Exploring data journalism practices in Africa: data politics,media ecosystems and newsroom infrastructures

Abstract

Extant research on data journalism in Africa has focused on newsroom factors and the predilections of individual journalists as determinants of the uptake of data journalism on the continent. This article diverts from this literature by examining the slow uptake of data journalism in sub- Saharan Africa through the prisms of non-newsroom factors. Drawing on in-depth interviews with prominent investigative journalists sampled from several African countries, we argue that to understand the slow uptake of data journalism on the continent; there is a need to critique the role of data politics, which encompasses state, market and existing media ecosystems across the continent. Therefore, it is necessary to move beyond newsroom-centric factors that have dominated the contemporary understanding of data journalism practices. A broader, non-newsroom conceptualisation beyond individual journalistic predilections and newsroom resources provides productive clarity on data journalism’s slow uptake on the continent. These arguments are made through the conceptual prisms of materiality, performativity and reflexivity.

Keywords

Africa data journalism data politics market media ecosystems newsroom infrastructure power state

Introduction

Data journalism has revolutionised newsroom practices across the globe. This form of journalism enriches the production and distribution of news and has brought benefits to investigative journalism. The future is digital, and practices such as data journalism are becoming part and parcel of the journalism profession. In Africa, data journalism has the potential to reinvigorate the purpose and essence of journalism in a continent dogged by massive corruption (Mutsvairo, 2019), declining public trust in the media (Chibuwe et al., 2022), and a growing imperative for media that serve the public interest (Chibuwe et al., 2022). Research by Moyo (2019), Mutsvairo et al. (2020) and Muneri (2019) shows that the uptake of data journalism has been slow and remains less utilised across the continent. This extant research (see Moyo and Munoriyarwa, 2019; Mutsvairo, 2019) notes several factors hindering data journalism uptake. These include the scarcity of skills among journalists who are not trained for the practice (Munoriyarwa, 2022) and technology acquisition within newsrooms to support data journalism practices (Mutsvairo, 2019). Gondwe and White (2022) have noted that some newsrooms in Tanzania and Zambia, for example, are hopelessly ill-prepared for the practice because of journalists’ perception of data journalism and the lack of resources.

These existing accounts of the slow uptake of data journalism in sub-Saharan Africa are newsroom centric. They focus on impeding factors within African newsrooms, such as lacking newsroom infrastructure and skills (Gondwe and White, 2022; Mutsvairo, 2019). We agree on the broader implications of newsroom infrastructures to the uptake of data journalism. But we move beyond this to interrogate the role of media ecosystems, the market and the state in the uptake of data journalism. In this paper, we adopt a ‘jet-plane’ view (Chambwera, 2021) – that seeks to understand the slow uptake of data journalism at a macro-level, beyond the newsrooms themselves and the individual journalists’ predilections. This allows us to move away from ‘the helicopter view’ (Chambwera, 2021), which adopts micro-lenses, to understand data journalism uptake.

Therefore, this article discusses factors impinging on data journalism in Africa through the prism of data politics, media ecosystems and newsroom infrastructures. We argue that to understand the slow uptake of data journalism practices in Africa, we should link these three together, as exerting a huge influence on how data journalism is appropriated on the African continent. Data politics encompasses the political aspects of data and refers to how data is collected, accessed, controlled and capitalised (Ruppert et al., 2017). For this paper, we focus on data politics from two perspectives. The first pertains to the power of the market to extract and control big data for value generation, thus making it difficult for certain groups of people to access this data. The second aspect of data politics refers to the vast amounts of data that governments generate, collect, and store through their services to the general public. The lack of freedom of information legislation and open data initiatives in several countries impacts how Africa can access this data. Accessing information remains challenging in countries where freedom of information legislation exists due to excessive bureaucracy and mistrust.

Media ecosystems in Africa, characterised by fractured media-state relations and media capture, also impinge on the growth of data journalism on the continent. By media capture, we mean the [creation] of relations of patronage and clientelist media through legal and administrative regulation, financial and economic incentives and the manipulation of media ownership (Mabweazara et al., 2020).

Newsroom infrastructures also actively impede data journalism practices. Regarding human resources infrastructures, African newsrooms are not well-positioned to practice data journalism effectively. These three factors – data politics, media ecosystems and newsroom infrastructure – collude to create a constraining environment for data journalism practices. The argument we make and sustain is that data journalism in Africa is implicated in issues of media freedom. Data-driven journalism allows news organisations to hold state and corporate institutions accountable, strengthening the fourth estate role of the media. Hence, the advent of data journalism has led to new journalism practices that have consequences for democracy and society (Westlund and Lewis, 2017). Therefore, we view media freedom and media ecosystems as mutually influential in data journalism.

To support our arguments, we draw on Borges-Rey’s (2020) concepts of materiality, performativity and reflexivity. He used these concepts to examine the development of data journalism, its challenges, and its particularities in Scotland, Wales and Northern Ireland. Apart from Borges-Rey (2020), no known research has examined non-newsroom factors influencing data journalism uptake. Therefore, we find Borges Rey’s conceptualisation useful as a starting point for understanding non-newsroom factors’ influence on data journalism uptake.

Methodologically, the article relies on interviews with 13 journalists from eight African countries, namely Namibia, Liberia, Nigeria, Sierra Leone, Gabon, Cameron, Zimbabwe, Eswatini, and South Africa, interviewed at different times between July 2021 to May 2022. We purposively sampled prominent journalists involved in large data-driven investigative journalism projects like the Panama Papers¹, the Paradise Papers² and the Fishrot scandals,³ together with global partners. For data analysis, the article utilises thematic analysis (Braun and Clark, 2006). This method identifies, analyses, and reports patterns (themes) within data. Thematic analysis organises and describes data in rich detail. Through thematic analysis, commonly recurring themes are drawn from the data.

This article is organised as follows. In the next section, we discuss the concept of data journalism and provide an overview of the current state of the field. This is followed by our conceptual framework, where we conceptualise materiality, performativity and reflexivity. After that, we then present our findings. We first discuss the influence of data politics and all its three key components – power, the state and the market and demonstrate how their simultaneous coalescence influence data journalism. We then explore African media ecosystems and how they influence data journalism practices. Lastly, we explore the influence of newsroom infrastructures. We conclude by offering a nuanced discussion on how these factors link to materiality, performativity and reflexivity.

Data journalism: definitions and research

An era of big data has triggered constant changes in news media. The most significant changes include, among others, the move towards algorithmic and data-driven news production. Technological programmes like digital metrics and algorithms now produce data that augment journalism practice (Mutsvairo et al., 2020). There is no universally agreed definition of data journalism. Fink and Anderson (2015) note that definitions of data journalism are deeply contested and diffused. In some quarters, data journalism has been defined as finding stories in data – stories of interest to the public – and presenting these stories most appropriately for public use and reuse (the Heravi, 2017; Silicon Republic, 2016). Larrondo-Ureta and Ferreras-Rodríguez (2021) add that data journalism involves visualisation in storytelling, statistics, tables, 3D maps and other forms of data. Knight (2015) states that data journalism is closely intertwined with infographic and computer-assisted reporting practices in earlier newsrooms. De Maeyer et al. (2015) further add that data journalism has evolved to reveal two emerging trends. The first is the technological one, focusing on the data itself. Secondly, the journalistic approach focuses on the applications and implications of the advent of data in newsrooms. Thus, without a universally agreed definition of the practice, scholars have moved away from the dysfunctional maze to focus on what the practice does to news production. Some scholars emphasise the analytical affordances of data in journalism by integrating quantitative (graphics, visuals and statistics) with qualitative approaches to news (Lynn et al., 2018; Rogers, 2013). Other scholars have emphasised the contribution of data to serious newsroom practices like investigative journalism (Larrondo-Ureta and Ferreras-Rodríguez, 2021).

There are also enduring arguments about the skills that data journalists require (Hannaford, 2015; Loosen et al., 2017). Loosen et al. (2017) argue that a ‘complete’ and professional data journalist requires programming, online data search, visualisation, and verification skills. Tabary et al. (2016) add that data journalism is no longer possible with the status quo in journalism schools (J-Schools) because the new skills required by the data journalist include database management and statistical interpretation. Hannaford further argues that these skills have been domiciled in other disciplines like computer science and applied mathematics. Tabary et al. (2016) imply that the existing journalistic skills normatively taught in J-Schools would be sufficient for data journalism practice. According to Tabary et al. (2016), data journalism is too limited to automated visualisations and utilisation of already existing data sets, making no effort to change anything. Hence, existing journalism skills would adequately serve the new practice.

Most of the available research on data journalism has generally focused on a limited set of countries in the Global North (Appelgren et al., 2019; Fink and Anderson, 2015; Larrondo-Ureta and Ferreras-Rodríguez, 2021). Research on data journalism in Africa is still emerging, as outlined earlier in the paper.

Conceptual framework: materiality, performativity and reflexivity

Borges-Rey (2016, 2020) develops an epistemological framework of data journalism in Scotland, Wales and Northern Ireland using three conceptual lenses – materiality, performativity and reflexivity. He argues that the continual interaction of the materiality of data and data journalists’ performativity consequently mediates the reflexivity of these journalists. In other words, ‘data as a material entity intermediates the professional practice and mindsets of data journalists’ (2016: 834). These concepts provide a framework for understanding how data politics, access to data, media ecosystems, and newsroom infrastructures impact data journalism in Africa. The materiality of data refers to the origination of data itself (Tuin and Dolphijn, 2012). Borges-Rey states that data commonly originates through an in-house collation managed by news organisations, Freedom of Information (FOI) requests, open data regimes, private actors, and or leaks by whistle-blowers. For example, Borges-Rey (2020) notes that in Scotland, Northern Ireland and Wales, the Freedom of Information Act (FOIA) essentially infuses the material character of the datasets used as evidence by data journalists. Because of this dependence on FOIA, the public bodies’ bureaucratic mechanisms can potentially restrict data access (Borges-Rey, 2020). Sometimes, data is obtained after intense and prolonged negotiations with the body holding the data (Borges-Rey, 2020). In some instances, when it is finally received, it is in a format that is non-readable (Borges-Rey, 2020). All these challenges constrain the performativity of journalists in how they practice data journalism. The interplay between data materiality and data journalism performativity affects the reflexivity of data journalists.

The material identity of data depends on two major factors. Firstly, it depends on the human agency of journalists and secondly, it depends on the experts who contextualise the data and make it understandable (Borges-Rey, 2020). This process, called ‘datafied agency’, also states that third-party solutions are required for data journalism (Borges-Rey, 2020). These include algorithms, automated systems, related software, and the ability to code and programme. Software for data journalism is not always available and affordable. Neither are they always compatible with existing newsroom infrastructure, especially in the African context we examine in this paper. There are several key takeaways in the concepts of materiality, performativity and reflexivity critical to our argument. Amongst these important takeaways is that the chief definer of data journalism is access to data itself (materiality). There are also important issues of skills required to make sense of the data and newsroom infrastructures (performativity). The interaction between materiality (access to data) and performativity (ability to interpret and use the data) leads to journalists’ reflexivity (entrenched norms, values and mindsets). These concepts are central to our illumination of how data journalism as a practice in African newsrooms is impacted by access to data, media infrastructure and skills.

Data politics: power, the market and the state

Data journalism largely depends on access to data normally held by market and state actors. However, market actors are the main data brokers and collect vast amounts of data for commercial value. Tech giants, especially those based in the USA, heavily influence how data is generated and used and control the global digital ecosystem and, therefore, the entire movement of data (Kwet, 2019). These tech giants can use and transform big data into information and knowledge since they own data collection and processing infrastructures. In a process facilitated by data capitalism, ‘access to data, and the ability to transform raw data into useful information, is asymmetrical, and power lines in the institutions with the technical and economic resources to render it intelligible’ (West, 2019: 36). Although a successful data-driven story can rely on data sets of any size, data journalism primarily feeds off of a broad phenomenon of big data, which, as stated above, is owned and controlled by a few tech corporations in the Global North. National data gathered locally often sits in global technology companies’ data centres. In contrast, many countries in the Global South cannot access or benefit from it (De Bastion and Mukka, 2020: 29). Africa does not have the technology to create the cloud infrastructure needed to host data nationally or regionally. These political economy factors impact access to data by African journalists working on stories that require Big Data. As a result, African journalists cannot conduct large-scale investigative journalism projects such as the Panama Papers and the Luanda Leaks⁴ that rely on data journalism. One rare example is the South African Gupta Leaks.⁵ This cross-media collaboration exposed the corrupt activities of the Gupta brothers and their association with the then President, Jacob Zuma. Journalists in the Global North partner with private corporations with data collection and analysis capacities. African journalists cannot do this

In the Panama Papers and Luanda Leaks, African journalists who worked on these stories were part of the International Consortium of Investigative Journalists. This U.S-based nonprofit newsroom works with a local and global network of reporters and media organisations to investigate the most critical stories in the world. Without these Global North partnerships, African journalists cannot produce large-scale, data-driven investigative stories. Journalists face two key obstacles. First, due to the political economy factors discussed earlier, they cannot access big data to collaborate on any leaked documents they may receive. Secondly, they do not have the resources to convert large unstructured data into searchable data. Many African newsrooms have no access to data science techniques. For instance, in the Panama Papers, the amount of data involved in the leak was about 2.6 Terabytes and contained approximately 11.5 million leaked documents. This data volume is nearly impossible to investigate without access to advanced data forensics, as observed by Nigerian journalist Leakan Odiyepo:

During the Panama papers’ leak, African journalists had to depend on the software resources of their counterparts in Europe. We did not have well-equipped computers that had the right software. But this was not about the competence of the journalists, but software issue now. . . so it was about how our newsrooms are/ were equipped (Lekan Odiyepo Nigeria Punch; 20 January 2022).

Chief Bisong Etahoben, a renowned Cameroonian investigative journalist, working for Weekly Post, stated that without Big Data from outside, they could not have taken part in producing stories for the Panama Papers. ‘In my newsroom, we only had two trained data journalists. We had to depend on our counterparts in better-resourced newsrooms outside Africa. This is how we broke the Panama leaks as well’ (Interview, 3 February 2022).

The state also generates and collects vast quantities of data through everyday activities. Therefore, the public sector is also a big reservoir of data. This data represents a significant source of value for both the public and private sectors, including the media. Domeyer et al. (2021) state that despite the large amounts of data that governments hold and the potential benefits this offers, many governments worldwide remain unable to capture the opportunity. This is because ‘data are typically dispersed across a fragmented landscape of registers [. . .] which are often managed in organisational silos [. . .]and are routinely stored in formats that are hard to process or in places where digital access is impossible'. As a result, this data becomes difficult to access by citizens and organisations. There is also little transparency about what kind of data governments have. While many countries in the Global North have adopted modern data management policies to deal with this problem, in Africa, the lack of or poor implementation of information laws makes creating an environment that supports open data challenging. The Open Data Barometer report of 2019 states that ‘while some governments are advancing towards data openness, that remains the exception, not the rule’ and that most governments on the continent ‘still lack any kind of comprehensive guidelines, technical standards, and management procedures for their (open) data’ (Web World Web Foundation, 2019). Open data is not yet entrenched in law on the continent, and the legal frameworks supporting it are either incomplete or absent. Open Government Data (OGD) has been extensively defined as both a philosophy and a set of policies that promotes transparency, accountability and value creation by making government data available to all (Organisation for Economic Co-operation and Development, OECD, 2022). By their nature, governments and (quasi) public bodies produce and often commission huge quantities of data and information (OECD, 2022). When this data is made available to journalists and any other public members who request it, it is a huge possibility of making both government and these institutions more accountable to the citizens they are meant to serve. Journalists and citizens can scrutinise this data to hold these institutions to account. Therefore, government, state, and quasi-state institutions-held data constitutes an essential source of data journalism practices. In more liberal and democratic contexts, states should actively promote accessibility to this data, not only for them to be held accountable but also to reap the benefits of business promotion, innovation, and the creation of citizen-centric services that are most likely to result from public exploitation of OGD. In less politically transparent contexts like in Africa, state obscurantism discourages the access, use, reuse and (re)distribution of state-held data, which, in turn, discourages the practice of data journalism. One journalist notes:

When you investigate a story, you see a concerted effort by state agencies, parastatals, and government departments to keep you off from the data. . .Even audit reports for parastatals that should be readily available are hidden. I was frustrated when I was investigating the ZESA⁶ prepaid meter scandal. . .It sucked it a minister, a top-ranking permanent secretary, and senior executives at the parastatal. . .I could not even be allowed to access the audit report. . .it was damning. . .it took a leak later to get it. . . (NewsDay, Zimbabwean Journalist, 7 May 2022)

The lack of open data regimes in many African countries poses challenges for data journalism. In 2011, a group of leading African media and information stakeholders adopted the African Platform on Access to Information (APAI), which recognised the right to information as a human right. It was endorsed by the African Commission on Human and Peoples' Rights and the Pan-African Parliament.⁷ The platform stipulates that ‘information held by public bodies is public and should be subject to disclosure’. Since the adoption of APAI, the number of countries with freedom of information (FOI) laws in Africa has grown substantially from about four to almost 30 countries (CIPESA, 2020). However, enforcement and implementation of these laws have been lacking. But some countries in Africa have been slow in legislating FOIA laws. One such example is Namibia; despite its healthy freedom of expression ratings, the government has worked on a FOIA law since 2002. It has taken the country more than 20 years to debate the law. This shows the lack of urgency regarding FOIA laws. Namibian investigative journalist Frederico Links notes:

It has taken almost two decades to pass the FOIA bill into law in Namibia. As I speak right now, the bill is still under debate. The previous Attorney General of the country once said that a FOIA is not an urgent issue. The whole idea is to make it difficult for journalists to access data and any other information held by the state and its institutions. This is why criminal behaviour continues within the state (Frederico Links, freelance investigative journalist, 12 May 2022)

Delaying the passage of FOIA laws equally delays the establishment of a monitorial framework for journalists. But this is not only a Namibian peculiarity. In Eswatini (formerly Swaziland), an investigative journalist notes:

The FOIA bill has been in debate for the past 20 years. Parliaments have been formed and dissolved, and the FOIA bill is still stuck there. The state is quick to enact other laws, like security legislation. They take a few months, even weeks. For example, it took a few weeks for the king to pass a law on gender-based violence. I am not saying this is wrong. We encourage this, but the urgency should cut across all forms of legislation. . . (Metji, Eswatini independent investigative journalist, 3 May 2022).

Notwithstanding a supporting framework, in many countries, as reported by the Friedrich Ebert Stiftung (FES) African Media Barometer and the World Press Freedom Index by Reporters sans frontières (RSF), there are several incidents where journalists have been barred from accessing information from state institutions, government meetings and the courts, thus stifling the principle of open data. In addition, contradictory laws that restrict access to information are still on the books in some countries. For example, colonial-era laws such as the Official Secrets legislation gag both the public and journalists from accessing information considered secret by the state. Many countries, such as Kenya, Rwanda, Sierra Leone, South Africa, Uganda, and Zambia, have recently enacted data protection legislation. In contrast, others already had Protection of Information laws from the colonial era, including Namibia. Based on the EU Data Protection Directive 95/46/EC, these data protection laws provide the fundamental right to privacy. However, these data protection laws may create tensions between the right to privacy and the public interest. There is also fear these laws, operating in conjunction with repressive legislation, can be used to shield political elites from scrutiny. In South Africa, for example, the recently adopted Protection of Personal Information Act (POPIA) requires a person to seek permission from a source before quoting them. As a result, some lobby groups have requested that journalists be made an exception to this rule as long as they adhere to the Press Code. In Namibia, for example, the Protection of Information Act (1982) is routinely used to block government officials from giving information to journalists and allow for invasive communications surveillance (MISA, 2021). One journalist noted:

The Namibian laws actually block journalists from accessing information. If you look at the protection of Information Act (1982), you will notice that for journalists, it closes off any possibilities of getting information from sources that you want. . . (Immanuel Shinovene, The Namibian, 18 February 2022)

African media ecosystems and their influence on data journalism practices

Political economy approaches to journalism focus on forces outside the newsroom that impact and structure news production. These external forces and how they interact with the media constitute a media ecosystem. For this paper, these external forces include the state as a major producer of data, quasi-state institutions and private players, and holders of a huge corpus of data. The literature defines a media ecosystem as complex combinations of print, broadcast, digital, and social media (e.g. Zuckerman, 2021). However, this paper defines the media ecosystem as the complex interplay of journalists and other external actors who impact news production and outputs. In the case of this paper, we want to focus on specific aspects of the media ecosystem. We focus on media-state relations as constituting an impediment to data journalism practices. Media-state relations in many African countries are fractured (Chibuwe et al., 2022). The political elite dominating African states do not see the media as serving a public interest but as colluding with imperialist forces to recolonise Africa. In Zimbabwe, the private-owned media is often termed ‘neo-imperialist, foreign-funded and regime change media’, meant to overthrow the ruling party. This frosty relationship implies that,

The independent-owned media cannot get access to data held by the state or quasi-state institutions. The suspicion is that we want to write negatively about them. These institutions are scared to even share with us any data. We asked for data on COVID – 19 related expenditures and tenders from the Ministry of Finance. . .we never got it. . . Zimbabwe NewsDay journalist, 7 July 2021).

In Liberia, Bettie Johnson, a renowned investigative journalist working for ZAM media, noted that,

If your media house is not linked to the ruling political elites or connected to them somehow, you are never going to get the data you want to produce investigative journalism stories (Interview, 6 February 2022).

During their investigation of the now-infamous ‘Fishrot’ scandal in Namibia, journalists noted that they had to rely on sources from other media houses outside Namibia. Namibian authorities refused to release any data that would have helped the revelation of massive corruption by the media (Interview with Frederico Links, 6 February 2022).

The polarised pluralist model (Hallin and Mancini, 2004) of most African media systems militates further against the practice of data journalism. The polarised pluralist model of the media describes a media system characterised by strong links to the political world rather than the market and the public. The media under this system shows characteristics of an overt dependency on the state, political parties and wealthy individuals as proprietors. In addition to these characteristics, this model exhibits an absence of professionalisation of the media which, ultimately, stunts the development of an autonomous media system. In some African countries, the media exists in a polarised environment - ‘. . .often existing to mobilise readers to support political causes’ (Hallin and Mancini, 2004: 98).

Furthermore, ‘. . .they are instrumentalised by private interests with political alliances and ambitions who seek to use their media for political ends. . .’ (Papathanassopoulos, 2007: 194). In Cameroon, Chief Bisong Etahoben of the Weekly Post notes that,

When your publication is known to seek accountability, you will not have data from any state institution or any quasi-state institution. You will find that ‘politically correct’ publications are given access to data. . .but they practice ‘lightweight data journalism’ that validates the ruling elite and praises them even the smallest asinine project. . .is brought into the news. The actual (sic) data that shows patterns of misgovernance is denied to a journalist (Interview, 2 February 2022).

The polarised relationship between political elites and the media affects data journalism practices at a dual level. At the first level, the political elite uses their influence to sway public boards by sharing data with journalists, even if the data is in the public interest. In Guinea Bissau, one data journalist reporter notes that,

Even private actors with a vast corpus of data, like banks, are scared to share data because of the political consequences that might visit them if journalists use the data to reveal financial corruption or anything else. You will realise that the fear of sharing data with journalists permeates all players, private or public. . . . (Moussa Aksar: Journal ‘L’Événement’, Interview January 2022).

This relationship pattern between the media and the political elite can only be understood if we discern the nature of the state’s elites in many parts of the continent. Kandeh (1999) notes that in most parts of the continent, despite some improvements in governance, there are increasing levels of ruling elites looting the state and illegitimately appropriating state institutions to advance political domination of the state. Kandeh (1999: 353) notes how the state in Africa is an arena for the ‘criminal expropriation [in the engendering] context of elite parasitism and repression’. Under these circumstances of corruption, plunder of public resources, misappropriation of public finances, and many other governance ills, the dominant political elite has every reason to gatekeep any leakage of incriminating data to journalists. In South Africa, for instance, the team comprising 19 journalists from AmaBhungane Centre for Investigative Journalism (One of South Africa’s most known investigative units), The Daily Maverick and News24, who revealed, through excellent investigative journalism, the #GuptaLeaks, often complained of state quasi-state institutions, as well as private actors denying them access to data. Sam Sole, the most prominent investigative journalist in the team, notes, ‘Even when the data should be publicly available, there were always attempts to shield it from journalists who wanted it. . .’ (Sam Sole interview, 8 July 2021).

This data revealed a web of corruption within the state, now dubbed the state capture and subject to a presidential commission. The data, therefore, is ‘a protected species’ because it points out in quantitative terms how the state is robbed and the cumulative development of corruption in African states. Lekan Odeypor, a Nigerian data journalist, working for the Nigerian Punch, argues that access to data is sensitive and political because,

‘. . .if journalists get hold of the data, it can be used to reveal how these politicians are ransoming and looting the state. . . that is why there will always be a strategic resistance from powerful politicians who resist any sharing of data, especially with private-owned media houses. . .where they may not influence what finally gets published. . .’ (Interview, 20 January 2022)

At the second level, media capture is increasing (Mabweazara et al., 2020). We argue that media capture stands in the way of data journalism. For example, patrimonial media is entrenched mainly in the state-controlled media in Zimbabwe. What does this mean for data journalism practice? It means that either data journalism practice does not exist in the captured media or exists only in attenuated forms - that is, as far as the data and the stories accompanying it can bolster the agendas of the ruling elite. Thus, captured media become lapdogs rather than watchdogs. In the process, captured media entrenches and legitimises the enduring and embedded practices of corruption, state ransoming and looting by silence or omission. In the captured media of Gabon, for example,

Journalists do not use data journalism or any of its tools to sharpen their investigative reportage. They write with a political objective, to please those in power and sustain and legitimise that power, not to dig nor expose state and elite excesses and their predatory practices. . . . (Moussa Akbar, Journal L’Événement, 26 January 2022)

A private media journalist in Zimbabwe said of the captured public media,

There is no journalism happening in state-controlled media. They are all captured by ZANU PF (Zimbabwe’s ruling party). To even ask for data journalism is to ask for too much. Data journalism is evidentiary, and it only is so if it does not reveal wrongful acts and if it does not hold the elite to account, then it is not data journalism. Be honest, is this what is happening in the state-controlled media? Obviously not. . . (NewsDay journalist, 7 July 2021)

In 2021, a massive Personal Protective Equipment (PPE) tender scandal erupted in Zimbabwe, involving the president’s family and their close associates. The state media The Herald and The Sunday Times were conspicuous by their silence on the scandal, despite the matter dominating public discourse for over 2 months. This sums up patrimonialism and how it colours effective data journalism practices.

Thus, existing patterns of media capture turn data journalists into lapdogs that do not bark at those in power (Kovach and Rosenstiel, 2001) but rather into outlets that praise and worship power and do not hold it to account (Whitten-Woodring, 2009).

African newsroom infrastructures as impediments to data journalism

Shrinking newsrooms across the continent have impacted newsroom capital investments (Finlay, 2020). In South Africa, for instance, The South African National Editor’s Forum (SANEF) has noted that in the past 10 or so years, journalism jobs have been cut, and newsrooms have closed. Newsrooms have not had many technology investments as they battle to stay afloat. Data journalism, by its nature, requires newsrooms with powerful computers and access to big data analysis software (Bounegru and Gray, 2021). This is what is lacking in several African newsrooms. In Cameron, for instance, as stated earlier, journalists have to rely on newsrooms in the Global North for big data analysis. This is because ‘In our own newsrooms, we do not have those powerful computers with the software needed to make sense of the data’ (Chief Bisong interview, 2 February 2022). Most data journalists in Africa admitted that they engage with their counterparts in other parts of the world because of this lack of newsroom equipment. In Nigeria, Dapo, an investigative journalist, noted, ‘Often, we may get hold of large quantities of data, but we would require help with making sense of it’. Frederico Links, an investigative journalist in Namibia, agreed, ‘If you look at the vast quantities of data leaked in the Panama and Paradise papers, for example, I doubt there would be newsrooms in Africa capable of analysing that huge corpus without external help’ (Federico Links, 6 February 2022). Thus, the issue of newsroom resources rings high as an impediment to data journalism.

In addition to this, African journalists are not trained for data journalism. This is mainly because no known journalism schools in Africa provide training for data journalism. One journalist from Libya notes, ‘Most of what we call training is from in-house training and refresher courses offered by news-related organisations’ (Evelyn Groenink - Investigative Journalist: ZAM Magazine, 24 January 2022).

This points to the need to rethink the journalism curriculum in African universities. Since the advent of data journalism, several training opportunities in Africa have come from organisations other than J-Schools. Code for Africa affords the most available and pervasive training programmes. One of its flagship training programmes is the academy-AFRICA initiative that runs in partnership with the Google News Initiative and the World Bank. This initiative is also partnering with newsrooms and Hacks/Hackers communities in 12 African countries. Through this partnership, Code for Africa has built a StoryLab digital journalism laboratory, ‘which has physical hubs in four African countries along with an existing global network of roving data journalists and digital creatives who help newsrooms on projects ranging from investigative data analysis to drone journalism and 360° video’ (https://academy.africa/about). These donor-driven training programmes often derive their knowledge, approaches and methods from the West. These organisations’ training regimes foster a particular journalism practice mode that aligns with the West’s interests and is embedded in the liberal-democratic tradition (Berger and Foote, 2017).

The training initiatives in data journalism and other technology-based programmes are reminiscent of the technology transfer models of the 1970s and 80s steeped in modernisation rationality. In this transfer, expertise from the Global North is seen as superior. It is transferred to other regions without considering those regions’ cultural and social contextual circumstances. Yes, using data to enhance journalism is vital for all countries, but issues of resources and access should not be ignored. The one-size-fits-all programmes often offered by donor-driven interventions do not work in the long run. One critical debate in African universities is the contemporary relevance of the journalism curriculum in various universities. The colonial journalism curriculum is still intact in most African universities. These curriculums depend much on Western journalism scholarship (McMillin, 2007). The problems of using a colonial curriculum in journalism education to anchor post-colonial journalism practices are visible in the disjuncture between journalism education and contemporary journalism practices in Africa. Where efforts to factor in local contexts have been made, they have rarely gone beyond the Western cultural sphere of influence (Chasi and Rodny-Gumede, 2018: 107). By arguing for an African-focused training regime and approach to data journalism, we are not arguing for an ‘isolationist approach’ seeking to reject everything Western. There are data journalism skills that remain relevant across all contexts. Appelgren and Linden (2020) note that skills like database management, statistics and design, which are multidisciplinary, are in demand for the data journalist. This solid interdisciplinary underpinning is also lacking in the African context. Hence, programming, statistical analysis, and information visualisation skills are essential even under an African-specific journalism curriculum, as they are the lynchpin.

Materiality, performativity and reflexivity in data journalism in Africa

In line with Borges-Rey’s (2016, 2020) study, the interviews carried out for this paper show that journalists’ performativity is greatly affected by access to data, legal constraints, and human and material resources to practice data journalism. In the African context, data’s material identity depends on international organisations and journalists with access to data and the resources to interpret and analyse the data. Most large-scale data-driven investigative stories related to Africa were conducted with the International Consortium of Investigative Journalists. The performativity of journalists based in the Global North who have access to material resources and are often tech-savvy is higher than those on the continent.

Datafied agencies as spaces where data can be found, we have argued, exist in attenuated forms in many African countries. We can reiterate the dangers posed, for example, to whistle-blowers on the continent who, in other parts of the world, have been sources of most ground-breaking data-driven journalism stories. Whistle-blowers face serious threats and sometimes death, despite how on several occasions, anti-corruption organisations and other interested parties have advocated for stronger protection of whistle-blowers (Mail and Guardian, 2021). Whistle-blowers are better protected in the West than in Africa. In Africa, their value as news sources for data journalists is compromised. In addition to these, intra-newsroom factors within newsrooms impede data journalism. Borges-Rey (2020) refers to this as the material identity of data. We have noted here that limiting factors in the newsrooms include a lack of skills, training and absence of data-journalism focused expertise. Shrinking newsrooms compound this, and declining budget outlays have impacted newsroom acquisition of the necessary technologies, including coding software and automated systems.

The slow uptake of data journalism on the African continent, which we have attributed to a captured media, the politics, access to data by journalists, the influence of the political elite on institutions, and newsroom infrastructures on the continent, can broadly be attributed to contemporary politics on the continent. Most African countries are still run by semi-authoritarian regimes that maintain a facade of democracy but retain control over institutions (Amundsen, 2019; Schedler, 2013). This means that institutions, no matter how much they may be interested in helping journalism play its public interest role, are hamstrung by the influence of the political elite, who are suspicious that data can lead to an unravelling of webs maladministration, malfeasance and corruption. Admittedly, governance has improved in some parts of the continent. But there are still many countries where supposedly independent institutions are reluctant to make decisions without political influence. As explained above, the cases of Zimbabwe, Eswatini and Gabon demonstrate this trend. Thus, the lack of accountability politics, at a broader level, vitiates data journalism practices.

Conclusion

In this article, we have argued that to understand the quotidian practices of data journalism in Africa, we need to understand the peculiarities of data politics (market and the state) on the African continent, African media ecosystems and newsroom infrastructures. Furthermore, we have argued that there is a need to understand the effects of ‘data politics’ obscurantist practices on data journalism practices in Africa. Media capture, an existing reality in African media ecosystems, influences how data journalism is practiced on the continent. The often illicit and corrupt relationship between political elites in many African countries and institutions that hold the much-needed data is a huge definer of data journalism in Africa. This paper is based on several interviews with investigative journalists across sub-Saharan Africa. Future research might expand the unit of analysis by increasing the number of journalists involved to include those from Arab–Africa. This can be useful for understanding if there are certain peculiarities in these contexts.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Received funding from University of Johannesburg, University Research Committee (URC).

ORCID iD

Sarah Chiumbu

Notes

References

Amundsen

(2019) Political corruption in the Middle East. Authoritarianism and Wasta in a Global Perspective.

Appelgren

Linden

(2020) data journalism as a service: Digital native data journalism expertise and product development. Media and Communication 8(2): 62–72.

Appelgren

Lindén

van Dalen

(2019) Data journalism research: Studying a maturing field across journalistic cultures, media markets and political environments. Digital Journalism 7(9): 1191–1199.

Berger

Foote

(2017) Taking stock of contemporary journalism education: The end of the classroom as we know it. In: Goodman

Steyn

(eds) Global Journalism Education in the 21st Century: Challenges and Innovations. Austin: Knight Center for Journalism in the Americas, University of Texas at Austin, pp.245–265.

Borges-Rey

(2016) Unravelling data journalism. Journalism Practice 10:7.

Borges-Rey

(2020) Towards an epistemology of data journalism in the devolved nations of the United Kingdom: Changes and continuities in materiality, performativity and reflexivity. Journalism 21(7): 915–932.

Bounegru

Gray

(2021) Introduction. In Bounegru

Gray

(eds) The Data Journalism Handbook: Towards a Critical Data Practice. Amsterdam: Amsterdam University Press, pp. 11–23.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology 3(2): 77–101.

Chambwera

(2021) Understanding netflix’s foray into original productions in South Africa: A “Jet Plane” and “Helicopter” view. In Motsaathebe

Chiumbu

(eds) Television in Africa in the Digital Age. Cham: Palgrave Macmillan, pp.39–57.

10.

Chasi

Rodny-Gumede

(2018) Decolonising communication studies: Advancing the discipline through fermenting participation studies. In: Mutsvairo

(ed) The Palgrave Handbook of Media and Communication Research in Africa. Cham: Palgrave Macmillan, pp.55–71.

11.

Chibuwe

Munoriyarwa

Motsaathebe

, et al. (2022) Newsroom disruptions and opportunities in times of crisis: Analysing Southern African media during the COVID-19 crisis. African Journalism Studies 43(2): 53–70.

12.

CIPESA (2020) ICT Policy in Africa. Available at: https://cipesa.org/ (accessed 12 March 2021).

13.

De Bastion

Mukka

(2020) Data and the Global South: Key Issues for Inclusive Digital Development. Washington, DC: Heinrich-Boll-Stiftung, Available athttps://us.boell.org/en/2020/10/20/data-and-global-south-key-issues-inclusive-digital-development (accessed 16 March 2022).

14.

De Maeyer

Libert

Domingo

, et al. (2015) Waiting for data journalism: A qualitative assessment of the anecdotal take-up of data journalism in French-speaking Belgium. Digital Journalism 3(3): 432–446.

15.

Domeyer

Hieronimus

Klier

, et al. (2021) Government data management for the digital age. Available at: https://www.mckinsey.com/industries/public-and-social-sector/our-insights/government-data-management-for-the-digital-age (accessed 15 March 2022).

16.

Fink

Anderson

(2015) Data Journalism in the United States: Beyond the "usual suspects”. Journalism Studies 16(4): 467–481.

17.

Finlay

(2020) State of the newsroom: Before/After. Wits Journalism. Available at: https://journalism.co.za/wp-content/uploads/2020/11/State-of Newsroom_2019_2020_30112020.pdf (accessed 25 January 2022).

18.

Gondwe

White

(2022) Data journalism practice in Sub-Saharan African media systems: A cross-national survey of Journalists' perceptions in Zambia and Tanzania. African Journalism Studies 43: 21–36.

19.

Hallin

Mancini

(2004) Comparing Media Systems: Three Models of Media and Politics. Cambridge: Cambridge University Press.

20.

Hannaford

(2015) Computational Journalism in the UK newsroom: hybrids or specialists? Journalism Education 4: 6–21.

21.

Heravi

(2017) Teaching data journalism. In: Mair

Keeble

Lucero

(eds) Data Journalism: Past, Present and Future. Abramis Academic Publishing, pp. 221–228.

22.

Kandeh

(1999) Ransoming the state: Elite origins of Subaltern terror in Sierra Leone. Review of African Political Economy 26(81): 349–366.

23.

Knight

(2015) Data Journalism in the UK: A preliminary analysis of form and content. Journal of Media Practice 16(1): 55–72.

24.

Kovach

Rosenstiel

(2001) Are watchdogs an endangered species? Columbia Journalism Review 40(1): 50.

25.

Kwet

(2019) Digital colonialism: US empire and the new imperialism in the Global South. Race & Class 60: 3–26.

26.

Larrondo-Ureta

Ferreras-Rodríguez

(2021) The potential of investigative data journalism to reshape professional culture and values. A study of bellwether transnational projects. Communication & Society 34(1): 41–56.

27.

Loosen

Reimer

De Silva-Schmidt

(2017) Data-Driven Reporting: An On-Going (R)evolution? A Longitudinal Analysis of Projects Nominated for the Data Journalism Awards 2013-2016 (Arbeitspapiere des Hans Bredow Insituts, nr 41). Hamburg: Hans-Bredow-Institut für Medienforschung, University of Hamburg.

28.

Lynn

Hermida

Fulda

(2018) What makes for great data journalism? Journalism Practice 12(1): 115–135.

29.

Mabweazara

Muneri

Ndlovu

(2020) News “media capture”, relations of patronage and clientelist practices in sub-Saharan Africa: An interpretive qualitative analysis. Journalism Studies 21(15): 2154–2175.

30.

Mail & Guardian (2021) Story of a whistleblower: ‘Baba, do you think you are about to die? Available at: https://mg.co.za/opinion/2021-12-03-story-of-a-whistleblower-baba-do-you-think-you-are-about-to-die/ (accessed 21 December 2021).

31.

McMillin

(2007) International Media Studies. Chichester: John Wiley & Sons.

32.

Media Institute of Southern Africa (MISA) (2021) The state of press freedom in Southern Africa, 2019-2020. Available at: https://unesdoc.unesco.org/ark:/48223/pf0000377172 (accessed 20 March 2022).

33.

Moyo

Munoriyarwa

(2019) Data journalism in the age of big data: An exploration into the uptake of data journalism in leading south African Newspapers. In: Mutsvairo

Bebawi

Borges-Rey

(eds) Cham: Palgrave Macmillan, pp.85–105.

34.

Moyo

(2019) Data journalism and the Panama papers: New horizons for investigative journalism in Africa. In: Mutsvairo

Bebawi

Borges-Rey

(eds)Data Journalism in the Global South. Cham: Palgrave Macmillan, pp.23–38.

35.

Muneri

(2019) Prospects for data journalism in Zimbabwe: Challenges of engendering a democratic society and an informed citizenry in the digital age. In: Mutsvairo

Bruce

Bebawi

Saba

Borges-Rey

Eddy

(eds)Data Journalism in the Global South. Cham: Palgrave Macmillan, pp.39–52.

36.

Munoriyarwa

(2022) Data journalism uptake in South Africa's mainstream quotidian business news reporting practices. Journalism 23(5): 1097–1113.

37.

Mutsvairo

(2019) Challenges facing development of data journalism in non-western societies. Digital Journalism 7(9): 1289–1294.

38.

Mutsvairo

Bebawi

Borges-Rey

(eds) (2020) Data Journalism in the Global South. Cham: Springer Nature.

39.

OECD (2022) Open Government Data. Available at: https://www.oecd.org/gov/digital-government/open-government-data.htm (accessed 9 November 2022).

40.

Papathanassopoulos

(2007) The Mediterranean or polarized pluralist model countries. In: Terzis

(ed.) European Media Governance: National and Regional Dimensions. Bristol: Intellect Book, pp. 191–200.

41.

Rogers

(2013) Facts Are Sacred: Text Only Ebook. London: Faber & Faber.

42.

Ruppert

Isin

Bigo

(2017) Data politics. Big Data & Society 4: 1–7.

43.

Schedler

(2013) The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism. Oxford: OUP.

44.

Silicon Republic (2016) How is data journalism changing the newsroom? Q&A with Bahareh Heravi E. Burke, ed. Silicon Republic. Available at: https://www.siliconrepublic.com/innovation/data-journalism-media-newsroom (accessed 12 February 2022).

45.

Tabary

Provost

Trottier

(2016) Data journalism's actors, practices and skills: A case study from Quebec. Journalism 17(1): 66–84.

46.

Tuin

IVD

Dolphijn

(2012) New Materialism: Interviews & Cartographies. London: Open Humanities Press.

47.

Web World Web Foundation (2019) Open Data Barometer — African Edition. Available at: https://idl-bnc-idrc.dspacedirect.org/bitstream/handle/10625/57855/57924.pdf (accessed 15 March 2022).

48.

Westlund

Lewis

(2017) Reconsidering news production: How understanding the interplay of actors, actants, and audiences can improve journalism education. In: Goodman

Steyn

(eds) Global Journalism Education: Challenges and Innovations in the 21st Century. Austin, TX: The Knight Center for Journalism in the Americas at The University of Texas at Austin, pp. 409–428.

49.

West

(2019) Data Capitalism: Redefining the logics of surveillance and Privacy. Business & Society 58(1): 20–41.

50.

Whitten-Woodring

(2009) Watchdog or lapdog? Media freedom, regime type, and government respect for human rights. International Studies Quarterly 53(3): 595–625.

51.

Zuckerman

(2021) Why study media ecosystems? Information Communication & Society 24(10): 1495–1513.