Bringing online data to a rural, offline community (photo: Kyle James)

Data journalism has opened up new ways of telling compelling stories, finding connections and creating visuals that explain important, complex issues is easy-to-understand ways.

This has become possible through the huge range of digital information now available online. While much of that information is presented online and sometimes in newspapers and magazines, what about communities who don’t have much access to the internet or even printed media?

Are these communities, often poor and rural, simply going to miss out on the results and benefits of data journalism? Not necessarily. onMedia’s Kyle James looks at a project in Cambodia that has been linking online and offline communities – with promising results.

Early this year, 13 young Cambodians working in the media and communication sectors gathered in the capital Phnom Penh for a data-driven journalism workshop. They were looking for stories related to their country’s social and political development. But instead of hitting the streets with notepads, recorders and cameras, they flipped open laptops and started digging in databases, online archives and other digital information sources.

While data journalism is taking off in Cambodia like it is in many parts of the world, for a lot of people in the country, going online to see the results of data analysis or opening a paper to see a chart or graph is simply not possible. Some 80 percent of Cambodians live in the countryside—and many of them aren’t connected to the electricity grid. While some people there do get online with their smartphones, most don’t. Newspapers in Cambodia seldom reach many areas outside the main urban centers.

But it is possible to link up this information found online with people living in offline environments, as a project carried out earlier this year by the German development agency GIZ, DW Akademie and several local groups showed.

And it’s not exactly rocket science; it just involves getting data into the kinds of information channels rural people can access.

Getting good data

At the data-driven journalism workshop organized by DW Akademie, participants scoured online resources like news archives, NGO reports, government websites, social media and blogs. They also dove into open-data sites such as Open Development Cambodia (ODC), which aggregates and disseminates publically available data on development trends in the country. (Check out the onMedia post on ODC from last year.)

Crunching data at the data journalism workshop (photo: Ly You Y)

The results of their research were compiled and put into a multimedia, interactive, data-driven journalism blog. It features data stories on falling rice prices, the high number of accidents on Cambodian roads, sex workers, among other topics.

One data story on the blog is about the difficulties in getting passports and IDs in Cambodia. It’s a crucial issue since so many Cambodians migrate to Thailand for work. They often go without any legal documentation since they don’t know even how to get the papers. And the hurdles, including the costs, are often too high.

The report features graphs comparing passport fees in ASEAN countries, statements from citizens, info about “unofficial” fees (i.e. bribes often asked for by officials and passport brokers), and other data. The challenge was to get that information to the communities most affected as well as to government officials, not just to have it circulating among the young and tech savvy in the cities.

Hitting the target audience

Fast forward a few months and from a wifi-equipped seminar room in Phnom Penh to a clearing outside a pagoda accessible only by dirt road. Some 75 people, mostly rice farmers and unskilled laborers, gathered under a canopy that gave welcome protection from the blistering Cambodian sun.

Rural Cambodians listen to info on passports provided by data journalists (photo: Kyle James)

They were here for a public forum on passports and IDs and to hear from the four local officials who had come to speak, and hopefully listen. As the moderator pointed to a screen set up in front, the villagers craned their necks for a better look.

They saw bar graphs and pie charts from the data story that showed how passport prices compare in the region, how long people in different countries have to work to afford one, and the wide gulf between the official passport price and how much people actually end up paying for one.

The moderators had translated the information graphics into Khmer, put them on a USB stick, brought them to the provinces and projected them onto the screen.

No broadband required.

“The visualized data reached people in a friendly and understandable way,” said Tep Bunthan, a manager at the Women’s Media Centre of Cambodia (WMC), a radio and TV broadcaster, and one of the moderators. “The graphs helped people take part in productive discussions.”

After the information session, there was a lively round of questions and opinions for the local policymakers. Equipped with new information, the villagers asked the officials outright to explain why passports cost more in Cambodia than in richer Southeast Asian countries. Several of the officials offered practical advice on getting passports and how to avoid being price gouged by brokers. They promised to take action to make the process easier.

A radio call-in show got the data to a wide audience (photo: Kyle James)

The forum was recorded and broadcast live on WMC, greatly increasing its reach. A few weeks later the broadcaster held a radio call-in show on the same topic, again using information from the data workshop, but talking to national-level officials this time. The moderator also wanted to follow up on the forum. Was the government putting its money where its mouth was?

Finally, to complete the circle, portions of the recordings of the forum and the call-in show were edited down put on the WMC website via Soundcloud and will soon go on the data journalism workshop blog, which will remain online to serve as a resource for people researching development topics in Cambodia.

That means the data journalism workshop blog will benefit from the voices of real people who are affected by the issue. Its data stories will be that much richer.

It’s about access

The goal of the project is to provide better access to information to people who don’t have all that much. The information on passports and other development topics gathered by the young journalists in the data workshop was great, but it was unlikely to reach many communities outside the capital.

But simply putting relevant information onto a slide and taking it to the communities themselves and then creating a radio program got the data out there to a much wider audience. Sure, it took more effort than putting it on a blog (especially that last bit of very bumpy ride out to the village), but not everyone has a computer or smartphone or even knows how to get online, especially in many rural parts of the developing world.

And not everyone in rural Cambodia reads, so in this case it was important to have someone explain the charts. But the bar graph format helped make it clear to everyone, regardless of reading level.

Use the channels through which offline communities get their info (photo: flickr/Peter J. Bury CC:BY-NC-SA)

While this project has used slides and old-fashioned radio, you could also print out informational graphics on paper, hand them out or post them in a central gathering place (of course, taking the literacy rate into account).

Perhaps people in the area get their information by word of mouth. So, another possibility would be to share the data with a local village leader or someone who can be trusted to pass along good information to others.

Sometimes and in some places, a low-tech approach is the best one.

Linking these online and offline worlds has brought tangible benefits to rural Cambodians. At the same time, it’s letting people accessing the information on the internet hear to the experiences of people they probably wouldn’t have been able to otherwise. I’d say it’s a win-win.

The data journalism/information access project had a great deal of help from its cooperation partners: the Women’s Media Centre (WMC), Open Development Cambodia and the Cambodian Center for Human Rights.

Author: Kyle James

Photo – Fotolia/spaxiax

Should journalists learn computer code? That’s a controversial question at the moment. The general consensus seems to be that today’s journalists need at least a basic level of technical literacy so when they team up with developers, they can better understand the language they speak. To get journalists up to speed, onMedia has prepared an overview of the best websites to help learn how to write code without having to turn into a hard-core programmer.

The debate about whether journalists need to learn to code has been simmering for a while now but an article by Olga Khazan for The Altlantic triggered a heated discussion all over again at the end of 2013. For a more considered opinion, read these articles by respected educator Mindy McAdams and developer and journalist Noah Veltman. Whether you just want to make yourself more technically literate or want to increase your knowledge of computer languages, then the list below will help you get started.

Codecademy
Codecademy is the most hyped website for teaching yourself coding skills at the moment. The courses are free, it’s easy to get started and it’s a great place to learn the basics even if you have zero understanding of how coding actually works. Codecademy currently offers courses in the following six languages – HTML/CSS, Python, JavaScript, PHP, Ruby and APIs (to find out what these actually are, click on the links).

Coding for GOOD
If you have no idea what <html> means, how websites function or how you can embed Google search or Instagram streams on your blog, then Coding for GOOD is for you. The website integrates links to online resources such as Codecademy, W3schools or Google Maps API documentation (which you may need in order to create your own customized maps) to help you learn to code. At the end of every section it has a step-by-step challenge where you do the coding yourself. The course solves the frequent problem of beginners not knowing where to start.

Coursera
Coursera is an educational technology company that now offers over 500 free courses in different fields. To learn the basics of programming, take a look at the courses “Learn to Program: the Fundamentals” and “Programming Languages.” Check out more courses here. Recently, Coursera also started to offer specializations, which are collections of several paid courses devoted to one topic. If you want to dig deeper into data analysis and statistics (which is an extremely useful skill for data journalists), have a look at the specialization “Data Science” offered by John Hopkins University.

Khan Academy
Khan Academy is a good starting point if you want to learn some programming fundamentals. In the course “Python Programming,” you’ll learn the basics of this programming language and will know enough to be able to talk to developers in their language.

Skillshare
If you are more into web-based seminars (webinars) and courses with a human touch, have a look at webinars offered at Skillshare. It’s a platform for project-based classes offered by professionals in a given subject. For example, in the technology classes you can learn how to build a website using HTML and CSS, how to program iOS apps, which comes in handy if you are thinking of launching a news app, or how to code simple games using JavaScript. Each course costs about $20 to $40.

Udacity
Co-founded by Silicon Valley demigod Sebastian Thrun (who’s helping build Google’s driverless car and Google Glass), Udacity is one of the biggest providers of massive open online courses. The company recently changed its focus to offering courses under a monthly payment scheme, but says it will still make free course content available.

Udemy
Offering partially free and partially paid courses, Udemy is a combination of Udacity and Skillshare. Anyone can create and offer their own course. Start looking for coding courses in the “Technology” section. You can then sort the courses into paid and free ones. You can also chose popular topics such as CSS or JavaScript.

School of Data
School of Data is not actually about coding. The courses posted here are rather about how to crunch numbers, extract and clean your data. While doing it, you will learn some algorithms which will be helpful to understand programming as well. Learning is very task oriented and hands on, which is a great way to get the basic idea of how a particular tool can be used without going into detail. The course in data journalism will be especially interesting for reporters working with programmers.

Treehouse
At Treehouse, you can learn how to how to build websites and apps, write code or start a business. The website operates on a subscription model which means you pay either $25 or $49 a month for access to videos and tutorials. You can also practice live with the code challenge engine. They offer a free trial to see if it works for you. Also, the Treehouse team has created a coding game called CodeRacer so you can learn code in a fun way and compete with your friends.

Mozilla Developer Network
If you are a more text-oriented person, Mozilla Developer Network tutorials are the right choice for you. They offer material related to the construction of websites, web apps and games such as HTML, CSS and JavaScript, from introductory to advanced levels. The tutorials also contain useful links to other resources including “CodeAvengers” which is a fun way to learn to code web apps and games.

More resources

If you want to dig deeper into coding and creating websites and apps, you can also check out tutorials at Code School, the Google Developers University Consortium, Learn Code the Hard Way, HTML5 Rocks and LearnStreet.

For more visual content, check out Bitcast (software development screencasts) and TheCodePlayer (the videos walk you through creating the code from scratch).

There’s also MIT Open Courseware containing lectures from the Massachusets Institute of Technology. You might like to have a look at “Introduction to Computer Science and Programming.”

Bonus

If you are not sure coding is something for you or just want to have a try without enrolling into an online course, try out the short tutorial “Hour of Code” by the Khan Academy. Within an hour, you will learn enough programming to create your own computer-generated drawings. The course gives useful insights into how programming actually works.

Also, check out “The Journalist’s ‘Learn to Code’ Resource Guide” with a collection of learn-to-program resources and recommendations regarding free books and online tutorials.

Written by Natalia Karbasova, edited by Kate Hairsine

Twelve renowned journalists from Latin America and Germany are coming together to explore the future of data journalism. Under the motto “Public Data Goes Social”, the journalists and open data experts will meet for three days in Buenos Aires to discuss the most burning issues in the booming field of data journalism.

The talks are part of a Media Dialogue organized by DW Akademie and supported by Germany’s Federal Foreign Office. You can follow the event, which runs from 13-15 November, on Twitter  via @dw akademie and the hashtag #mdba.

There is a thriving community of data journalists in Germany and Latin America and the two regions are building a reputation for innovations in the field. Read more for a short introduction to the Media Dialogue participants.

Laura Zommer is Executive Director of Chequeado.com, the only NGO doing fact checking in Latin America, and a professor for the Right to Information at the University of Buenos Aires . Her area of expertise are anti-corruption, accountability and open data.

Data projects you like: ProPublica, Poderopedia.

My favorite data visualization: How rumors and unverified information spread on Twitter.

Data set I would like to work with: The designation of judges and prosecutors in the last 15 years in Argentina in order to analyze, for example, what educational background they have or the gender or regional ratio.

Data which shouldn’t be open: Sensitive information that can lead to discrimination, such as sexual orientation, religion and health issues (but only when such data is tied to an identity, obviously not when it’s purely are statistical). Also the names of juvenile offenders.

Something every journalist should know about data journalism: It’s a new way of telling stories that allows you to involve citizen in the process.

Question for my colleagues: Do you have any advice on how to do great data journalism even if the data is not always the best?

___________________________________

Mariano Blejman is a Knight International Journalism Fellow at the International Center for Journalists, manages @Mediafactory_vc, runs the Buenos Aires chapter of Hacks/Hackers and is also editor-in-hack at the investigative journalist web platform @analiceme. Previously he was an editor at Pagina/12.

Data project I like: Dollars for Doctors by ProPublica.

My favorite data visualization: Betrayed by our own data by Malte Spitz at ZeitOnline.

Data set I would like to work with: Internet cables in Latin America, something like Submarine Cable Map.

Data which shouldn’t be open: Health.

Something every journalist should know about data journalism: Data journalism isn’t magic, it’s about building teams.

Question for my colleagues: What is more difficult? Developers learning to write or writers learning to code?

___________________________________

Christina Elmer is a data and science journalist at Spiegel Online. Before that she was part of the investigative reporting team at Stern magazine and worked as an infographics editor for the German press agency, dpa. She also trains journalists in data reporting and online research.

Data projects I like: Interactive German federal election map at Spiegel Online’s data blog and the regional analysis of health indicators at Stern magazine (partly also covered online).

My favorite data visualization: I find the visualization of the drones attacks deaths in Pakistan very impressive, well composed and narrated in an absolute meaningful way.

Data set I would like to work with: Of course, the US surveillance data, to at least be able to show the full extent of their international intelligence actions.

Data which shouldn’t be open: Private information that individuals don’t want to be published.

Something every journalist know about data journalism: Data journalism is more an enrichment than a replacement of fundamental journalistic research methods.

Questions for my colleagues: What was your most encouraging data journalism moment and what motivates you day after day?
___________________________________
 

Gustavo Faleiros is a Knight International Journalism Fellow and environmental journalist specializing in data-driven journalism. In 2012, Faleiros launched InfoAmazonia, a digital map that uses satellite and other publicly available data to monitor the Amazon rainforest. This year he has created a multidisciplinary team called Environmental News Lab and has been launching news projects based on the experience of using interactive maps and earth science data.

My favorite data visualization: the interactive feature Racing Against History by the New York Times.

Data set I would like to work with: Carbon monoxide concentrations.

Data which shouldn’t be open: Personal data, private data.

Something every journalist should know about data journalism: That it is just a name for the much larger transformation happening in all journalism processes, from sourcing information to distributing it.

Question for my colleagues: What are the most innovative forms of data distribution you have seen done by a media company or a journalistic project?
___________________________________

Sylke Gruhnwald is a data-driven reporter with the Swiss media outlet, Neue Zürcher Zeitung. You can find her projects here and here.

My favorite data visualization: Kant – you can read more about it here. Besides this, I prefer bar or line chart over bling-bling maps or fancy stuff.

Data sets I would like to work with: Commercial registry, Swiss income data (Salarium), NSA files

Data which shouldn’t be open: Very personal data such as credit card details, health records, etc.

Something every journalist should know about data journalism: Data driven journalism is not only about fancy visualizations; it is about storytelling with data.

Questions for my colleagues: Why and how did hacking in Latin America develop to the current state? How can one individual foster that development?
___________________________________

Stefan Plöchinger started as a local reporter for Germany’s biggest broadsheet newspaper, the Süddeutsche Zeitung. He was then a political journalist at the Munich tabloid Abendzeitung, a production editor for the now defunct Financial Times Germany, and copy chief and managing editor of the German news site, Spiegel Online. In 2011, he returned to the Süddeutsche Zeitung as the editor of the online edition and managing editor for online business.

My favorite data visualizations: US Gun Deaths and I also love the Partly Cloudy app just because it’s a clever redesign of a classic topic

Data set I would like to work with: Traffic jam data over the years to compare the difference made by building new streets or new railway lines – does reshaping traffic systems reduce traffic or lead to more traffic?

Data which shouldn’t be open: Intimate and private data of individuals who have committed no crime whatsoever

Something every journalist should know about data journalism: It’s not about data alone. It’s really about a new way to explore stories that otherwise couldn’t be told or researched.

Questions for my colleagues: What is your experience with readers – which projects did they like the most and which were a lot of work but didn’t resonate?
___________________________________

Juan Carlos Simo works as a journalist at the daily paper, La Voz del Interior in Córdoba, Argentina and is also part of the national journalist’s NGO, Foro de Periodismo Argentino (Fopea). Even though says he’s not a data-journalism specialist, he has worked on several data projects. One of them is Tragic Roads, a long-term project to build up data about road safety (see for example one of the monthly reports on the subject). Another is about deforestation in Córdoba.

My favorite data visualizations: I love this interactive cartogram made by Andy Tow for Cippec, a local NGO because I think is quite effective at showing the inequity in Argentina and One Race, Every Medallist Ever from the New York Times

Data sets I would like to work with: Homicides related to organized crime in Argentina and femicides. Unfortunately, these aren’t available.

Data which shouldn’t be open: Personal data that people want to remain private. In Argentina, the government publishes people’s addresses and pictures on a website for election purposes.

Something every journalist should know about data journalism: Data is just the beginning. Without journalism, there is no social use of it.

Question for my colleagues: How would you convince editors and publishers of the need to invest in creating specialized areas of data journalism in local media?
___________________________________

Julius Tröger originally started off writing for the German daily newspaper, Berliner Morgenpost but now also codes for them too. His latest data journalism projects are an interactive 3D visualization and analysis of the flight traffic above Berlin and an interactive map of the German Bundestag election results in Berlin by polling station.

My favorite data visualization: Gay rights in the US by Fielding Cage and Gabriel  Dance from the Guardian US.

Data set I would like to work with: Data as granular and open as this for all German states

Data which shouldn’t be open: All personal data.

Something every journalist should know about data journalism: Don’t underestimate the advantage of working with web developers in the newsroom.

Questions for my colleagues: I would like to ask about the state of open data in Argentina and in other Latin American countries and find out how my colleagues deal with it in their daily work?
___________________________________

 Miguel Paz is a Knight International Journalism Fellow at the International Center for Journalists and founder and CEO of Poderopedia, a data journalism
website that maps the political and economic links between members of the elite. Paz is also the president of Poderomedia Foundation, which fosters
innovation within the news ecosystem and promotes the use of new
technologies to increase transparency. He is the co-creator of the
Hacks/Hackers Chile chapter and co-creator of OpenDataLatinoamerica.org.

My favorite data visualization: A code status dashboard I saw in a newsroom. It was a small black screen on the wall that worked like a semaphore for the website performance team. Green lines of code meant everything is ok. Yellow lines of code meant beware. Red lights was “run Forrest run”!

Data sets I would like to work with: The NSA archive and metadata on Barack Obama’s emails

Data which shouldn’t be open: Who am I to say? I leave that to judges and ethic masters

What every journalist should know about data journalism: If you can think, read, write, dont’t be lazy and stay hungry, you can do this and any sort of journalism: keyboard journalism, phone journalism, recorder journalism. Names are just references to tools which are often outdated by our working practice. Someday “data journalism” will sound as dumb as “phone journalism” sounds now.

___________________________________

Sebastian Vollnhals lives in Berlin where he works with OpenDataCity. His projects include Castorticker, a platform of citizen journalism about nuclear waste transports in Germany and Lobbyplag, an evaluation of the influence of lobbyists on the European Parliament.

My favorite data visualization: The display of messages broadcast with Castorticker on a map.

Data set I would like to work with: Beauty and value can be found everywhere so I’d like to work with a data set I’d least expected to work with.

Data which shouldn’t be open: Data openness should always empower people and not be a tool for oppression. Therefore a world without inequality and oppression needs to be built where all data can be free without causing harm. Until then, the question if data should be open is a question about power.

Something every journalist should know about data journalism: Data contains wisdom and it’s your quest to find and share it. Wisdom brings power and it’s your responsibility to use it wisely.

Question for my colleagues: What is the question that needs to be answered next?
___________________________________

Stefan Wehrmeyer works for the Open Knowledge Foundation Germany as a developer and project lead on FragDenStaat.de, a German Freedom of Information Portal. He also likes to hack on public transport data and government transparency.

Data projects I like: OpenSpending and FarmSubsidy are both great projects I worked on that improve transparency of public money in Europe and around the world.

My favorite data visualization: Of course I appreciate good visualizations but as a developer I’m more intrigued by what possibilities are still out there and what techniques are coming up. One example are vector tiles that will change Open Source mapping. You can see an example here.

Data set I would like to work with: A GTFS (public transport schedule) file of Buenos Aires to integrate into Mapnificent.net.

Data which shouldn’t be open: I wouldn’t mind if all publicly available data was openly licensed. Of course, not all data should be publicly available.

Something every journalist should know about data journalism: It’s still journalism.

Question for my colleagues: How can we collaborate better on global stories between different countries?
___________________________________

Florencia Coelho is the Digital Media Research and Training Manager at La Nacion news organization in Argentina and a member of La Nacion’s data team, @LNdata Editor. She coordinated the translation of the Data Journalism Handbook into Spanish Coordinator and in her past life was a lawyer.

Data project you like: One of my favorites (an oldie but goldie) is the Las Vegas Sun’s 2010 Do No Harm series on hospital care. They analyzed almost 2,900,000 hospital billing records revealing more than 3,500 preventable injuries, infections and surgical mistakes and identifying more than 300 cases in which patients died of preventable mistakes. Six laws where passed after their project.

My favorite data visualizations: Visualizing The US Electric Grid by NPR from 2009 and Wind Map, a newer and cooler dataviz by Fernanda Viegas and Martin Wattenberg.

Data set I would like to work with: Argentinian hospital care data like the Do No Harm series.

Data which shouldn’t be open: Personal private data and sensitive information regarding national defense. We also need some creative solutions to certain data such as census information and health information from NGOs so that people aren’t identified (people living in rural areas with very low population can be easy to identify, for example).

Something every journalist should know about data journalism: Maybe today you think these skills are kind of nerdy but they will soon be in your daily essential journo toolbox.

Question for my colleagues: What about making a wiki list of ethical and legal circumstances which arise from the data journalism process?

Go to the event blog

Justin Arenstein is one of the driving figures behind the Hacks/Hackers movement in Africa, where there are currently 13 chapters. Arenstein, a South African, is currently a Knight International Journalism Fellow in charge of the Digital Innovation Program at the Africa Media Initiative. He also is a consulting strategist for Google on data and digital journalism issues (Twitter: @justinarenstein). DW Akademie’s Kate Hairsine talked to him about Hacks/Hackers in Africa.

Is Hacks/Hackers in Africa the same as in the United States?

In North America, Hacks/Hackers is purely bringing together journalists and developers. In Africa, we have widened the definition of what media is. We have defined media as any civic engagements that strengthen civic discourse so this include bloggers, for example. In addition, in Africa we see Hacks/Hackers as having four constituencies. As well as journalists and developers, we have included social justice activists because these are people who sit on a lot of the data. We have also very explicitly included user-interface experts – these are people who design the way normal citizens interact with data. Their inclusion is incredibly important not just because of the technology itself but also because of the challenges you face in Africa in getting information out to average citizens.

What benefits can the Hacks/Hackers model bring to journalists in Africa?

It is a way for conservative and often resource-strapped media companies to outsource their research and development. You get journalists who understand what technology is about coming to these meetings and they suddenly find it’s a meeting of minds – there are people from a technology space and a data space and a mobile space. Then they take this synergy back to their newsrooms and suddenly you have a completely different voice at the newsroom desk.

Invariably in newsrooms in Africa and around the world, news agendas are shaped by people who already have power and influence. These are people in governments or people working for large corporations or working for large non-governmental organizations. The voices that we are getting from this new kind of emergence disruptive space are more crowd-sourced and more kind of grassroots because of the people who get together in these meet-ups. We are also showing in many of the newsrooms – specifically in Kenya and South Africa – that some of these new technologies these guys are experimenting with dramatically increases the impact and reach of these traditional media.

What kinds of things are newsrooms in Kenya and South Africa doing?

Kenya is a bit ahead of the other chapters simply because they have been running longer. They were first established in June 2012 (Twitter: @HacksHackersNBO). In Kenya, some of the hungrier young media, not the ones who dominate the media sector but the up-and-coming challengers, are involved. And specifically, the Star newspaper which is a blue-collar newspaper targeted at a younger audience, has seen a lot of their journalists get involved in the Hacks/Hackers chapter. And they have then come back to their newsroom where they have built things like a crowd-sourcing citizen journalism app using a mobile android-based app as well as an SMS app. (The app can be downloaded for free on Google Play here and you can read more about the app here.) The app does two things: firstly, it allows members of the public to ask questions. So a journalist will say, “I am sitting in a press conference with the minister of education. What do you as the public want me to ask as a question?” That changes the power dynamics completely. It is no longer a bunch of high priests at a political desk in a newsroom deciding what the questions should be. It’s the actual readers who are deciding and who have a direct channel to the interviewee. The other thing the app does is allow the news desk to actually commission people to go out and cover stories. So they’ll put out a call saying: “we are covering the Kenyan army in Somalia and we are looking for these kinds of images from these front-line provinces.” And then citizen journalists who are registered can respond.

One of the ideas behind Hacks/hackers is that developers can help a journalist make data easier to understand by putting the data into a visual form, for example. It’s a fantastic idea to train people to use data, but is there actually any data in Africa for them to work with?

In many African countries, the data is actually locked in paper. So as part of these hacks and hackers activities, we have data liberation scrapathons. That means for a week at a time, people will get together and they’ll target a specific institution or a specific issue and they will go out and source documentation. Then they’ll then scan the documents and scrape the information off that documentation, turning it into structured data that people can start building stuff with. The data is then deposited into a crowd-sourced data portal, Africa open data, which is now the largest data portal in Africa. It’s already three times the size of the Kenyan government’s open data portal. It shows that there is data available in Africa, we simply need to find it, digitize it and clean it for people to be able to use it. Then there is the question of whether the data is credible.

How do you check that?

This is where connections with social justice and other grassroots organizations are essential. For example, where we have scraped data about the GIS locations of rural schools from official government records, we can then turn to other organizations and ask if there is in fact physical buildings where governments say they are. In the case of Kenya, we found that there usually are physical structures but they are nowhere near where the government thinks there are. Sometimes they are up to 10 to 15 km away. Why is that important? Because when the government does its budget resource allocation planning, it uses this data. And if this data is flawed, it means their resource development plans, such as where they are putting in roads, pipelines for water, or electricity, are not going to attend the beneficiaries. So it is essential that there is accuracy. We are finding that keeping these social justice people in the loop helps verify such information.

How much do journalists skills need to be developed though for them to be able to work with data?

That is one of the big challenges. Science-based education in Africa is a rare thing. As a result, Hacks/Hackers has taken a slightly different approach to the US and European approach. In the United States, we hear a lot of talk about computational journalists. These are journalists who are learning to do programming themselves. In Africa, we are trying to build multidisciplinary teams instead. We are bringing in the narrative journalists – the storytellers, and we are pairing them up with people who have some technologist skills.

Why are Hacks/hackers putting such an emphasis on data?

Data journalism gives newsrooms new tools to parse or to analyze or to wrangle information. So suddenly, rather than getting a press release saying there is an epidemic of cholera, the newsroom can also now delve into records from the United Nations and the World Bank and civil society and actually analyze the trends and also what potential causes could be. You can start by mashing up the data, by incorporating climatic information, socio-demographic information, all kinds of information to try to get down to root causes. The resulting story is potentially far richer and far deeper and far more meaningful that the newsroom could deliver otherwise.

Africa Check was launched in 2012, and is a non-profit organisation led by the media development agency AFP Foundation and the Journalism Department of the University of the Witwatersrand in Johannesburg.

The organisation not only examines the public claims of politicians, leaders in civil society, government agencies and NGOs, but also checks the facts journalists use in their stories.

Peter Cunliffe-Jones of the AFP Foundation is the Director of Africa Check and says getting the facts right is the “essence” of journalism. But under the pressure to feed 24 hour news, and working across more topics, journalist may feel they “don’t have the time and the expertise, or even know where to look” to verify information.

Reading an Africa Check report is a little bit like a mini-lesson in the basics of journalism. You also see that this project has the potential to have a high impact on African media and perhaps serve as a model in other countries or regions.

Take this recent example of mis-reporting the HIV rate among South African schoolgirls that appeared in The Sowetan newspaper.

It is an alarming headline that gained a lot of attention. The South African Press Association (SAPA) picked up the story citing The Sowetan, and then the story was reported by media outlets around the world and via social media networks.

By checking quotes with the Health Minister’s staff and using a publicly available source, Africa Check’s investigation found that the real rate of HIV among SA schoolgirls is half of what the media reported: “HIV prevalence among young women aged between 15 and 19 was around 12.7% in 2011”.

Africa Check reports feature a prominent box with key facts

Peter Cunliffe-Jones says it took just a few minutes to check the key facts in this story.

Peter Cunliffe-Jones: “Accuracy of our reports in media houses is fundamental to the role, purpose and survival of journalism.”

To help journalists improve their skills the Africa Check website has some useful tips for fact checking and a list of online references – potentially making it a go-to website for African journalists to start their search for sources of data.

Africa Check’s focus at the moment is South Africa, but Cunliffe-Jones says the organisation will soon broaden its coverage to countries across southern Africa. He says it’s encouraging that South African media outlets are now starting to cite Africa Check’s reports, or have corrected their stories following an Africa Check investigation.

So how does Africa Check fit into media development within Africa?

By having the project based in the journalism department of Wits University, Cunliffe-Jones says Africa Check can carry out fact checking, but also offer teaching and training. Moreover, the project is “fostering a culture of fact checking” which is fundamental to good journalism.

Watch this short video below (from Vimeo user Pokitin) for some more thoughts from Peter Cunliffe-Jones on fact checking in journalism and the Africa Check project.

Peter Cunliffe-Jones from Pokitin on Vimeo.

Author: Guy Degen