Identifying Contested Spaces with Digital Tools

By Meredith Broussard

I had one personal goal for the Engine 30 fellowship: I wanted to get lightning-fast at using data visualization tools. When you work alone at home, as I do most of the time, you don’t have tech support. Every time I face a technical problem with a data viz tool, I have to go find someone to troubleshoot it– and by the time I’ve tracked down someone who can help, I’ve generally lost interest or passed my deadline. It slows me down, and I hate it. On-demand tech support is one of the major advantages to being in a traditional newsroom. For reals. One of the great pleasures of the fellowship was being surrounded by talented tech people who know their stuff and like to help others learn.

So, I got here, and we were presented with three concepts: contested spaces, arts education, and the question of why art here. “Where is here?” I asked myself. If here is LA, can you find any dataset that shows the prevalence of art? What kind of art are we talking about? What qualifies as art? What criteria are we using to say what is and isn’t art, and furthermore do I trust the people who developed those criteria? You see how the problem got very thorny, very quickly. The topic needed to be narrowed down.

Contested spaces seemed like it might be easier to write about. In order to write about spaces in general, you can tell a story about one specific place. Some students and fellows were interested in LA’s graffiti murals. I liked that because it was more specific than trying to talk about all art, everywhere throughout time. Talking about all art is possible, but it’s extremely difficult and time-consuming to be precise about it, and I only had 10 days to make whatever I was going to make. So. I decided to snoop around and find out what I could about the numbers that go along with graffiti murals in LA.

Data visualization was essential to the reporting process. I’m not from LA, I’m from Philadelphia. If you asked me to write about murals in Philadelphia, I could come up with half a dozen off the top of my head and could identify the three people it would be most helpful to talk to in order to report a story quickly. However, I don’t know people in LA, I don’t know its neighborhoods, and I didn’t have a car. I needed data to narrow down the possibilities.

Planning my time for the week, I quickly realized I had two options. Either I could get lighting-fast at tools, or I could report and write and photograph and edit and produce a story. I couldn’t do both. Learning something well takes time. I decided to concentrate on the learning aspect and use the hypothetical story about contested spaces as a framework for learning.

I started with the police. If I could find areas where there were graffiti arrests, I could probably find graffiti murals. If I could find murals, I could find artists.

It took a while to find the appropriate data and the tools to interpret it. The LAPD has terrific resources about graffiti on its website. It also has a terrific web tool called Crimemapping, run by a company called the Omega Group, that lets you pick types of crimes and see where they happened.

Crimemapping is a very effective data visualization tool, and it’s ideal for the general public. Because I am a journalist with very specific needs, however, it didn’t suffice. One example: I could only get Crimemapping to show me incidents since May 2012, and I wanted to see all of 2012. I also couldn’t immediately see how to determine which specific spots had multiple arrests.

I snooped around until I found downloadable data files with the raw data from Compstat, the system that large police forces use to organize and categorize and track crime. If you saw The Wire, Compstat plays a large role in the police department. That’s always tickled me: Compstat is a real thing that people use, and sometimes they use it just like the characters do in The Wire. David Simon, Ed Burns, and colleagues are creative geniuses.

Here’s a picture of what the raw data looked like:

It’s pretty incomprehensible. Clearly, you need a data viz tool to interpret it.

I found a smaller data set to practice on: the LA Sheriff’s Department allows you to download a CSV file with crime data for the past 30 days. Again, I didn’t need something as complex as Crimemapping for this project. I just needed to find contested spaces in a city that’s unfamiliar to me.

I put the past 30 days’ numbers into a tool called Tableau Public. It’s free and is popular among some journalists.

Here’s the first thing I made:

By looking at the hotspots, the biggest dots, I could quickly identify the locations where the greatest number of vandalism arrests were made in the past 30 days.

The next step would be to download data for the entire year, and then ideally all the data for previous years so I could look at the way the vandalism numbers changed over time. That would more difficult to map because of the change-over-time element. But since I’d invested time in learning the tool the day before, I could quickly put together a model for the data for the entire year.

This project is incomplete. I stopped working on this project once I had a rich data visualization about all kinds of vandalism in the past 30 days. It should be noted that “vandalism” includes four categories of offenses: vandalism felony, vandalism misdemeanor, vandalism graffiti/tagging, and vandalism jail. If I were doing this story for real, not just as an exercise, I would need to filter out the graffiti incidents from all the vandalism incidents.

Even though it is incomplete, this project worked well as an exercise. I looked at the map of Los Angeles, looked at the neighborhoods, and saw where the greatest number of vandalism arrests happened. As a journalist, this suggested to me that these geographic areas were contested spaces. The tool allowed me to draw the conclusion that either art or vandalism was occurring in these spaces. The next natural step would have been to go to one or more of the contested spaces. I would talk to community members, talk to artists, talk to the police, snoop around in public records a bit more, throw in some urban sociology, and weave all of this into a delicately nuanced story that included the voices of representatives from every group interested in a particular contested space in Los Angeles.


The data viz was the tool that allowed me to identify where certain types of contested spaces happened in a city not my own. It was a handy shortcut. It also made for a nifty visual addition to this story. Another nice feature of a data viz: if I do a data viz for research purposes, I can easily repurpose it to run as art alongside my eventual story.

This whole project was an exercise in how to do my brand of journalism better and faster. (In a big picture, my work is about the sociology of everyday life. I approach it as an arts and culture reporter with investigative training.)

I didn’t draw any conclusions about contested spaces in Los Angeles this week. I did learn how I could use a tool to identify contested spaces. I also came up with 547 ideas of new stories I can do on my home turf. I also learned some new storytelling techniques that I plan to use in my own work. I’ll use those storytelling techniques in my work as an educator, helping my students to conceive new ways of thinking about stories.

That feels empowering.

Comments are closed.