Professor Edmondson during a workshop with international scholars from the Study of the U.S. Institute 2015 at Ohio University. Photo by Camilo Pérez.

Aimee Edmondson: “All journalists need to get in a data state of mind”

How to create a data journalism class from scratch? Do the students get frustrated trying to get data from public officials? What has changed and what has remained in the data driven journalism for the past two decades? Aimee Edmondson, a professor at Ohio University, is one of those enthusiastic journalists that can convince anyone — even reporters with math anxiety — that numbers and statistics can be a fun and compelling way to tell stories.

by Patricio Contreras | Spanish version

Aimee Edmondson is an experimented journalist and an Associate Professor at the Scripps College of Communication at Ohio University, in the lovely town of Athens, OH. She majored in journalism at Louisiana State University and worked as a reporter in Georgia, at The Augusta Chronicle. Then she got her masters degree at the University of Memphis. She lived there for a decade, working for the metro daily newspaper of the city.

After that she decided to get a Ph.D. on the University of Missouri. For her it was a perfect fit because the nonprofit Investigative Reporters and Editors organization (IRE) was based there. A good combination of academic environment and professional insights.

In the Fall of 2009 professor Edmondson began her data journalism class at Ohio University. There she has been an advisor of the student newspaper The Post. She also coordinates an international contest of the American Journalism Historians Association. And every year, during Spring Break, she travels with her students to the Conference of the National Institute of Computer Assisted Reporting (NICAR), an event aimed at beginners and also advanced reporters.

We talked to her about the data journalism class she teaches at Ohio University, the changes to the topic through the years and how the interaction with young reporters has helped her to acquire new knowledge to integrate to her syllabus. The following is an edited transcription of a conversation held at her office on July, 2015.

— How do you produce a new course on a subject that is still new?
I found it can be incredibly hard, very difficult, to create a course from scratch. Luckily I had a data journalism class at the University of Memphis. It was very basic. I had some training at a conference in 1993. So I had known about spreadsheets and data journalism since the early nineties. And then I did a little bit of self taught type data journalism. I took what I knew from the industry, from the field, that background, and said “I know what they need to know. How do I teach it? How to get them ready to go out into the field?” Part of that was finding the appropriate training from myself. In the IRE teaching resources site there are sample syllabus. I was also very lucky being at the University of Missouri. They taught it, the data journalism class, as a semester long course. So I took those syllabus and took the course myself in an abbreviated form while I was there, so I didn’t need the course myself to get my Ph.D. but I felt like I needed it for teaching the class that I wanted to teach.

— What are the basic skills you are teaching?
The first thing you need to do is you need to understand all journalists, no matter the beat, doesn’t have to be just investigative, need to get in a data state of mind. One of the things you must think is “is there data on that, how do I get the data.”

» Then, we spend a lot of time talking about how to acquire data, either download it from the web, build you own spreadsheets or negotiate with public officials for the data. Even though the public records law says this is public — x dataset — a lot of time they turn down. The assignment is to “identify a public record, a spreadsheet, file a freedom of information act request,” and we work this through the class and then write a journal: “then the public official told me no. And then I responded this way. And then they told me no again. And then I tweeted my response and finally at the end I got the data.”

» And for us that’s just the beginning of getting the information. Keep asking, keep asking. So we go through data state of mind, negotiating and acquiring a dataset, analysing the data and then incorporating the data in a story. So, how to combine traditional shoe leather reporting with the data? How do you put a face on the story? Who do you interview? That kind of things. And then how to write the story so it’s not loaded with numbers.

The basics skills of the spreadsheets will remain the same and get better. A lot of the skills I used in 1993 with spreadsheets, with Excel in particular, have stayed the same. The innovation, the real innovation now, is in the visualization.

— During all this years have you seen an evolution of the interest that the students show about reporting with data?
Yes, it was so hard to get them to sign up for the class at first. They are scared of numbers, they think spreadsheets are boring. I couldn’t filled the class for a number of years. And now I can, the class fills. And we also eliminated some of the prerequisites so they can get the class faster. One thing that we’ve done is we’ve really sold the class to students. Because it’s an elective they can take all this other things. If you are a job candidate and you are up for a newspaper job and this person has the exact same qualifications as you, but they took data journalism, they are getting the job. You need this to make yourself more marketable in today’s environment.

— During the semester, do students get frustrated with the data requests or when they have to deal with the datasets? How is the reaction of the students while learning?
We have a very newsroom environment: I am the editor, they are the reporters and the classroom is a newsroom. Yes, it can be so frustrating. Public officials are turning them down left and right. So, they are getting there grousing and complaining, “this is awful!” But it’s real world. Be frustrated and complain with your colleagues in the class and your editor, your professor, because when you get in the real world this is what it’s like, it’s a struggle. The data negotiation part is the most frustrating for them. They get the skills pretty good so they are usually ready to do the analysis by the time they get their spreadsheet.

— What is the final assignment?
There is a final story where they have to turn in the dataset, which I have approved before, they write the story, it’s ready to publish. What happens at the end of the semester is they have identified where they want the story published and ideally what we are trying to do is flood the Athens market with data stories. Lots of times there are big projects that took fifteen weeks to do. And so, school is out, they finished it, they turned it in the last day of class and so they’ll wait polished it with my feedback over the break and then it’ll run in, like, The Post or WOUB.

— Do you make changes to the syllabus every year?
The syllabus has pretty much stayed the same. And when you see it you’ll see this three big assignments: the data negotiation, I make them build their own spreadsheet from scratch and then the big story at the end. And then we do quizzes on the skills, how to use Excel and how to use Access. The syllabus has stayed the same but I’ve tried to update my skills and show them new things or the latest things. It’s not so much the data analysis part that it has changed over the years, though the software has changed and that’s very hard for a faculty member. But the other thing that has changed a lot since I started teaching this was the data visualization options. Tableau didn’t exist, Piktochart, all this different things didn’t exist. Staying on top of that is a challenge, but a lot of time students find new things and introduce them to me. And I think it is nice to kind of check your ego at the door. Students are teaching me as much as I am teaching them.

— Why news organizations are paying so much attention to data journalism? Is there a struggle between the “he said, she said” reporting and the data reporting?
Smaller media companies are less likely to do the data driven stories. They have maybe people who are inexperienced or at the end of their careers who never learned it. Bigger newsrooms have more of a tradition of investigative journalism, they’ve got more resources. I hate to keep circling back to IRE but I think that Investigative Reporters and Editors has been an amazing driving force since the 1970s for investigative journalism. The training that’s been going on since the 1990s has been massive. And so I think the culture has just kind of built up to that organization first and foremost. That truly has been the driving force in the U.S. for the data journalism.

— Working at Ohio University gives you more space and time or freedom to innovate than if you were working in a newsroom today?
Yes. Not having to feed the beast, write the daily stories.So yeah, a university professor who is a journalist also, definitely has more time to get these skills. The other thing that is new and different is there are more resources online than ever before. There is so much online now that I think I have more time and interest, really, to be honest, to get online and train myself to stay on top of the latest things.

— Do you ever think about “what is next”, how is going to be data journalism, data reporting, tomorrow? What will change, what will remain?
The basics skills of the spreadsheets will remain the same and get better. A lot of the skills I used in 1993 with spreadsheets, with Excel in particular, have stayed the same. The innovation, the real innovation now, is in the visualization. You don’t have to be a coder, you don’t have to know HTML, CSS, Javascript, JQuery, Python, Ruby. You can use — and great media outlets use — Google Maps, Tableau, Piktochart, Fusion Tables. The other thing I think is getting bigger and better is finding connections, especially social media networks, but finding connections between a Senator and a group that they might be organized with, like, maybe on a board of directors, donors. How to find those connections to find stories through a lot of different data sets. “Oh, this Senator is connected to this donor, connected to this person, connected to this special interest and he voted on a bill this way at the State House.” That to me is appearing to be another big frontier.