Teaching Data Journalism logo

Click the the image to read our entire series. Public domain image via DARPA

I have stopped using New York Times data visualizations in my training presentations to educators and students. Don’t get me wrong. They’re spectacular. This one setting winter Olympic event finishes to music completely changed my understanding of timed events. I learned about the nightmare of balancing the federal budget. And I figured out why World Cup soccer confuses me.

But given my suspicion that far too few journalism programs are including data training across their curricula, most importantly at the entry level, I need to dial down the “wow.” Why show elite marathoners producing interactives like this (below) when I’m a toddler and my audience is just learning to crawl?

Screenshot of New York Times interactive graphic

Instead, I prefer to show regional papers like the Appleton (Wisconsin) Post-Crescent, which gathered a great deal of its own data to show heroin overdoses on the rise in Wisconsin (below).

Screenshot of Post-Crescent data pacakge

A simple comparison of food costs and benefits is visually revealing.

A simple comparison of food costs and benefits is visually revealing.

Yet even that visualization is created with a sophisticated tool called Tableau. It’s nowhere near the spot we should all start from to introduce data in our classes. We should start as simply as this basic graph (left) from Amy Eisman’s HungryDC project at American University.

I led off this week saying data is arguably the single most important trend in media work right now. I’m closing it with a clarion call to journalism programs: Include data literacy and computational skills as core learning objectives and key curricular elements, starting with every entry-level class.

Get ‘em young

Nick Penzenstadler, now with the Post-Crescent, got an early taste of data journalism in my introductory media practices course and went on to intern, workshop and self-train his way to an outstanding career working regularly with data in his reporting. He argues that every student must have exposure early in their training with opportunities to advance and specialize.

“We had some hands-on experience with ranking, sorting and filtering, and that gets you pretty far in the battle,” he said. “We had an exercise where you ask questions of the data. What can I ask at the start? How much? Who has most? What’s the highest? What’s the lowest?”

This simple start opens a door for students. If I could wave a curricular magic wand, programs would start with these basics at the intro level, include more sophisticated data projects in intermediate and advanced reporting courses, and add one class or workshop series dedicated to data journalism for those who truly want to specialize. These skills are critical in the field today.

“I think the problems we see with journalism are the lack of context and the lack of support for claims,” Penzenstadler said. “Data is a way to add context and support. This is how you separate yourself from bad journalism.”

Let me be clear: Neither I nor the journalism program at the University of Wisconsin-Madison gave Penzenstadler all the tools to work on this compelling database project on unsolved murders in the state. What we gave him was a look at data’s potential, understanding of key concepts and an introduction to basic tools like spreadsheets and mapping.

And let me emphasize: I am not a data and visualization pro. I cannot teach at the level of the folks you’ve seen in this week’s EdShift series, say, Derek Willis or Meredith Broussard. But I am absolutely comfortable that I can teach important concepts and skills at introductory and intermediate levels and envision course design for the advanced stages.

Data Basics for Intro Courses

Introductory reporting courses should include data basics as surely as they include interviewing. Some instructors shy away from this instruction because they don’t feel comfortable themselves.

But believe me it isn’t difficult to get yourself up to speed. And if you take a collaborative approach with your students, you can learn together.

What to cover:

  • Statistical concepts: means, medians, outliers, surveys and sampling, generalizability
  • Analytical skills: correlation vs. causation, change over time, context, common mistakes (I get great traction with students on pieces like this)
  • Spreadsheet basics: cleaning, sorting, counting, filtering, doing calculations (add, subtract, multiply), writing basic functions (mean, median)
  • Interviewing data for insights and stories

How to get yourself ready:

Assignment ideas

For me, the toughest part of learning new things for teaching is assignment design. So let me offer a little boost. You can remix any of these into your own class.

Interviewing data

Thanks to Derek Willis, I use congressional vote data on a specific issue to introduce the concept of asking questions of data. The assignment ties these questions to the most used spreadsheet functions to seek answers. This is a useful assignment to follow training in Excel and Google Spreadsheets.

Mapping

Take locational data that’s of interest to students and use it to create maps using Google Fusion Tables. I have used campus security data mandated by the Clery Act to map crimes around campus and had students gather incident reports on underage drinking and house party busts to map hotspots.

Timelines

Showing change over time can be a powerful element of storytelling. Thanks to the Knight Lab at Northwestern University, we all have a terrific tool called TimelineJS. You simply enter structured data in a spreadsheet to create interactive timelines like this one from UW-Madison student Polo Rocha.

Charts and Graphs

Colleges and universities are awash in data. I take reporting students through exercises creating charts and graphs on such issues as enrollment and graduations rates across different demographics, size and funding of various majors, and grade inflation. That last one always makes for lively discussions. Google Charts is self-explanatory and enables students to get to cruising altitude quickly.

Evolutionary resources

It’s a good thing the data journalism community is as robust and collaborative as it is. Helpful tools, training and people are easy to find.

Let’s change together

Many of you who will read this piece are already part of the choir I’ve been preaching to for a while now. You come to EdShift because you’re open to experimentation and innovation. You’re part of the evolution of journalism education, and I’m happy you’re here to read.

But what I really need you to do is share.

The ideas and steps I’ve outlined here are by no means the only way forward. But I’d like you to take this gospel and share it. Fire it at a curriculum committee member. Give it to your new faculty hire. Send it to a grad student assistant. Tweet, like, post, promote. Try to get as many people as possible to see that we need to improve data literacy in our curricula.

Then I’d like you to share what you do. Have training materials or cool projects your students have completed? Add them in a comment below, so the whole community can benefit and maybe remix them in their own classrooms. You can also put them out on Twitter with the #EdShift hashtag.

Finally, check my work. Did I leave out key concepts to teach? Comment on those. Disagree that data is essential across the curriculum? Let’s open a conversation on #EdShift.

I recently told a funder that I can be accused of being overly optimistic about change afoot in journalism education. But I do see it. I think more people are embracing shifting winds and charting new waters. Dealing with data is a key place we can stimulate progress.

Kathleen Bartzen Culver (@kbculver) is an assistant professor in the School of Journalism & Mass Communication at the University of Wisconsin-Madison, teaching and researching at the intersection of ethics and digital media practices. Culver also serves as associate director of the Center for Journalism Ethics and education curator for PBS MediaShift.

Related