The news media and election campaign coverage : a case study of the Ghanaian news media and the 2012 General Election

(1)

The news media and election campaign coverage. A case study of the

Ghanaian news media and the 2012 General Election.

Collins Atta Clancy Opoku

10832831

University of Amsterdam

(2)

ABSTRACT

Applying western European and American theoretical frameworks to the understudied Ghanaian context, this study examines media ownership and whether it is related to biased coverage, horserace coverage and conflict framing of election campaigns by the Ghanaian news media. Drawing on data from two online news websites in a content analysis study, this study found that contrary to expectations ownership did not strongly influence how the Ghanaian news media covered the election campaign. Furthermore, despite commercial ownership, the studied news media covered the 2012 election campaign in an issue frame manner instead of a horserace manner, fulfilling their primary normative responsibility of informing citizens about their leaders and the alternatives at their disposal in the democracy. However, the two main political parties in the country were the most visible in the media, raising concerns about the viability of the other parties in the democratic landscape. The findings are discussed in the light of extant knowledge about media ownership and its effect on news coverage and framing of presidential election campaigns.

(3)

INTRODUCTION

"For an election to be truly fair, different parties and candidates should have equal access to the media. Neither state power or the power of money should determine that one party gets a hearing whilst another is denied it. The media must actively seek the truth on the public’s behalf, and be free to tell it as they see it.” Kofi Annan, former Secretary General of the United Nations (Accra, 6th December, 2007).

The 2012 Presidential elections in Ghana was the sixth democratic election to be held in the country since it returned to civilian rule in 1992. The election, described as a close contest between the ruling National Democratic Congress (Democratic Socialist) and the main opposition New Patriotic Party (Democratic Capitalist) by analysts and pollsters, ended up being challenged in court by the vanquished opposition. The election was unique because it was the first time that a sitting president and candidate for the ruling party had died barely four months before the elections, to be replaced by the then vice president, and also the first time that a presidential election result was challenged in court. Media coverage of the election campaign, and the ensuing legal battle was very prevalentin both local and international media, with the Electoral Commission admitting to accrediting the highest number of media houses and personnel for an election since 1992 (Gyimah-Boadi &

Brobbey, 2012). Eventually the Supreme Court of Ghana affirmed the results of the Electoral Commission in declaring Mr. John Mahama, of the National Democratic Congress the winner of the elections. Media plays an important role in democracy especially in this era of heavily mediatized politics. One key feature of democracy is elections and normatively the media is expected to provide the citizenry with requisite information for the evaluation of leaders and making of electoral choices (Strömbäck, 2008) in every democracy. It is based on this expectation that the media is usually assessed on how well they play their role of providing

(4)

the electorate with the information they require to assess their leaders and alternatives available to them. However, scholarship shows that commercial media ownership, an emphasis on horse-race-style news coverage, and an over-emphasis on conflict during

election times can be problematic for democracy (Dunaway, 2008). Such literature, however, focuses overwhelmingly on western democratic contexts. The Ghanaian political

communication environment, though vibrant, has been significantly understudied.

One notable exception to this isa study from 2002 positing that the media in Ghana played a critical role in the outcome of previous elections in Ghana; an assessment of the coverage of the 2000 election campaign, for example, pointed to a general positive tone and disproportionate amount of coverage in favour of the then- ruling National Democratic Congress (Temin & Smith, 2002). This was a period when there were few privately owned media organisations in the country. Since the repealing of criminal libel laws in July 2001, there has been a boon in the media industry with an influx of private media firms all over the country. The internet was also young and online news websites were not so crucial and common as they are today.It is worth exploring, therefore, whether these significant changes in the media landscape changed the media coverage of the 2012 election. Furthermore, this study aims to complement Temin & Smith’s (2002) analysis and speak to the communication science literature by incorporating other relevant features of campaign coverage—conflict framing, horse-race coverage, and a look at media ownership structure’s effects on content. To do so, the study analyses media content from two comprehensive news websites to find out how they portrayed the political actors in the election, the tone of their coverage and the focus of their coverage; whether an issue based coverage or a horserace coverage. The study seeks, therefore, to answer the following research question:

(5)

THEORETICAL FRAMEWORK

The role of media in democracy

In today’s ever changing technological world, the media’s ability to reach citizens far and wide has made it a key component of not only the information dissemination process but also the democratic system of every country (Pavlik, 2000). The growing influence of the media in the democratic dispensation of countries has seen them move to the centre of the social process in what is mostly referred to as the mediatization of politics (Strömbäck, 2008). Asp and Esaiasson (1996) define “the mediatization of politics as a three-stage process in which there is a development toward increasing media influence” (p. 80–81) As a result of the increasing mediatization of politics, the media has assumed a greater role and influence by shaping and framing the process and discourse of political communication as well as the society in which the communication takes place (Ferree, Gamson, Gerhard & Rucht, 2002). This situation has seen the media turning out to be the most important source of political information for the electorate, media practices being governed by media logic rather than political logic and political practices governed by media logic rather than political logic (Strömbäck, 2008).

The role of the media in a competitive democracy is profound. A competitive democratic model is defined as the “the democratic system which has an institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote” (Schumpeter, 1975, p. 269). It is often referred to as electoral democracy (Sartori, 1987) and elections are very important in this democratic model. In the competitive model of democracy, “citizens are expected to choose between competing political elites either retrospectively (the sanctional variant) or prospectively (the mandate variant). In order to do that people need information and

(6)

knowledge about important societal problems, about how society works, how the country is governed and by whom, and about differences between the political alternatives” (Strömbäck, 2005, p. 334).

This has led to a natural proclivity for a symbiotic relationship to develop between media and political parties. The media need exciting political campaigns in order to market their product and sell advertisements in their newspapers, television and radio broadcasts (Temin & Smith, 2002). Candidates on the other hand have come to the realisation that the media provides easy publicity and perhaps the most efficient means of getting their message out to the electorate to win their votes (Gyimah-Boadi, 1999). In effect, the primary function of the media in a democracy is to provide the citizenry with requisite information for the evaluation of leaders and making of electoral choices (Strömbäck, 2008). However recent studies and trends have shown the media is gradually shirking this responsibility in their coverage of presidential election campaigns (Kaplan, Goldstein & Hale, 2005).

Does ownership matter?

One reason the media may be seen to be shirking this informative responsibility relates to the fact that in the process of carrying out their duty in a democracy, media comes up against some external factors that influence their work. Some of these factors are the competitive media market, made up of advertisers and audiences as well as the ownership structure of media companies (McManus, 1995). Media ownership and structure plays an important role in how every news organization is run. A considerable amount of scholarly work has examined whether the institutional characteristics of media companies influence a particular set of company behaviours, as well as the volume, slant, and tone of coverage of the news by these media brands. The findings are mixed. Ownership structure can influence

(7)

the goals and operation of organizations, because the characteristics of any organisation’s ownership structure impacts its goals and behaviours (Miller, 1992; Moe, 1988).

Particularly relevant to this study are findings related to public versus private ownership of media companies. Findings from studies on corporate governance point to the fact that it is more difficult to monitor managers in publicly owned business corporations when compared to privately owned ones because of the different and diverse nature of shareholders. Large group of different shareholders cannot easily monitor or interfere in the daily operation of their firms (Agrawal & Knoeber, 1996; Hansmann 1988).

This logic is applicable to media companies too. In privately controlled media

organization, managers and subordinates need to be "responsive to the preferences of a single owner, which might be driven by motivations other than profit maximisation" (Dunaway, 2008, p. 1194). Bovitz, Druckman and Lupia (2002) have shown, for example, that single private owners can control the ideological direction of news coverage. In a typical business under private corporate ownership, a manager is retained to run the firm according to the directives of the owner (Chomsky, 2006). In summary, ownership structure can influence coverage because a single private owner as opposed to a corporate structure can more easily assert his or her preferences over the daily operation of the firm (Hamilton, 2004). On the other hand, with public corporate ownership there is more separation between ownership and control, and diverse shareholder owners are limited in their ability to direct and interfere with the management of the firm (Agrawal & Knoeber, 1996; Himmelberg, Hubbard & Palia, 1999).

Differing opinions like the RAND Corporation’s comprehensive review of research on the subject posit that media ownership barely plays a role, if at all, in affecting the content of newspapers (Baer, Geller, Grundfest, & Possner, 1974). Other recent studies could not find

(8)

any difference in the amount of public affairs programming produced by news organisations depending on the type of ownership under which they operate (Spavins, Denison, Roberts, & Frenette, 2002).

Studying a notable recent change in media ownership, McKnight (2010) argues that News Corporation is unique among media conglomerates in its commitment to Rupert Murdoch’s ideological beliefs, providing evidence that Murdoch is willing to let some of his newspapers lose great sums of money in the service of the promotion of his beliefs. Wagner and Collins (2014) posit that the Wall Street Journal’s editorial page is a very different place under Rupert Murdoch/ News Corporation’s ownership than it was under Bancroft family ownership. Government action, already addressed with a healthy skepticism from the Wall Street Journal’s Bancroft-era conservative editorial page, became even more negative in the Murdoch era.

Media ownership affects the goals and operations of media companies and there are owners who are willing to use their outlets to promote their beliefs even at the expense of profits in some cases. Media ownership influence in content is more likely to occur in a privately owned media company than a publicly owned one. This trend has a link to a situation in Ghana, where an owner of a media conglomerate has openly said and used his media conglomerate to promote his capitalist ideals and beliefs. This is worth taking a closer look at, since Ghana is a developing country with a media system different from the cases cited above.

Of media ownership and bias in news coverage

Media ownership is a critical component of journalistic work especially when it has the capacity to influence the direction of media coverage. Media bias is defined as "the perception that the media is reporting the news in a partial or prejudiced manner, to push a

(9)

specific viewpoint, rather than reporting the news objectively or when the media seems to ignore an important aspect of the story" (D’Álessio & Allen, 2000, p. 134). Williams (1975) argues that bias is prevalent only when it is volitional or willful; it is influential, must be threatening to widely held conventions and must be sustained rather than an isolated incident. An example of this kind of bias is partisan or ideological bias; bias as a result of a person’s ideological beliefs or partisan stance (Hofstetter, 1976) which meets all the criteria set by Williams. Partisan or ideological bias is mostly present in presidential election campaign coverage and must be seen as a genuine worry (D’Álessio & Allen, 2000). The period of election campaign coverage is seen as an ideal time frame within which to study traces and pattern of partisan or ideological bias. The period of an election campaign provides a suitable time period and relevant variables to look at how partisan or ideological beliefs manifest in the media coverage (D’Álessio & Allen, 2000). Journalism in most western democracies is populated with people who see themselves as being politically liberal (left leaning) than average (Schneider & Lewis, 1985). In practice, bias takes the form of one party being covered more extensively than others, being covered more negatively in tone than others or being covered more critically than others.

However, all the literature reviewed above cites examples from Western democratic and media systems. The democracies in the Western world are well developed with media systems different from young democracies in third world countries with different media systems (Hallin & Mancini, 2004). As a result, scholars still know very little about the relationship between media ownership, media bias and political election coverage in

developing countries. My study attempts to fill this gap by focusing on Ghana, a young but thriving democracy on the west coast of Africa. Before posing my first hypotheses, then, I will outline the relevant specifics of the Ghanaian case.

(10)

Ghana’s 2012 Elections as a case study

The role of the media in elections in developing countries in Africa is an important area often overlooked by academics (Temin & Smith, 2002). Unfortunately the existing academic literature on the subject is thin, considering the frequency with which elections are being held on the African continent. By looking at the situation in Ghana, my study

contributes knowledge about the media ownership and coverage of election campaigns in a developing country specifically one with a vibrant press system, and which is ranked 31st out of 202 countries and judged as "partly free" on the Freedom House 2016 index. Ghana unlike the United States has a very vibrant public broadcaster operating together in a common market with private broadcasters.

Ghana is a young thriving democracy in Africa which held its sixth presidential and parliamentary elections in December 2012 after returning to civilian rule in 1992. The elections, contested by seven political parties and an independent candidate were regarded as a close contest between John Mahama of the ruling National Democratic Congress (social democrats) and Nana Addo Dankwa Akuffo-Addo of the main opposition New Patriotic Party (liberal capitalist). The 2012 general elections were not only competitive in the political arena but also in the media arena. The Electoral Commission of Ghana admitted to

accrediting the highest number of media organizations across all spectrums and personnel ever in their history of organizing election since 1992 (Gyimah-Boadi & Brobbey, 2012), with the stakes very high, as the ruling party won the previous election in 2008 with a difference of only forty thousand votes. The ruling National Democratic Congress had lost their leader, candidate and sitting president John Atta Mills four months before the elections and had to replace him with the vice president John Mahama. Many electoral observers and posters had billed the election as a very close contest going into the polls (Gyimah-Boadi & Brobbey, 2012).

(11)

The high stakes nature of the election was such that vituperation during political discussions became common. Political talk shows mostly degenerated into accusations and shouting contests. The opposition New Patriotic Party believed they lost the 2008 election because of the National Democratic Congress’ strong arm tactics on the election day

(Gyimah-Boadi & Brobbey, 2012). As a result they called for extra vigilance this time round from their supporters. The usually calm Nana Akuffo-Addo called on his teeming supporters to be ready to "fight to the death." The National Democratic Congress also instituted a "Heroes Fund" meant to serve as compensation for party supporters or their relatives in the event they die or suffer bodily harm as a result of election violence (Gyimah-Boadi & Brobbey, 2012).

The modest beginnings of broadcasting in Ghana date back to the colonial era. In 1935 the then- Governor of the Gold Coast (as Ghana was then called), Sir Arnold Hodson established the first wired radio distribution system in Accra as an extension of what was, at that time, called the Empire Service from London. The Accra network, which was later named Station ZOY, started service on 31 July 1935, and is what is presently known as the Ghana Broadcasting Corporation (Ansah, 1986). The public broadcaster was the only medium of broadcast transmission until 1992 when the country had returned to civilian rule and the ban on private ownership of media had been lifted. Since then there has been a massive influx of privately owned media organizations across the country, a development many academics attribute to the repealing of the criminal libel laws in 2001(Temin & Smith, 2002).

Two of the biggest media conglomerates in the country are Multimedia Broadcasting Group and Despite Media. These are also the two outlets examined in this study, for reasons outlined below. The Multimedia Group is owned by a Harvard Business School trained entrepreneur, Kwasi Twum, who believes in motivating people and maximizing the potential

(12)

of the private sector in every economy (Multimedia Ghana, 2012). He has been running the media conglomerate for the past twenty years and is credited with owning the first private radio station (Joy FM) in the country. Twum’s liberal market leanings have seen him being accused consistently by the current Ghanaian government of serving the opposition New Patriotic Party who share similar liberal capitalist market ideologies, with his media brands (Awuni, 2015). He runs www.myjoyonline.com; a Web-based news channel, targeted at local and international audiences. During the 2000, 2004 and 2008, 2012 elections, it was the undisputed point of reference for the campaign and election results (Multimedia Ghana, 2015).

Despite Media Group was established in 1999. Peacefmonline.com is part of the Despite Group media brands including UTV, Hello FM, Okay FM, Neat FM and Peace FM. It is the English news media website of a local language (Twi) broadcasting media group aimed at both the young and old and literate as well as illiterate populace. Their objective is to demonstrate singular support for the local music industry which was on a rapid decline when they began operations, put Ghanaian culture above all else and use over 70 per cent airtime to broadcast programmes in Twi. They have evolved into a content company, the home of Ghana's Most Popular programming and a multimedia leader with a solid presence in the country (Peacefmonline, 2016). It belongs to Dr. Osei Kwame a self-confessed socially oriented philanthropic billionaire business man. Because of his love for socialist interventions and perceived business connections within the ruling democratic socialist National Democratic Congress, he has being accused of using his brands to aid them and portraying the opposition New Patriotic Party negatively (Allotey, 2016).

The owner of Myjoyonline.com has strong capitalist beliefs and his media empire espouses capitalist ideals. Coupled with McKnight (2010)’s assertion that media owners use their media brands to pursue their ideological beliefs even at cost to profits, I propose;

(13)

H1: Myjoyonline.com was more favourable in tone to the New Patriotic Party in their coverage as compared to Peacefmonline.com because of the owner’s capitalist leanings.

Wagner and Collins (2014) also assert that media owner’s leanings and political beliefs are reflective on how critical they are or otherwise of certain governments and their policies. As a result I propose;

H2: Myjoyonline.com were more critical of the ruling National Democratic Congress in their coverage as compared to Peacefmonline.com.

H3: Myjoyonline.com’s coverage of the campaign focused on more capitalist related topics as compared to that of Peacefmonline.com.

Horserace coverage of presidential election campaigns.

Two other critiques of news media performance in western democratic elections may also be relevant to the Ghanaian case, and merit investigation. The first is the focus on “horse race” coverage. Horse race coverage is media coverage which focuses primarily on strategies campaigns are employing or who is winning or losing in the polls (Graber, 2002). Western campaign news increasingly focuses more on the "game frame" or horserace news coverage (discussing who’s ahead, who’s behind, and campaign strategies and events) than on substantive policy issues (Graber, 2002). This is as a result of audience preferences for such kind of news on elections as compared to policy issues. The audience will look out for such news regardless of where it is in a particular news bulletin or the page it is in, in a newspaper (Andersen & Thorson, 1989) and find it more interesting than issues of policy covered in these election campaigns(Iyengar, Norpoth & Hahn, 2004).

Aware of the audience interest and preferences and their quest to maximize profits, media owners increasingly urge and fashion strategies and targets around coverage that

(14)

audience prefer so that they could cash in (Wagner 2007). The situation in Western democracies might not necessarily be the case in third world countries, but with the increasing ‘’Americanization’’ of politics and democracy in the world (Negrine & Papathanassopoulos, 1996), and the ownership structure of the media outlets under examination here, it is worth examining the situation in Ghana.

Based on the trend from Western democracies and media systems cited in the

literature reviewed and the fact the chosen outlets are commercial media organizations which seek to maximize profit by satisfying audience’s needs and preferences, I propose;

H4: Peacefmonline.com and Myjoyonline.com covered the election campaign in a more horse race than issue-based manner.

Conflict Framing

The second criticism on the media's coverage of election campaign worth examining is the use of conflict framing. Framing is a widely used concept in the literature (e.g., Entman 1993; Iyenegar 1991; Grabber 1993; Zaller 1992). In spite of the large volumes of literature available on framing, there is not a single definition for framing.

News frames have been identified as ‘’conceptual tools which media and individuals rely on to convey, interpret and evaluate information (Neuman, Just, & Crigler, 1992, p. 60). They set the parameters ‘’in which citizens discuss public events’’ (Tuchman, 1978, p. 6). It is also seen to be selecting "some aspects of perceived reality" to enhance their salience "in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation" (Entman, 1993, p. 53). Frames tend to help audiences "locate, perceive, identify and label" the information flow around them (Goffman 1974, p. 21) and to "narrow the available political alternatives" (Tuchman, 1978, p. 156). In

(15)

summary, frames come in the form of generic or issue specific frames. Generic news frames are general and not confined to a specific issue whilst issue specific frames are media frames in specific news stories, for example the Iraq War in 2003 was an issue specific frame of war on terror up against occupation. There is quite a number of generic news frames used in the news media and one of such is the conflict frame. Semetko and Valkenburg (2000) define conflict frame as a frame "that emphasizes conflict between, individuals, groups or

institutions as a means of capturing audience interest" (p. 95). In their work on news frames used by the United States news media; Neuman et al. (1992) discovered that the American news media rely on a few key and central frames in reporting a wide range of issues and the conflict frame was the most commonly used. In Europe, Dutch news media also used conflict frames. "They occurred most often in the serious news outlets in the press and also where there was more political news" (Semetko & Valkenburg, 2000, p. 106).

Available literature argues that the discussion between political actors in the news often reduce long substantive issue based debates to an overly simplistic conflict. Presidential election campaign news, for instance, is framed largely in terms of conflict (Patterson 1993). There exist conflicts between governments and opposition parties in the news (Semetko & Valkenburg, 2000). The news media have received their fair share of criticisms for their over emphasis on conflict in their news coverage of political events and actors. They have been criticized for inducing public cynicism and mistrust of political leaders (Capella & Jamieson, 1996). Like other jurisdictions, I expect the two privately owned news media outlets to cover the presidential election campaign largely in a conflict frame to attract the attention of

citizens. I propose;

H5: Myjoyonline.com and Peacefmonline.com covered the 2012 Presidential Election in a conflict or contested frame.

(16)

METHODOLOGY

Research Design

To study the news coverage of the 2012 general election campaign in Ghana, a quantitative content analysis was carried out on the two leading news websites in Ghana based on their visitor traffic statistics from Alexa’s Actionable Analytics for the Webdata (National Media Commission, 2014). Content analysis allowed me to assess the tone of coverage towards each candidate/party over time, as well as which sources were quoted and what level of disagreement was present in the media coverage of the election campaign. Content analysis is a reliable method for observing such patterns because “it is a technique which aims at describing, with optimum objectivity, precision, and generality, what is said on a given subject in a given place at a given time” (Lasswell, Lerner & de Sola Pool,1952, p. 34) .

The selected websites; (Myjoyonline.com and Peacefmonline.com) serve as the official online versions for the two largest media groups in Ghana. These two websites were chosen because of the traffic; six thousand three hundred (6,300) and three thousand four hundred and twelve (3,412) average unique visitors per day for Myjoyonline.com and Peacefmonline.com respectively, shows how comprehensive they are (National Media Commission, 2014). The period of focus in the study was the last twelve weeks of the 2012 election campaign, a period spanning 1st September 2012 (the date by which all contesting parties and independent candidates were officially registered with the Electoral Commission of Ghana (Gyimah-Boadi & Brobbey, 2012)) to the 6th of December, 2012, the last day of campaigning in the election which was held on 7th December.

The sample was gathered in multiple stages. In the first stage all articles that were archived in the “politics" section of these websites were selected. Using keywords such as the

(17)

names of all candidates and contesting political parties, “2012 elections” and key policy phrases eg. (“free SHS”) in the search terms,; articles that spoke about the political parties and their representatives that fell within the period under consideration were selectedfrom that archive. The units of analysis in the study were the individual stories selected for the study. The focus was on all news stories talking about the political parties and candidates in the 2012 electioneering campaign. I selected 130 articles that spoke about the elections from Myjoyonline.com but took out 21 after an audit as they focused on the Electoral Commission and had very little on the parties or their representatives instead. It was a similar situation with Peacefmonline.com, where i selected 320 but had to settle on 250 as 70 focused on the Electoral Commission instead of the parties or their representatives. In the end the size of the sample was 359 news articles, made up of 109 articles from Myjoyonline.com and 250 from Peacefmonline.com. The average length of a news article was three hundred and seventy (370) words.

My study encountered challenges of gaps as one of the outlets myjoyonline.com had very few articles published on the election campaign in September available with over 85% of their sample coming from October, November and December. Even though the gaps existed it did not affect my study as the outlet's coverage of the campaign picked up in the last week of September. The first three weeks of the campaign coverage focused on the Electoral Commission and three other political parties who had been disqualified from contesting the elections for failing to meet some statutory requirements. They challenged this decision in court and it was not until the 20th of September that the court ruled to uphold the decision of the Electoral Commission.

(18)

Coding

To effectively operationalize and test the hypotheses, a codebook was developed based on relevant literature and a close reading of a subset of the texts. The first variable in the study codebook was the visibility of political parties and or their representatives in the story.For the articles selected, the headline and lead or opening paragraph were coded sentence by sentence to find the visibility of political parties and or their representatives. Representatives were classified as presidential candidates including independent candidates, parliamentary candidates, political party officials and the political party itself.

The first three actors were coded per news story because the focus was on which actors were most visible from the start of each news article. To assess the tone of coverage, any slight hint of positive and negative portrayal within the story were coded for. Any form of critical references to policy statements and performance in government of political parties and actions of their members or representatives respectively were coded for. The

measurement was done based on how the news article portrayed the political party or its representative in the news story. For example, through their actions or policy statements were they characterized as the wrongdoers or the right people (Brodie, Brady & Altman, 1998)?

Next, up to five sources were coded for in each news story, classified between as belonging to one of various political or non-political categories. Additional codes measured the presence of ruling and opposition party sources in the same story, whether there was mention of a direct conflict of (policy) opinion in the story and the overall balance of tone and actor visibility present in the story. Finally, codes were designed to distinguish between the reference of policy information versus more strategic/"horse-race” coverage (Iyengar et al, 2004). A second coder coded a randomly chosen 10% of the texts, and the results of the intercoder reliability test showed that most variables achieved satisfactory reliability on the Scott’s Pi measure (at least .7) . A few variables had values between .6 and .7: and a few

(19)

variables did not meet satisfactory levels (e.g. graphics or illustration to show polls leader). Therefore, in my analysis i did not use those variables that did not meet satisfactory levels and interpreted those with satisfactory levels (.6 and .7) with caution.

A full copy of the codebook is available in the appendix as well as a table of the satisfactory variables used.

RESULTS

Before starting with the hypotheses, it is interesting to see a bit of overview of the coverage. In the twelve-week campaign period the two news websites had visible coverage of the political parties and candidates in the 2012 presidential election. Out of 359 news articles studied, at least one of these candidates and/or their political parties were mentioned in the headline 357 times making up 99.4% of the sample. The most visible parties mentioned in the headlines were: National Democratic Congress represented by Mr. John Mahama (mentioned in 42.9% of the headlines); New Patriotic Party represented by Nana Akuffo-Addo (41.5%); Progressive Peoples Party represented by Dr. Papa Kwasi Nduom (5.3%); Peoples National Convention represented by Mr. Hassan Ayariga (5.3%); and Convention Peoples Party

represented by Dr. Abu Sakara Foster 2.2%. The two main political parties in the election the ruling National Democratic Congress and the main opposition New Patriotic Party (the

opposition party with the most seats in parliament) made up 84% of the total political party mentions in the headlines of the news articles studied. Since Ghana reverted to civilian rule these two political have both served as governing party twice, with both winning all the elections that have been held in the Fourth Republican constitution promulgated in 1992.

Visibility is also indicated by the sources quoted in the coverage. A varied range of sources were used by these news websites during their coverage of the election campaign, although 84.6% of these sources were from the contesting political parties and 16.4% from no

(20)

political parties. The most sourced political parties during the election coverage were the New Patriotic Party(sourced in 39.8% of articles), followed by the National Democratic Congress (30.4%); again the remaining parties were much less visible in coverage or sourcing, with the Progressive Peoples Party at 6.4%, Peoples National Convention at 3.9% and the Conventions Peoples Party at 1.7%. Again, here the ruling National Democratic Congress and the main opposition New Patriotic Party were the most dominant. They combined to make up 71.3% of the 84.6% of the political party sources used by the media. On the other hand the sources with no political parties ranged from foreign politicians (e.g. diplomats) 0.3%, official public sources (e.g. police sources) 1.4%, official private sources (e.g. think tanks) 11.7%, media sources 1.9%, citizens 0.3% and celebrities 0.6%. Most of the sources used were elite sources (versus everyday citizens).

As a final look at the general coverage, it is interesting to see what policy areas were discussed in connection with the election. Unlike party visibility, there was no clear dominant policy discussion in the run-up to the election. The most visible policy issue that

characterized the coverage of the election campaign was education (mentioned in 15% of stories), followed by security (8.9%), economy (5%), infrastructure (5%) and health (1.7%). We can see that 65% of stories contained some other policy issue, but the discussions ranged quite widely.

Turning now to the hypotheses, the first hypothesis predicted Myjoyonline.com because of the owners’ capitalist leanings was going to be more favourable in tone to the economically liberal New Patriotic Party in their coverage as compared to

Peacefmonline.com. To assess this, I ran an independent sample t-test comparing each outlet’s balance toward the NPP on a five-point scale, where higher scores indicate more positivity. The results show that there was a marginally significant difference between outlets, but in the opposite direction as expected: Myjoyonline.com had a mean positivity

(21)

toward the NPP of 3.26 (SD=1.5), whereas Peacefmonline.com had a mean positivity toward the NPP of 3.66 (SD=1.7) (t=1.79, df=254, p=.074). This indicates that Peacefmonline.com was actually more positive toward the NPP than Myjoyonline.com, although the difference is not very big and is only marginally significant. A Chi-square showed that there was a

significant difference in the presence of positive portrayals of the New Patriotic Party by the two outlets (χ2= 4.26, df = 1, p<.05): 58.1% of Myjoyonline.com articles contained a positive portrayal, whereas 71.4% of Peacefmonline.com articles did so. Another Chi-square showed that there was a significant difference in the presence of negative portrayals of the New Patriotic Party by the two outlets(χ2= 3.77, df = 1, p=.052): 56.8% of the articles from Myjoyonline.com showed negative portrayals of the New Patriotic Party as compared to 43.4% from Peacefmonline.com This refutes H1.

The second hypothesis is a counterpart to H1, and predicted that Myjoyonline.com would be more critical of the ruling National Democratic Congress in their coverage as compared to Peacefmonline.com. Again, I ran an independent sample t-test to assess this, comparing each outlet on its mean level of criticism of the NDC on a scale of 1-3, where higher scores indicate more criticism in an article. The mean score for Myjoyonline.com was 1.74 (SD=.76) and for Peacefmonline.com was 2.0 (SD=.89). This difference was also statistically significant (t=2.67, df=179, p=.008). Again, this difference is surprising and in the opposite direction of what was predicted; as a result, H2 is also rejected.

The third hypothesis predicted that Myjoyonline.com’s coverage of the campaign would focus on more capitalist- related topics as compared to that of Peacefmonline.com. A Chi-square showed a marginally significant difference between Myjoyonline.com and Peacefmonline.com’s inclusion of capitalist- related topics (e.g. increasing the scope of the private sector development fund, privatization of the national electricity company and increased private public partnership in the housing and infrastructure acceleration program);

(22)

(χ2= 2.15, df=1, p=.07). 8.3% of the articles from Myjoyonline.com covered capitalist related topics whilst 4.4% of those from Peacefmonline.com. Although both news outlets only covered these topics rarely in the coverage, we can see that Myjoyonline.com did so twice as often as Peacefmonline.com. This result supports H3.

The fourth hypothesis predicted that both Peacefmonline.com and Myjoyonline.com would cover the election campaign in more of a horserace manner because they are two commercial outlets owned by a private investor with a commitment to maximizing profits by satisfying audience’s needs and preferences. Horserace coverage (and its counterpart, issue-based coverage) manifested in three ways in the sample: the use of quotes from

manifestos/campaign speeches (taken to indicate more issue-based coverage); the presence of discussion of campaign strategies/campaign teams in the news articles; and the presence of discussion of poll standings in the news articles. Because these variables are all independent indications of horse-race coverage (or its counterpart), and were not expected to perform as a scale, they were analyzed separately. A Chi-square comparing the use of quotes from

manifestos/campaign speeches showed significant difference between the two outlets (χ2=12.9, df=1, p<.001): 39.4% of the articles from Myjoyonline.com had quotes directly from campaign speeches or manifestos whilst that of Peacefmonline.com was 60.0%. This was contrary to what was expected. Another Chi-square comparing the presence of the coverage of news on campaign strategies or the campaign team showed a non-significant difference (χ2= 1.60, df=1, p= .20): 23.9% of the articles from Myjoyonline.com covered news on campaign strategies or campaign team of the candidates in the election whilst 30.4% of the articles from Peacefmonline.com did same. A third Chi-square comparing the coverage of performance in polls by the two outlets showed a non- significant difference between the outlets (χ2= 1.10, df=1.20, p=.300): 5.5% of the articles from Myjoyonline.com covered candidates performance in polls whilst 3.2% from Peacefmonline.com covered it. Thus we

(23)

see that for these three horserace variables, the only significant difference emerged on the use of quotes from campaign speeches or manifestos.

One final variable helped measure the balance, per article, between horserace and issue coverage. This variable indicated whether an article was entirely issue-based, mixed, or entirely horserace-based in its coverage of the election. 5.5% of the articles from

Myjoyonline.com were entirely horserace in coverage as opposed to only 2.8% from Peacefmonline.com. This result was statistically significant (χ2= 10.72, df=3, p= .013). Coupled with the results from the preceding paragraph, this result rejects H4.

The fifth and final hypothesis predicted that both Myjoyonline.com and

Peacefmonline.com covered the 2012 Presidential Election in a conflict or contested frame, because of the existence of a trend of most presidential election campaign news being framed in terms of conflict. Articles were coded for whether they contained explicitly conflicting views in their coverage, as well as whether they contained government and opposition sources. I initially ran a correlation to see if the presence of government and opposing sources in an article was correlated with the explicit presence of conflicting views in an article. This was the case; the Pearson’s correlation was r=.842 (N=358, p<.001) showing positive correlation between them. As a result, I recoded the two variables into a single variable where 1=yes; conflict framing present and 0= no; conflict framing absent. A Chi-square comparing the presence of conflicting views and frames in their coverage of the election showed a non-significant difference between the two outlets (χ2= 5.30, df=1, p.= 0.21): 16.7% of the articles from Myjoyonline.com and 8.4% for Peacefmonline.com. The two outlets are not significantly different from one another, but neither outlet included very many conflict frames/conflicting views, because less than 1 in 5 articles contained such ideas. This suggests that the coverage wasn’t as conflict-based as I expected overall. So again H5 is rejected.

(24)

DISCUSSION AND CONCLUSION

The goal of this study was to assess the Ghanaian news media coverage of the 2012 presidential election campaign by comparing the coverage of two online news websites. The study specifically looked at ownership and whether it is related to biased coverage, horserace coverage and conflict framing of the elections by the news media. The results presented above revealed some interesting differences as well as similarities.

The results showed that the Ghanaian news media fulfilled their primary normative function of providing the citizenry with requisite information for the evaluation of leaders and making of electoral choices (Strömbäck, 2008) in their young democracy. 68.8% of the articles from Myjoyonline.com were entirely issue based focusing on the policy issues that were topical during the election whilst that of Peacefmonline.com was 74.4%. By covering the election in that manner it is assumed that the media provided the citizens with the information they required. This is a healthy development for both the media and the Ghanaian society as the citizens are empowered by the media to be aware of their options, discuss them and make the right choices between political elites for themselves and their country. The two media outlets, rather than emphasizing horserace coverage and strategy, often- sourced directly from the campaign speeches of candidates and manifestos of political parties during their coverage of the election. By doing this they served as the conduit between the citizens and the political elite. In the process of doing this they assume a greater role and influence, by shaping and framing the process and discourse of political communication as well as the Ghanaian society (Ferree, Gamson, Gerhard & Rucht, 2002). The media through the increasing mediatization of politics shape up the discourse of political communication and by covering the elections in an issue based manner, the Ghanaian news media impliedly assumed a greater role and influence in the political communication process during the election campaign in 2012.

(25)

This is quite remarkable because we are in an era where campaign news at least in Western societies increasingly focuses on the "game frame" (Graber, 2002), as well as the era of increased "Americanisation" of democracy (Negrine & Papathanassopoulos, 1996); a period where media focus on candidates, their performance in polls, their campaign team and strategies, as it happens in the American democratic system is being imported to other

developing democracies in third world countries. So in general, one might expect

commercialization of the media (Iyengar et al, 2004)and the resulting emphasis on audience tastes and preferences for horserace news (Andersen & Thorson, 1989), to drive these two commercial outlets to focus on more horserace coverage than issue coverage. Quiet

remarkably the Ghanaian news media covered the 2012 election in a more issue frame manner than horserace.

The results point to the fact that contrary to widely held misconceptions in Ghana (Tettey-Alarti, 2011), Myjoyonline.com is not positively biased towards the New Patriotic Party because of the owner’s capitalist leanings. The ruling National Democratic Congress’ officials and some members have frequently used Peacefmonline.com to criticize

Myjoyonline.com, threatening to boycott their platform for propagating the opposition’s agenda (Awuni, 2015). The opposition New Patriotic Party also recently served notice to boycott all media platforms of the Despite Media group (owners of Peacefmonline.com) because of biased coverage as a result of the owner’s business connections with government and some government officials (Allotey, 2016). The results directly refute these accusations, showing that, at least during the run-up to the 2012 election, Myjoynline.com was less positive and more negative towards the NPP compared to Peacefmonline.com. Both outlets were critical of the National Democratic Congress with Peacefmonline.com being even more critical. In effect, results suggest that the ownership of these two outlets barely played a role, if at all, in affecting the content of the news websites. This is arguably “good news” for

(26)

scholars who are concerned about the role of ownership in affecting and biasing news

coverage (Miller, 1992; Moe, 1988; Dunaway, 2008; McKnight, 2010). It seems that, at least in Ghana in 2012, ownership does not predict biased or overly horserace-focused coverage of politics.

These findings may also be reflective of the conditions the Ghanaian citizens were in at that time. Ghana was in a serious economic and energy crisis at the time of the elections (Gyimah-Boadi & Brobbey, 2012). The year 2012 saw a sustained period of power outages which affected businesses, a fast shrinking economy and increased cost of living leading to hardships for many citizens. The election was a high-stakes one (Gyimah-Boadi & Brobbey, 2012), and for the press to have focused on the issues and be critical of the then regime is an indicator of how well the media played their watchdog role in the democracy (Sartori, 1987).

However, the picture is not only positive; such critical coverage could also potentially harm Ghana’s young democracy; scholarship suggests that coverage which tends to be critical of political leaders and political systems induces public cynicism and mistrust of political leaders (Cappella & Jamieson, 1996). These attitudes, in turn, can lead to low political knowledge, voter disgust, and a lack of political participation (Cappella & Jamieson, 1996).

Both news outlets were expected to cover the election campaign in a conflict frame, conforming to the trend of most presidential election campaign news being framed in terms of conflict (Patterson, 1993). However, both outlets had less than 20% of the news articles studied having the conflict frame in their coverage. This is very encouraging for Ghana’s young democracy and stability. Elections in Ghana is mostly a tribal or religious affair, (Gyimah-Boadi, 2009) something which can easily inflame passions and trigger violence. As a country in a sub-region with notable cases of electoral violence (La Cote D’Ivoire) and civil

(27)

strife (Liberia), higher coverage of election campaigns in a conflict frame and politics as a zero sum game has the potential to erode all the modest gains made through a post electoral conflict.

Another interesting pattern, but also more concerning for Ghana, is that the two main political parties absolutely dominated overage in terms of visibility. This suggests that additional parties that have not been in power have very little voice or opportunity in the media to get their message across. This is concerning from the standpoint of democracy campaigners as the country is increasingly leaning to a two party state system. It is either the National Democratic Congress or the New Patriotic Party. Citizens are restricted to two options even though the country is not a two party state. This is increasingly worrying as the trend of media coverage points to the strengthening of the two main parties as against the five other parties.

Even though the results may be satisfactory overall, from a normative perspective, this study has its limitations. Assessing the role of ownership and biased coverage, horserace coverage and conflict framing in presidential election campaigns by studying two commercial outlets limits the strength of the study. Comparing commercially owned outlets to publicly or government-owned outlets would be an important next step to broaden the findings.

Additionally, some of the variables, as reviewed in the methods section, had low values on intercoder reliability. This means we should interpret some of this data cautiously. Finally, by focusing on only one election period, the study was limited as Ghana has had four successful elections since the boon of the internet and three elections since criminal libel laws were repealed. Comparing the media coverage over two or three election campaign period could have strengthened the study.

(28)

In light of these limitations, I recommend further research to look into more media coverage of elections in Africa in general and Ghana in particular by comparing coverage of commercial broadcasters and public/state broadcasters. A framework should be developed theoretically and empirically to propose grounded hypotheses that take into account the political system, party system, media system and the journalistic norms and values in these third world countries with young democracies and media systems different to that of the developed Western democracies. There is always difficulty when applying western

theoretical frameworks to non-western cases (Strömbäck & Dimitrova, 2006); more studies need to be done to compare election coverage as they would serve as an extension of the existing empirical database and lay the foundation for further theorizing and framework for comparing election coverage by media outlets in these developing countries.

In general my study found out the Ghanaian news media coverage of the 2012 presidential campaign was dominated by two main political parties; the ruling National Democratic Congress and the main opposition New Patriotic Party. They were the most visible and sourced political parties during the election campaign coverage. There is good news for the Ghanaian media and democracy as contrary to trends in Western democracies, the Ghanaian news media coverage of the election campaign focused on issues rather than horserace. No single issue dominated the election coverage as a wide variety of policy areas were covered by the media. Ownership structure or owners beliefs did not influence media coverage of the election. Outlets were not biased in their coverage contrary to expectations and as well as it had been publicly claimed.

(29)

REFERENCES

Agrawal, A., & Knoeber, C. R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31(03), 377-397.

Alarti-Amoako, F. T. (2011, October 3). NDC Boycotts Joy FM. Retrieved from

https://www.modernghana.com/news/353752/ndc-boycotts-joy-fm.html

Allotey, G. A. (2016, April 3). NPP boycotts Okay, Neat FM. Retrieved from http://citifmonline.com/2016/04/03/npp-boycotts-okay-neat-fm/

Andersen, K., & Thorson, S. J. (1989). Public discourse or strategic game? Changes in our conception of elections. Studies in American Political Development, 3, 262-278.

Annan, K. (2000) ‘Africa’s thirst for democracy’, Evening News, December 6, p. 7.

Ansah, P.A.V. (1986) ‘Broadcasting and Multilingualism’, in G. Wedell (ed.)

Making Broadcasting Useful: The African Experience.

Asp, K., &Esaiasson, P. (1996). The modernization of Swedish campaigns: Individualization, professionalization, and medialization. Politics, media, and modern democracy: An

international study of innovations in electoral campaigning and their consequences, 73-90.

Awuni, M. A. (2015, April 22). Hanna Tetteh, Joy FM and the hounds. Retrieved from http://m.myjoyonline.com/marticles/opinion/manassehs-folder-hanna-tetteh-joy-fm-and-the-hounds-

Baer, W. S., Geller, H., Grundfest, J. A., & Possner, K. B. (1974). Concentration of Mass Media Ownership: Assessing the State of Current Knowledge.

(30)

Bovitz, G. L., Druckman, J. N., & Lupia, A. (2002). When Can a News Organization Lead Public Opinion?–Ideology Versus Market Forces in Decisions to Make News. Public Choice, 113(1), 127-155.

Brodie, M., Brady, L. A., & Altman, D. E. (1998). Media coverage of managed care: is there a negative bias?. Health Affairs, 17(1), 9-25.

Cappella, J. N., & Jamieson, K. H. (1996). News frames, political cynicism, and media cynicism. The Annals of the American Academy of Political and Social Science, 71-84.

Chomsky, D. (2006). “An interested reader”: Measuring ownership control at the New York Times. Critical Studies in Media Communication, 23(01), 1-18.

D'Alessio, D., & Allen, M. (2000). Media bias in presidential elections: A meta‐analysis. Journal of Communication, 50(4), 133-156.

Dunaway, J. (2008). Markets, ownership, and the quality of campaign news coverage. The Journal of Politics, 70(04), 1193-1202.

Freedom of the Press 2016 Table of Scores. (2016, February 12). Retrieved from https://freedomhouse.org/report/freedom-press-2016/table-country-scores-fotp-2016

Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58.

Ferree, M. M., Gamson, W., Gerhard, J., & Rucht, D. (2002). Four models of the public sphere in modern democracies. Theory and Society, 31(3), 289-324.

Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Harvard University Press.

(31)

Gyimah-Boadi, E. (2009). Another step forward for Ghana. Journal of Democracy, 20(2), 138-152.

Gyimah-Boadi, E., & Brobbey, V. (2012). Countries at the Crossroads: Ghana. Freedom House.

Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and politics. Cambridge University Press.

Hamilton, J. (2004). All the news that's fit to sell: How the market transforms information into news. Princeton University Press.

Hansmann, H. (1988). Ownership of the Firm. Journal of Law, Economics, & Organization, 4(2), 267-304.

Himmelberg, C. P., Hubbard, R. G., & Palia, D. (1999). Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics, 53(3), 353-384.

Hofstetter, C. R. (1979). Perceptions on news bias in 1972 presidential campaign. Journalism and Mass Communication Quarterly, 56(2), 370.

Iyengar, S. (1994). Is anyone responsible?: How television frames political issues. University of Chicago Press.

Iyengar, S., Norpoth, H., & Hahn, K. S. (2004). Consumer demand for election news: The horserace sells. Journal of Politics, 66(1), 157-175.

Kaplan, M., Goldstein, K., & Hale, M. (2005). Local news coverage of the 2004 campaigns: An analysis of nightly news broadcasts. A report for the Lear Center Local News Archive, A project of the USC Annenberg School and the University of Wisconsin.

(32)

Lasswell, H. D., Lerner, D., & de Sola Pool, I. (1952). The comparative study of symbols: An introduction (No. 1). Stanford University Press.

Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common knowledge: News and the construction of political meaning. University of Chicago Press.

McKnight, D. (2010). “Rupert Murdoch’s News Corporation: A Media Institution with A Mission.” Historical Journal of Film, Radio and Television 30 (3): 303–316.

McManus, J. (1995). A market‐based model of news production.Communication Theory, 5(4), 301-338.

Miller, G. J. (1992). Managerial Dilemmas: The Political Economy of Hierarchy. New York: Cambridge University Press.

Moe, T. (1988). ‘‘The New Economics of Organization.’’ American Journal of Political Science 28 (4): 739–77.

Multimedia Group Limited (2012, February 6). Retrieved from http://www.multimediaghana.com/about.html

Negrine, R., & Papathanassopoulos, S. (1996). The “Americanization” of Political

Communication A Critique. The Harvard International Journal of Press/Politics, 1(2), 45-62.

Page, B. I., & Shapiro, R. Y. (2010). The rational public: Fifty years of trends in Americans' policy preferences. University of Chicago Press.

Patterson, T. (1993) Out of order, New York: Knopf.

(33)

Peacefmonline all about us (2016, May 15). Retrieved from http://www.peacefmonline.com/services/about/

Sartori, G. (1987). The Theory of Democracy Revisited Part One: The Contemporary Debate, Vol. 1 and Part Two: The Classical Issues, Vol. 2. New Jersey: Chatham House.

Schneider, W., & Lewis, I. A. (1985). Views on the news. Public Opinion,8(4), 6-11

Schumpeter, J. A. (2013). Capitalism, socialism and democracy. Routledge.

Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication,50(2), 93-109.

Strömbäck, J. (2004). Den medialiseradedemokratin, om journalistikensideal, verklighet och makt.

Strömbäck, J. (2005). In Search of a Standard: four models of democracy and their normative implications for journalism. Journalism Studies, 6(3), 331-345.

Strömbäck, J. (2008). Four phases of mediatization: An analysis of the mediatization of politics. The International Journal of Press/Politics, 13(3), 228-246.

Spavins, T. C., Denison, L., Roberts, S., & Frenette, J. (2002). The measurement of local television news and public affairs programs. Washington, DC: Federal Communications Commission.

Temin, J., and Smith, D., A. (2002). Media Matters: Evaluating the role of the media in Ghana’s 2000 elections, African Affairs, p. 585-605.

(34)

Wagner, M. W. (2007, October). The utility of staying on message: Competing partisan frames and public awareness of elite differences on political issues. In The Forum (Vol. 5, No. 3).

Wagner, M. W. & Collins, T. P, (2014). Does Ownership Matter?, Journalism Practice, 8:6, 758-771,

Williams, A. (1975). Unbiased study of television news bias. Journal of Communication, 25(4), 190-199.

Zaller, J. (1992). The nature and origins of mass opinion. Cambridge University Press.

APPENDICES

Appendix A

CODEBOOK

H1 Myjoyonline.com were more favourable in tone to the opposition party and more critical of the ruling party in their coverage as compared to Peacefmonline.com because of the owner’s capitalist leanings.

H2 Myjoyonline.com’s coverage of the campaign focused on more capitalist or free market related topics as compared to that of Peacefmonline.com

H3 Peacefmonline.com and Myjoyonline.com covered the election campaign in a more horse race than issue-based manner.

H4 Myjoyonline.com and Peacefmonline.com covered the election in a contention or contested frame

(35)

ID0 Coder: Your first name

………

V01: The source of the article

1. Myjoyonline.com

2. Peacefmonline.com

V02: The headline of the article

………..

V03: Date article was published

………

V04: Is any political party or an actor of a political party mentioned in the headline of the article?

1. Yes

2. No

V05a: If yes in V04, code the party/affiliation of the first political party/actor mentioned in the headline

1. NPP/Representative of NPP

2. NDC/Representative of NDC

3. CPP/Representative of CPP

(36)

5. PPP/Representative of PPP

6. GCPP/Representative of GCPP

7. UFP/Representative of UFP

8. Independent Candidate

9. Other:………..

V05b: If yes in V04, and more than one actor is mentioned, code the party/affiliation of the second political party/actor mentioned in the headline

1. NPP/Representative of NPP 2. NDC/Representative of NDC 3. CPP/Representative of CPP 4. PNC/Representative of PNC 5. PPP/Representative of PPP 6. GCPP/Representative of GCPP 7. UFP/Representative of UFP 8. Independent Candidate 9. Other:………..

(37)

V05c: If yes in V04, and more than two actors are mentioned, code the party/affiliation of the third political party/actor mentioned in the headline

1. NPP/Representative of NPP 2. NDC/Representative of NDC 3. CPP/Representative of CPP 4. PNC/Representative of PNC 5. PPP/Representative of PPP 6. GCPP/Representative of GCPP 7. UFP/Representative of UFP 8. Independent Candidate 9. Other:………..

Now read the full article. The following variables pertain to the full article, not just the headline (but the headline can be included as well).

V06a: Is the NPP/ NPP representative positively portrayed at all in the story?

To find out whether the political party or actor is positively portrayed in the story, we look out for whether the story itself is a positive story or one that puts the party or actor in good light—even momentarily. This includes the role they play in the story and how they are essentially characterized as good. Words like caring, sensitive, responsible, proven,

concerned, etc. should provide an indicator of how they are portrayed by the news story. A positive portrayal can be just a short portrayal in the news story; don’t worry about overall

(38)

balance yet (see variable 08 for that), but just code whether there is ANY positive portrayal of the actor in the story.

1. Yes there is a positive portrayal present somewhere in the article.

2. No there is no positive portrayal present in the article.

99. The NPP/NPP representative is not portrayed at all in the story.

V06b: Is the NDC/ NDC representative positively portrayed at all in the story?

99. The NDC/NDC representative is not portrayed at all in the story.

V06c: Is the CPP/ CPP representative positively portrayed at all in the story?

99. The CPP/CPP representative is not portrayed at all in the story.

V06d: Is the PNC/PNC representative positively portrayed at all in the story?

(39)

99. The PNC/PNC representative is not portrayed at all in the story.

V06e: Is the PPP/PPP representative positively portrayed at all in the story?

99. The PPP/PPP representative is not portrayed at all in the story.

V06f: Is the GCPP/GCPP representative positively portrayed at all?

99. The GCPP/GCPP representative is not portrayed at all in the story.

V06g: Is the UFP/UFP representative positively portrayed at all?

99. The UFP/UFP representative is not portrayed at all in the story.

V06h: Is the independent candidate/ representative positively portrayed at all?

(40)

99. The independent candidate is not portrayed at all in the story.

The next variables are similar, but for negative portrayals.

To find out whether the political party or actor is negatively portrayed in the story, we look out for whether the story itself is a negative story or one that puts the party or actor in bad light at all. This includes the role they play in the story and how they are essentially characterized as the bad guys. Words like terrible, shocking, corrupt, reckless, negligent, irresponsible, hardship, suffering, pain, agony, etc. should provide an indicator of how they are negatively portrayed in the story. A negative portrayal can be just a short portrayal in the news story; don’t worry about overall balance yet (see variable 08 for that), but just code whether there is ANY negative portrayal of the actor in the story.

V07a: Is the NPP/representative negatively portrayed at all in the story?

1. Yes there is a negative portrayal present somewhere in the article.

2. No there is no negative portrayal present in the article.

99. The NPP/NPP representative is not portrayed at all in the story.

V07b: Is the NDC/representative negatively portrayed at all in the story ?

(41)

V07c: Is the CPP/representative negatively portrayed at all in the story?

99. The CPP/CPP representative is not portrayed at all in the story.

V07d: Is the PNC/representative negatively portrayed at all in the story?

99. The PNC/PNC representative is not portrayed at all in the story.

V07e: Is the PPP/representative negatively portrayed at all in the story?

99. The PPP/PPP representative is not portrayed at all in the story.

V07f: Is the GCPP/representative negatively portrayed at all in the story?

(42)

99. The GCPP/GCPP representative is not portrayed at all in the story.

V07g: Is the UFP/representative negatively portrayed at all in the story?

99. The NPP/NPPUFP/UFP representative is not portrayed at all in the story.

V07h: Is the independent candidate/representative negatively portrayed at all in the story?

1. The Independent candidate is not portrayed at all in the story.

V08a: What is the overall balance between positive and negative portrayals in the story, regarding the NPP/representative of the NPP?

1. Story is entirely negative toward the NPP

2. Story is slightly more negative than positive toward the NPP

3. Story is balanced toward the NPP – both negative and positive portrayals are relatively equally present

(43)

4. Story is slightly more positive than negative toward the NPP

5. Story is entirely positive toward the NPP

99. Story does not mention/portray NPP at all.

V08b: What is the overall balance between positive and negative portrayals in the story, regarding the NDC/representative of the NDC?

1. Story is entirely negative toward the NDC

2. Story is slightly more negative than positive toward the NDC

3. Story is balanced toward the NDC – both negative and positive portrayals are relatively equally present

4. Story is slightly more positive than negative toward the NDC

5. Story is entirely positive toward the NDC

99. Story does not mention/portray NDC at all.

V08c: What is the overall balance between positive and negative portrayals in the story, regarding the CPP/representative of the CPP?

1. Story is entirely negative toward the CPP

2. Story is slightly more negative than positive toward the CPP

3. Story is balanced toward the CPP – both negative and positive portrayals are relatively equally present

4. Story is slightly more positive than negative toward the CPP