By Jayden Khatib.
Getafe, Madrid.
Data journalism is an increasingly popular but often misunderstood mode of storytelling in newsrooms across the world. St. Louis, Missouri-based Chalkbeat data and graphics reporter Kae Petrin shows how data journalism can enhance reporting and be much simpler to begin doing than it may first appear.
Courtesy Kae Petrin
Computer-assisted reporting, data journalism’s predecessor, has been around in newsrooms since the 1950s. Early examples include CBS attempting to predict a presidential election with a mainframe computer in 1952 and The Detroit Free Press’s Phillip Meyer’s use of a mainframe computer to utilize survey data about a series of local riots.
Later, in the 2000s and 2010s, journalists started using more data visualization in addition to the old CAR techniques, birthing modern data journalism. Today, the subfield combines elements of journalism, design, statistics and computer science and is popular across the world.
Getting into Data Journalism
Petrin first became interested in using data storytelling to enhance reporting as a college student. However, their alma mater, Washington University St. Louis, does not have a journalism school, much less a data journalism program, so they began their journey of self-education on the topic by taking design classes and eventually completing a non-profit coding boot camp that teaches Python, front- and back-end web development, SQL, and Java.
According to Petrin, although going to a boot camp is uncommon, it is very common for data journalists to come from non-traditional backgrounds, rather than a formal data journalism program.
“I don't meet that many people who've done coding camps specifically, but I think non-traditional training is pretty common,” said Petrin. “I've talked to a lot of people who started out in programming, or maybe were an accounting major, or they started out in journalism and then learned the skills later.”
Early on, Petrin had to create their own opportunities for data storytelling. They started their career as an editorial assistant at a local magazine in St. Louis that didn’t have a data visuals team and started sneaking elements of data journalism like Tableau graphics into stories as a way of building up their portfolio.
Eventually, they moved on to St. Louis’s public radio station, where they got to use Python, SQL, and JavaScript to create investigative reports for web and radio, before landing at Chalkbeat, a U.S. nonprofit newsroom covering education at the national level as well as in eight local communities.
Petrin explained that working in data journalism is slightly easier in an entirely digital newsroom like Chalkbeat as opposed to a radio station or a magazine because online journalists have more freedom to work outside of a traditional radio or magazine production schedule, which may be one reason why data journalism jobs are more common in newsrooms with large online presences.
Petrin created this interactive graphic to show how COVID-19 testing is changing in New York City Public Schools. Screen-recorded by Jayden Khatib.
The Day to Day Responsibilities of a Data Journalist
The data visuals team at Chalkbeat takes requests from the rest of the newsroom to create graphics or analyze data and works on their own projects, meaning that Petrin’s job can be very different day-to-day.
“This week, I've pretty much spent the whole week doing my own reporting and editing. I've just been tinkering with an interactive graphic we’re working on as a team don't really have any deadlines,” said Petrin. “And then other weeks, like eight graphics requests will come in simultaneously, and we're all rushing, you know, to try to get everything done.”
One aspect of data journalism that students and people new to the subfield often find daunting is finding the initial data to work with. While education journalism is traditionally considered an less prestigious beat than, say, politics or business, something that outlets like Chalkbeat are trying to change, it presents an unusually high amount of opportunities to work with data and graphics to help readers understand stories.
“There's a lot more mandated data production [in education] than there is on a lot of other beats,” said Petrin. “Sometimes it comes from the places that we know are going to be releasing enrollment or demographic data at a certain date, and we keep an eye out for it.”
However, sometimes the data isn’t readily available, and in that case, Petrin and the team at Chalkbeat often end up filing FOIA, or Freedom of Information Act, requests.
FOIA is a U.S. federal law that gives the public the right to request access to documents from federal agencies. Individual states have their own, typically similar public access laws, and on both the state and federal levels this process can be costly, slow and time-consuming. Still, FOIA requests are a key resource for data journalists.
“A lot of the reporters at Chalkbeat are doing sort of their independent investigative work,” said Petrin. “If they run into something that's data specific or they're having trouble getting a specific database, [the data visuals team will] help them revise the FOIA.”
Petrin created this interactive quiz for an article about COVID exposure in New York City Public Schools. Screen-recorded by Jayden Khatib.
Advice for Aspiring Data Journalists
According to Petrin, it is important for people interested in data journalism to explore exactly what type of data journalism they’re most interested in. The job title “data journalist” can refer to someone like a newsroom developer who mainly works on the technical parts of a data-reporting project as well as journalists who incorporate data analysis and/or visualization into their original reporting.
One other thing that Petrin emphasizes for newer data journalists is to make sure that your soft skills are as strong as your technical skills because things like “knowing how to partner with reporters” can sometimes “even be more important than knowing the right programming language to make the graphic.”