On Jan. 25, the Global Editors Network (GEN) held a live Slack chat on how to conduct collaborative data journalism projects. Pierre Romera, Rachel Glickhouse, Arun Karki, and Gian-Paolo Accardo led the discussion, which was moderated by Marianne Bouchart and Simon Rogers.
The leaders gave advice on launching a collaborative data journalism project, common mistakes that people make when carrying out a project and useful tools for data journalism collaborations.
First things first
What do you do when you and your team uncovers data that they realize deserves a deeper look? How do you move forward and launch an investigation?
“I’d say, start with contacting the people you’ve already worked with,” wrote Accardo, editor and cofounder of VoxEurop. “Then start building up a formal consortium or network of partners.”
Bouchart, founder of HEI-DA, which is a media nonprofit building data journalism around the world, also asked the leaders the most important thing to get right when collaborating on a data journalism project.
Accardo highlighted the need for establishing clear roles for everyone involved. “Clear roles, responsibilities and authorities probably has the additional benefits of reducing overall stress in the organisation,” he wrote.
It’s important that everyone involved in the collaboration understands the data that was collected and how to interpret it, said Glickhouse, who runs the Documenting Hate project at ProPublica. The project aims to collect reliable data on hate crimes and incidences of bias for use by journalists, researchers and other organizations.
To make sure that the entire team properly reads the data, Glickhouse recommended giving everyone materials that clearly outline all of the necessary details, and to offer continuous training for the entire team.
It’s also important to consider the people that you’re going to work with, said Romera, chief technology officer at the International Consortium of Investigative Journalists. Working with people you trust is critical, as any collaboration requires open dialogue. “We developed a concept of ‘radical sharing’ where every journalist must play his/her part by sharing all findings,” he wrote.
He added that collaborators should be invested and enjoy working on the project: “A collaborative project is a lot of work so you better enjoy it.”
One of the common mistakes that Glickhouse regularly notices is that collaborative partners assume that everyone who shows interest in an investigation will participate. “DocHate has a lot of partners who join but not everyone finishes onboarding and even fewer end up doing reporting,” she wrote.
Many long-term projects also run the risk of fading enthusiasm from team members, according to Accardo. In this case, it’s critical to have strong leadership to keep contributors motivated. “Involvement will highly depend on the coordinator’s ability to run the project and keep up the pace and motivation,” he wrote.
Glickhouse emphasized that a leader should not just be the person with creativity, but someone who can handle the day-to-day details of the project. This includes assigning roles to team members, monitoring deadlines and making sure that accurate records are being kept.
When Bouchart asked the speakers about tools that they would recommend that facilitate collaborative data investigations, many of them suggested Google. Apps such as Google Docs or Google Sheets make it easy for multiple collaborators to edit a file at the same time and to see the changes in real-time.
Karki, the founder of the Center for Data Journalism Nepal, also suggested using encryption software, which can be particularly useful when you’re dealing with sensitive information. “We have mostly used PGP and encrypted file sharing in emails,” Karki said. “These [files] can be shared on Google Drive or Dropbox but only for limited [periods] of time.”
However, Accardo cautioned against relying entirely on online communication, recommending some classic methods that can prevent the problems that modern technology has yet to fix. “Phone, one-on-one or even in person meetings solve many writing or language-related issues,” Accardo wrote.
This article was originally published by the IJNet. It was edited and republished on Data Journalism Awards website with permission.
This article is based on a monthly Slack discussion organized by the Data Journalism Awards. Join our Slack channel for the next discussion.
Image CC licensed by Unsplash via Markus Spiske.