004) has explained, “the mission of newspaper enterprise includes both commercial and social facets” (p. 109). Importantly, the big counties make a significant contribution to the gross domestic product (GDP)—with seven of them having a combined GDP of more than $500 million (Bundervoet et al., 2015; Republic of Kenya, 2017).
Last, surprisingly, though ethnicity is a major determinant of who gets elected into political offices in Kenya, the variable wasn’t a significant predictor of visibility of the 114 politicians in the newspaper news. What this suggests is that in the media coverage of politicians during non-election cycles, the power of ethnicity is watered down as it competes with multiple other variables which might be more attractive to journalists’ coverage of political actors. As such, the findings suggest that journalists might have given more attention to variables like commenting on devolution and corruption, and criticizing national government, but did not consider affiliation with a big ethnic group as a strong factor for the inclusion of governors or senators in the news. A good example is Mombasa Senator Hassan Omar Hassan who received high visibility (ranking sixth overall among the 67 senators), even though he belongs to a small ethnic group in Coast region. Hassan might have been propelled to the top of news headlines because he was affiliated with Wiper Democratic Party—a major political party of which he was the secretary general. He also was a vocal legislator on topical issues—ranking fifth on commenting on devolution and criticizing the national government, and seventh in commenting on corruption. The same is true for Tharaka-Nithi Senator—Kithure Kindiki who also belongs to a small and marginalized ethnic group (Tharaka) in Eastern Kenya. He ranked third overall—something which could be attributed to the fact that he was the leader of majority in the Senate. He also ranked fifth in commenting on devolution—meaning he was also vocal voice on issues of great nation interest. For both Hassan and Kindiki holding leadership positions, means they had a greater influence on the policymaking and legislative processes.
Conclusion
Content analyzing 2,794 news articles, this research examined a 4-year newspapers coverage of 114 politicians (47 governors and 67 senators) in Kenyan politics between 2013 and 2017. Within the theoretical perspective of media coverage based on “news values,” the study investigated whether ethnic group size, county size, party size, political leadership, commenting on devolution, commenting on corruption, and criticizing the national government predicted the visibility of the politicians. Findings show that the strongest predictors of governors’ coverage were commenting on devolution, commenting on corruption, and county size. On the contrary, criticizing the national government and commenting on devolution were the determinant of the senators’ coverage in the four newspapers. Therefore, the study makes an important contribution to the increasing literature on media coverage as a mirror of political reality. This theoretical model assumes that the news media cover politicians proportionately to the degree of their political work (McQuail, 1992; Tresch, 2009). Hence, it is safe to say that Kenyan journalists (guided by media routines) cover politicians in a proportionate manner by monitoring their political engagements—a conclusion in agreement with Ireri and Ochieng (2020), and Ochieng and Ireri (2022) studies on the visibility of MPs and governors over the same period.
Future research should consider including television news to measure visibility of governors and senators, one of the limitations of this study. As Coleman and Wu (2015) have argued, stories on television news enjoy “picture superiority” as they are deemed credible and interesting. Yet with television news gaining more audience in Kenya (see Media Council of Kenya, 2020, 2021); because of the expanding middle class, it is important for future research to include broadcast coverage of politicians. This could offer contrasting insights whether variables that predict coverage of political actors in print media are similar or dissimilar with those of broadcast news. The emergence of digital communication, also offers a useful space to gauge the visibility of politicians considering the wide-array of digital media outlets available in Kenya media landscape. Future research could also consider examining whether the nature of coverage (valence reporting—negative vs. positive) influences visibility of politicians in news media—an area which has been ignored in research focusing on visibility of political actors in media. With strong evidence that media coverage in Kenya tend to focus more on politicians’ activities rather than news values, conducting a qualitative study among political reporters and senior gatekeepers will offer more useful insights as to why this theoretical context is prevalent—coverage as a mirror of political reality. Considering that Kenya media is not ideologically differentiated, future research should examine and establish the place of the other two theoretical perspectives in the visibility of political actors—media logic and media economy—which have received less attention in research of this nature.
