College professors share tips, ideas for teaching data journalism

By D.B. Narverson

Veteran educators offered three key tips for teaching data journalism during a CAR Conference session Thursday:

1. Teach so that each lesson builds into the next, and review often.
2. Use topics of interest to students, whether it’s restaurant health inspections, the plight of journalists overseas or a stray dog.
3. Create situations where students can easily find stories in data. Don’t just throw them in the deep end. Instead, construct scenarios and projects for teaching that are more simple and clear-cut than what you normally work with.

Meredith Broussard of New York University and Brant Houston of the University of Illinois moderated a series of 5-minute talks from several data journalism teachers.

Houston said it’s important to be a good translator when teaching students to use data and write data-driven stories. He said to break it down, repeat things often, and go slowly.

“The best first approach is to use Excel,” Houston said. “But real data is confusing, so cook it up, get clean generic data with easy, visible patterns, and build in a payoff for using this cooked data.”

“Cooking the data” means changing names of items or people in spreadsheets you already have. Ensure there is a clear pattern in the numbers and an "easter egg" they can find so that students quickly see where the story is heading.

Scott Klein, assistant managing editor at ProPublica, said it’s important to build each lesson into the next and to have quizzes and assignments that require them to recall what they’ve just learned. The quizzes make the knowledge stick. Klein said it’s important that quizzes on previous lessons aren’t intimidating to students or heavily weighted in the grading scale. The idea is to keep them thinking and using what they’ve learned.

Danielle Cervantes, a professor at Point Loma Nazarene University, said she uses a story about a stray dog she and her sister found as an introduction to a project. Her students then have to use data and documents to find the dog’s history.

Students love dogs, Cervantes said, and the project forces them to follow a due diligence checklist.  The takeaway was to teach using a topic that interests students and provide them with a clear process for reporting and research.

Broussard said it’s important to create a place – often through servers on the university campus – where students can create and publish online without releasing their unrefined work to the rest of the world.

Jeff Kelly Lowenstein, a lecturer at Columbia College Chicago, said he uses food inspection data to get students interested. He teaches them to make sure they understand each piece of the data, as well as how to investigate the numbers and ask questions.

Aimee Edmondson, a professor of media law and data journalism at Ohio University, said she tries to pull students in by getting to their hearts. She uses a data set from the Committee to Protect Journalists that shows how many journalists have been killed in the line of duty.

Christian McDonald from the Austin American-Statesman said he uses regular expressions to teach his students how to sift through and work with data.

Matt Waite, professor of practice at the University of Nebraska-Lincoln, has his students use Jupyter Notebook. The online program allows them to code and manipulate data while keeping track of each step they make.

D. B. Narveson is a mass communication senior at Louisiana State University. She is editor of LSUNOW.com and a member of the Manship News Service Statehouse Bureau covering the Louisiana Legislature.