The Potential and Reality of Data Journalism in Developing Media Markets

"In the rush to embrace the open government and open data movement, donors have favored activities like boot camps, hackathons and conferences seem to promise a quick and sexy route to instilling data journalism skills. Yet these approaches make the mistake of assuming that the primary barrier to data journalism lies in technology and awareness."

Centring around a case study of data journalism in Myanmar, this report from Internews explains the role of data journalists and open data, laying out key considerations that can help predict the success or failure of new data journalism initiatives. Noting that information has the potential to help communities understand challenges in areas such as health, elections, conflict, land rights, and the environment, civil society organisations (CSOs), governments, and universities release data on a daily basis. However, most people do not have the skills or inclination to engage with data directly. That's where data journalists and the open data community come in. The core argument in this paper is that data journalism initiatives need to be designed holistically in ways that strengthen the fundamentals of public journalism practices while simultaneously addressing low levels of data and digital literacy (often at issue among the media in countries where democracy is under-developed) and a host of data access issues.

Data journalists are not merely messengers of breaking news events. Rather, they become "change agents for government accountability" in that their role becomes investigating the root cause of an issue in the public's interest. Public service journalists need to have the orientation, critical thinking skills, and drive to process and transform data sets into information that citizens need to know to engage with their governments. In seeking to train journalists for such a task, therefore, it is vital, according to author Eva Constantaras, to start with establishing a foundation: review basic numeracy, introduce digital research tools and, practice identifying and integrating basic statistics into a coherent narrative. Only then, she contends, does it become possible to assess a journalist's capacity for learning, step by step, the "hard" technical skills (e.g., scraping, cleaning, analysing, and visualising data) they will need to become truly effective practitioners. (On the other hand, if no access to information law exists to facilitate access to data, journalists may end up with skills but without the data to cover issues that matter to the public.)

Specifically, Internews' long-term approach encompasses not only training in-depth reporting and data skills but also harnesses the support of data scientists and developers, provides one-on-one editing and mentoring, and includes production support and dissemination of participants' products in the open data community. The capacity building involves using human-driven narrative and appropriate visualisations to present the data in ways that make complex issues comprehensible at the community level. Key aspects of the Internews approach are described, and focus on: inclusivity, accessibility, innovation, purpose, adaptability, and profitability.

The process begins with the journalists, training them on data literacy basics and laying the groundwork for more balanced, purposeful, and objective reporting in countries with a weak history of independent journalism. A second step is to link journalists with members of the open data community so they have access to these data sets. Finally, publishers and editors must be educated to understand that data journalism can be a valuable asset in the face of digital convergence. "Truly strategic approaches that aim for lasting impact will need to situate themselves more thoughtfully in the broader media and data landscapes of the country, and be more realistic about the levels of time and investment required for this complex development exercise to bear fruit."

Constantaras provides case studies of Internews' data journalism efforts in Kenya, Afghanistan, Latin America, and Myanmar. With regard to the latter, beginning in August 2015, the United States (US) Embassy in Rangoon, Myanmar's Public Affairs Office supported a series of Internews workshops for journalists, civil society actors, and developers and graphic designers to learn data analysis, storytelling, and visualisation skills. As a result of the workshops, in the lead-up to the landmark 2015 general elections in Myanmar, trainees produced a total of 41 data-driven stories with visualisations and 12 advocacy publications covering topics ranging from the numbers of voters and Ministers of Parliament (MPs) affected by flooding and conflict to the challenges of tax evasion and land registration. There are, however, a vareity of challenges going forward in this country with regard to data journalism projects. For example, data availability is an issue. However, several efforts are described here that bode well for more data in the future: (i) Mohinga.info [PDF] is the government's development aid portal designed to enable the people of Myanmar to better see the impact that aid activities have in their communities. (ii) The Open Myanmar Initiative is a CSO data project to track the participation and effectiveness of every parliamentarian. (iii) Phandeeyar, a new technology hub in Myanmar, has collaborated with Open Development Mekong, a regional open data portal, to launch an open data portal for Myanmar that will aggregate data that is publicly available in the country. (iv) Incubated by Phandeeyar, The Asia Foundation, and several other CSOs and technology groups, Mae Pay Soh is a project that digitalised and open-sourced the candidate list for the last elections and held a hackathon for civic hackers to make voter education apps that used the candidate data. Despite these promising endeavours, there is weak journalism capacity, a greater number of journalists and media outlets than the market can sustain long term, and a lack of metrics around media consumption.

Constantaras offers recommendations to accelerate Myanmar's open data movement, including:

• Invest in government capacity to collect and publish data and educate relevant officials on the role of open data in a democratic society.
• Educate CSOs about the linkages between advocacy and open data.
• Educate publishers and editors about business models for digital media.
• Create a core cadre of public-service-oriented independent media that collaborate actively with the open data community.
• Provide concrete incentives to the open data community.
• Design, implement, and evaluation a variety of data journalism capacity building activities. Examples include: (i) a 6-week fulltime Data News Lab model in which participants drawn from media, CSOs, and tech communities can split their time between learning and production in a simulated newsroom environment. (ii) An in-house training model designed to establish a permanent data desk within media outlets that have committed to integrating data journalism into the newsroom (7 Day, Myanmar Now). (iii) A one-year part-time fellowship that selects Myanmar's strongest journalists. (iv) A programme to integrate data journalism into the university communications curriculum. (v) Support for more extensive data journalism initiatives, such as partnerships with CSOs for fact checking, parliamentary monitoring, and public service tracking.