A couple days after the Bay Citizen Bike Accidents Tracker launched, Steven Vance, a young blogger, hardcore biker, and urban planner in Chicago, released his own version for his city and gave us a nice shoutout for helping to inspire his tracker:

Anywhere you've got a city with bikers, I think you'll find demand for this kind of vizualization and analysis. in the days after our launch, I saw retweets about the project from all over the world asking for this kind of tool where they live (Milan, Berlin, Portland OR, New York, and Mexico City, among others). So I was really glad that Steve took the next step and built his own, and wanted to pick his brain on the process in the hopes that others can do the same.

Q&A

Why'd you make this project? What's the "problem" you're trying to dig into?

I made this project for two reasons: One is to continue practicing my GIS skills and to learn new software and new web applications. The second reason was to put the data out there. There's a growing trend for governments to open up their databases, and your readers have probably seen DataSF.org's App Showcase. But in Chicago, we're not seeing this trend. Instead of data, we get a list of FOIA requests, or instead of searchable City Council meeting minutes, we get PDFs that link to other PDFs that you must first select from drop down boxes. But both of these are improvements from before. 

I would love to help anyone else passionate about bicycling in Chicago to find ways to use this data or project to address problems. I think bicycling in Chicago is good for many people, but we can make it better and for more people. 

How'd you get the data?

I asked the Illinois Department of Transportation for it after the Chicago Police Department sent me a response to my FOIA request that they don't keep "quick access" records of bicycle crashes. IDOT obliged and sent me the data. 

Did the data come to you pretty clean?

The data came very clean. I think some of the fields were miscoded (as strings instead of integers), but it was an easy fix. The data I received had many parts of the crash report already removed (but you ask about that in a different question). I'm not sure if all the records were there. Someone tweeted at me to say she couldn't find her crash in the map. I'm not sure how to explain that.

What do you wish was different about the data?

IDOT did not supply the exact time of the collision (only the hour of the day), the address of the collision, the direction the bicycle was traveling, the direction the other vehicle (or pedestrian) was traveling, or if the police assigned "fault." I'm sure there are other pieces of information that would be useful. Other information was probably removed for privacy reasons, like the police report number.

How did you build the project?

I used uDig to quickly copy and paste rows from one year and paste them into the shapefile of another year. I then used QGIS and the fTools plugin to add XY coordinates to the encoded points (no addresses were provided - if they were I would have geocoded it with BatchGeocode myself). I used QGIS again to reproject it as WGS84 for Google's sake. QGIS exported the data into a clean CSV file which I uploaded to Google Fusion Tables. From that point, it was a simple matter of copying the map's embed code and pasting it into my website. 

What jumped out at you when first saw the mapped data?

I've been bicycling in Chicago since 2006 when I moved here to start school at UIC. I then worked for the Chicago Department of Transportation surveying locations all around the city (I can actually show you a map of the 2,000+ places I visited) for new bike racks. I believe I know the city's layout, traffic and transportation quirks, and bicycling conditions quite well. I don't think we can draw many conclusions from this map except to target our infrastructure expansion (new bike lanes, and hopefully some protected or separated bike lanes in some areas) as well as our education efforts for people who bike and people who drive. 

I think some enterprising students could delve further into this data and perform some more advanced analysis using GIS. The people in Chicago responsible for implementing bicycle facilities and education could use that research to guide current or new strategies to encourage more people to bicycle in the city, and to reduce the number of collisions. 

What kind of response are you getting so far?

Within minutes after I tweeted a link to the map on Thursday morning, someone responded "It's amazing how some streets are just solid crashes like Milwaukee Avenue, Damen Avenue, Chicago Avenue." Milwaukee Avenue seems to have dot after dot after of crashes. People are recognizing what they already seemed to know: A ton of people bike here on this street, or there on this street. But I don't think they knew that so many collisions have occurred. 

So, what's next?

I hope to motivate other people to take a tougher stance on bicycling in Chicago and demand more facilities for bicycling and safer conditions for bicycling, whether that be removing a parking lane on Milwaukee Avenue to provide a buffered bike lane, or increasing traffic enforcement. I also want people to see what's possible with the data that is being collected, but also see that there is some data that isn't being collected but should. For example, how many people are bicycling down Milwaukee Avenue each day? Is the ratio of people crashing to people bicycling higher (worse) than on other streets?