Digital Publishing

How Newsrooms are Using Data to Report on COVID-19

Posted

Raise your hand if you have Johns Hopkins University’s coronavirus world map in your bookmarks?

For reporters covering the coronavirus pandemic (and let’s face it, most of us are in one form or another), the university’s interactive dashboard has become a daily stop. It provides a powerful, real-time look at the devastating toll COVID-19 is having across the globe. Outlets ranging from CNN, the Washington Post, the Associated Press and newsrooms in-between utilize the data provided by Johns Hopkins to inform their reporting.

The federal government also appears to regularly rely on the data provided by the university, which isn’t surprising because it’s unlikely anyone reading this story has actually visited the COVID Data Tracker maintained by the Centers of Disease Control and Prevention. In fact, then-acting Homeland Security Deputy Secretary Ken Cuccinelli was mocked back in February for turning to Twitter to ask for coronavirus data when John Hopkins’ map briefly went down due to a power surge.

The dashboard, which as of this writing has amassed north of 820 million pageviews since January, was conceived over a cup of coffee back in January between Lauren Gardner, an associate professor of civil and systems engineering at Johns Hopkins, and Ensheng Dong, a first-year Ph.D. student.

What began as Dong manually entering case data has since been replaced with a largely automated process that pulls numbers directly from health departments across the globe. Nevertheless, it takes more than three dozen people to monitor the data feeds for anomalies and errors because there’s no global standard for countries reporting their figures. That includes here in the U.S., where a vacuum of leadership in Washington, D.C. has led to a Wild West of data reporting, with states, counties, and municipalities all doing their own thing.

“There has been no standardized method established by state or federal government, which is a difficult position for reporters,” said Doug Donovan, who won a Pulitzer Prize as a reporter at the Baltimore Sun prior to joining the university’s media relation’s team back in 2019. “Typically, reporters need to know how such critical information is gathered and reported to determine if public officials are following the rules.”

The variances in reporting can be mind-numbing. Massachusetts reports confirmed cases by county, but aggregates deaths for Dukes and Nantucket Counties together. Rhode Island reports confirmed cases by county, but only reports deaths at the state level. Utah merges data from smaller counties together and reports them as county groups.

Despite the hurdles, the team of experts at the university manages to find the commonalties in data needed to compare and contrast different parts of the country. But don’t ask Donovan how the school’s scientists do it.

“That’s way above my pay grade,” he joked. “When you have the applied physics lab involved, you’re dealing with some complicated stuff.

Donovan mostly spends his day dealing with reporters and handling media requests, and said journalists have been valuable when it comes to determining which data to break out and report on. One example was a push for the university to report national data on a county level, something it wasn’t doing at the start of the pandemic.

“It became clear through media requests that a county by county breakdown was going to be the best way for them to utilize the information,” Donovan said. “That definitely got a sense of urgency behind it. Even a lot of the state health departments wanted to do it that way as well.

Despite the popularity of John Hopkins’ Coronavirus Resource Center (my go to is the university’s testing trends tool, which tracks cases and tests), Donovan said there are many features that reporters tend to overlook. For instance, journalists are able to download all the raw data amassed by the university using GitHub, which contains every reported data point for every county on every date since the pandemic first began to hit the U.S.

“The best reporters have explored that data and analyzed it on their own rather than calling us to do that for them,” Donovan said. “We can only point them to where the raw data exists so that they can conduct their own analysis and make their own determinations on how best to use it for their respective audiences.”

After initially amassing their own data about the outbreak in China, the New York Times turned to Johns Hopkins when cases began to spill out into Europe and other parts of the globe. But according to Archie Tse, the graphic director at the Times, the newspaper began tracking cases individually when they started to pop up in the U.S.

Originally tracked in a Google spreadsheet by national reporter Mitch Smith, what began as 15 cases quickly ballooned into hundreds, then thousands. Tse said he and his team wanted to do a more detailed map of the outbreak in the U.S., but at the time Johns Hopkins wasn’t providing the data on the country level, so they gathered it themselves.

Months later, Tse said on any given day there’s probably around 70 people working to maintain and monitor the data, as well as manage the more than 60 pages of coronavirus data active on the Times’ website. The project became so massive Tse said they created an election results style publishing system to make it easier to maintain all the coronavirus tracking pages.

“You have to attend to these pages. You can’t just let them sit on autopilot,” Tse said. “You need to examine them in the context of what’s happing right now, and figure out if we’re doing the right thing right now for our readers.”

Amassing all the data in the U.S. themselves allowed the Times to uncover certain facts about the virus early on, including its deadly impact on nursing homes and long-term care facilities, which as of this writing have been associated with more than 40 percent of all coronavirus deaths in the country.

“All the data on the clusters we’ve been doing have been feeding a lot of stories on meatpacking plants, colleges, and other outbreaks around the country,” Tse said. “It’s given us a lot of fuel for stories that the national desk has been doing, which has been a huge plus for us.” 

Rob Tornoe is a cartoonist and columnist for  Editor and Publisher, where he writes about trends in digital media. He is also a digital editor and writer for the Philadelphia Inquirer. Reach him at robtornoe@gmail.com.

Comments

No comments on this story | Please log in to comment by clicking here
Please log in or register to add your comment