For Matthew Kauffman, Covering Education Means Covering Data

The award-winning investigative reporter explains why data matters, particularly on the education beat.

The education beat is fertile ground for data journalism. Covering K-12 issues typically requires reporters to make sense of annual standardized test data, district budgets and local property tax rates. Schools and districts also generate reams of data, addressing everything from student discipline to daily attendance and graduation rates.

This wealth of publicly available information can be a gold mine for stories, but reporters may feel intimidated or overwhelmed, and never do any serious analysis or data crunching.

EWA member Matt Kauffman, an award-winning investigative reporter at the Hartford Courant, recently recounted how he got started diving into data-driven reporting and offered advice for others. (Kauffman responded by email to our questions.)

When did you start producing data-driven reporting and why? How did you get started?

As a young reporter in the mid-80s, I had the good fortune to work in the same newsroom as Brant Houston, a data journalism pioneer who literally wrote the book on computer-assisted reporting. Back then we had these enormous databases – every licensed driver in the state, every registered car, every state employee’s regular and overtime pay, every criminal conviction, every penny donated to political candidates. I didn’t have the skills to work directly with those datasets, but I could see how important data wrangling would be in journalism (and beyond) and I developed an interest in stories that illustrated the power of data.

When Brant left, another great data journalist, Jack Dolan, took his place and he got me on the road to working with Excel and Foxpro, a database manager that uses Structured Query Language. I was mostly self-taught, with the help of Brant’s book, and also did a weeklong computer-assisted reporting “boot camp” put on by Investigative Reporters and Editors. From Jack I learned there was a lot of good journalism to be found by analyzing data not just in one big table, but across multiple databases.

Jack and I teamed up on a story about licensed contractors who were supposed to be vetted by the state for criminal histories. We had a database of contractors and a database of criminal convictions, and with SQL it took a manner of minutes to identify contractors who were also felons. That was the beginning of our reporting, not the end, but that data step allowed us to produce a strong story that changed state policy.

Why should reporters incorporate more data into their stories?

I’m not a programmer or coder or web developer or software engineer. I just have some data skills – and a data mindset – and that’s something all reporters should develop, even those who have told themselves they’re “not good at math.” (C’mon – where’s that growth mindset!) Data helps you find stories and elevate stories. While the data are not the story, there are tips to great stories hiding in data, and there’s no question the world of education is awash in data. Student and faculty demographics, test scores, discipline, teacher quality, budgets, AP classes.

One need look no further than the Houston Chronicle’s “Denied” project, or the Tampa Bay Times’ “Failure Factories” series to see the power of data in education reporting. But beyond those high-impact projects, having the mindset to add a single data point can elevate a story. A story about special-education spending in one district will always be stronger with the added perspective of spending in nearby districts or statewide, or in other states. A story about one high-profile crime at a college will always be stronger with data on the total number of crimes reported by the school; data that’s easy to acquire – for reporters who have the mindset to go get it.

Why does data matters on the ed beat?

Covering education today means covering data. But it doesn’t have to mean a trip back to college classrooms and a degree in computer science. The reporters who have mastered programming languages and statistical software and can do regression analysis in their sleep are true rock stars.

The sophisticated analysis by Josh Benton and Holly Hacker years ago in “Faking the Grade” [published by The Dallas Morning News] is a classic. But for most of us, a decent understanding of Excel will handle the great majority of data work the average journalist will be called on the tackle. And for those ready for more, SQL is fairly intuitive, and it’s easy to find quality online courses and tutorials in various programming languages. What’s important is to recognize the importance of data journalism, and take the first step to developing the skillset and the mindset to pursue data stories.