Our Journocode founder Marie-Louise has had quite a year: she started January as a data journalism trainee at Berliner Morgenpost. Now, she leads the interactive team at major German publishing house Funke. Yesterday, she was at the News Innovation Forum in London to talk about the things she has learned about doing data journalism.
What was then
In 2012, Simon Rogers gave a TED Talk explaining why data journalism is the new punk. Using data to tell stories was not widespread in the newsroom yet. Journalists became data journalists by accident. They started out as video journalists or worked at the newsdesk when they slided into using data or experimenting with all the possibilities the world wide web offers.
What is now
This has changed. Where self-made data journalists took their first shaky steps a few years ago, now there are established teams of specialists: journalists work side by side with statisticians, designers and web developers, collecting and analyzing large amounts of data or handcrafting interactive maps and new storytelling formats. And there are newcomers like me, who were trained to become data journalists by schools or universities, who write and code as well. At the same time, however, there are still many newsrooms where colleagues who want to work with data are hardly given enough time and resources to do so.
So, if we look at the development of data journalism in the past years: What have we learned? First things first:
Data journalism is journalism. A data journalist still is primarily a journalist. We differ from other journalists in our toolbox, but not in our motives, ethics or topics.
Data journalism is data and technology driven.We use data not only as one of the sources for our stories. Data is our main source, we run on data. We may even discover stories in a data set only while looking at it, unaware of the gems hidden in it. We have the freedom and the skills to pioneer the use of new technologies for journalistic purposes. We never stop looking for new shores.
Data journalism has a wide range. Everything from the quick bar chart visualizing crime statistics to the month-long investigative analysis of bank records can be labelled as data journalism. This is why the structure, workflow and output of data journalism teams can widely vary.
Data journalism is a full-time commitment. Note that, right now, I am not talking about the basic data handling and visualization skills I think every journalist should have for their daily beat. I am talking about the heavy-duty data wrangling, the bigger stories that require more work. You don't do these kinds of stories on the side, between an interview and a shift at the Newsdesk. If newsrooms want well-done data-driven stories, they need to make time for their reporters to pursue them.
Data journalism is transparent. Transparency of sources is one of the cornerstones of journalism, but it is often neglected. Through its deep connections to the scientific and open source community, data journalists are often inherently more conscious of the need for transparency. That can mean everything from listing and linking data sources to publishing step-by-step making ofs or the whole programming code.
And last but not least:
Data journalism is the reader's darling. Whether it is a fancy visualization, a new perspective on a current topic or the possibility to break a complex topic down to your own front door: Readers love data journalism! We see this in the comments section, we can see it on social media, we see it looking at our engagement metrics: Investing in a data team can be the most successful image campaign to excite and retain both readers and advertisers.
Learnings to apply in 2019
That is what the last years have taught me about data journalism. What does this mean for the future of data journalism? Where are we going, what do we do next? I have four main learnings for us to apply in 2019.
Number 1 - Identify your team’s concept:
It may sound banal, but: Knowing what your goal is is the first step towards actually achieving it. In many instances, neither reporters nor their editors know what exactly they mean when they start doing data journalism. They may be driven by the buzzword, by a fancy map they saw in a rival newspaper, or the drive to try out a hip new technology. My tip: Set a goal first, a vision. Do you want to reach a maximum number of people? Do you focus on long, engaging stories? How visual do you want to be? How investigative? Developing a concept will guide your stories, help you judge your team’s success and communicate its value to anyone who needs to be convinced.
Number 2 - Collaborate, both internally and externally:
Data journalism is a team sport. This discipline is not a single-player game. It is based on the collaboration of several disciplines and people, so best you can do is build a multi-disciplinary team that is dedicated to data journalism. Their input, the exchange and the collaboration with both internal and external colleagues, whether from print or radio, another newsroom or another country, is a motor for great data stories. We can see this in big collaborations like the Panama Papers, but also in smaller projects like my team’s project on face recognition where we collaborated with the team of our weekend magazine.
Number 3 - Unleash the potential of data journalism for local reporting.
Many of my favorite data pieces cover neither international nor national topics. Projects like our noise pollution map tell a local story with data. You will also find the most exciting, granular data treasures at the local level. Interactive, data-driven graphics help your readers explore stories that are close to their hearts in a depth that they never could with text or audio. This is added value that readers will reward with trust and engagement with your publication. Which is exactly why more local newsrooms should dare to invest in data teams.
Number 4 - Hire the nerds, hire diverse.
Young people are driven to data journalism, scientifically interested people are driven to data journalism. Half of the data teams in Germany are headed by women, the background, experience and age range within teams varies strongly. Journalism, no matter which kind, runs on the diversity of journalists. So, when it comes to setting up a data team: hire the nerds, hire diverse and don’t be afraid to hire yet unexperienced talents.
If we apply these learnings, I believe that data journalism can have an exciting, diverse and multifaceted future. What worries me more, actually, is the lack of basic data literacy in the newsroom in general. The reason why more data journalism teams have formed in the past years is that our world is becoming increasingly digital and data-driven. As journalists, it is our job to explain and report to the world, to bring up the truth. How can we do that if we are data illiterate?
Many journalists reporting on the latest polls don’t know what a representative study looks like. Many journalists reporting on Facebook’s role in election campaigns don't understand what an algorithm is. For a long time, journalism was considered a discipline in which math was the last thing you need. And especially at the news desk, there is not much time to double-check any methods or samples. But basic data literacy is not for data journalists alone, it is a skill that every journalist should have and will need even more in the future.Awesome images: Phil Ninh 💝