How the Chicagoans use ride-hailing service during the May 28 — June 1 unrest?
By Hanzhang Yang
Hanzhang Yang is an urban planner and rookie data analyst.
Protest and Ride-hailing
The George Floyd protests in Chicago are a series of civil disturbances currently occurring since May 26 in the city of Chicago, Illinois.[1] During the unrest, how Chicago residents and visitors use the ride-hailing service (Uber, Lyft, Via, etc) and its pattern become my interest in this data exploration.
Chicago’s “Transportation Network Providers — Trips” is an open dataset reporting all ride-hailing trips since November 2018. So far, this dataset already includes more than 169,000,000 rolls of entries, or, 169 million trips.
The image above contained 849,799 points, but why the dots seems so scattered? Because Chicago protects TNP and Taxi trips privacy by grouping the geolocation of pickup points and dropoff points to census tracts or even community area level. To know more about how they protect data privacy, check this post.
Patterns of ride-hailing during protest
This time, I selected two time periods as my research timeframe. The first is from May 24 to June 6, 2020. The second time period is from May 28 to June 1, covering the acme of the protest.
From the above two tables, I noticed that statistically, there are no big differences between “normal days” which already bothered by COVID-19 and the days during the protest. So I move to looking at the daily number of trips per community area during the protest, these trips are assigned to their pick up community areas.
Although the overall patterns and hotspot of TNP trips did not change during the protest (May 28 — June 1), the number of trips decreased by about 30%. So how these trips distributed temporarily?
By using moving average, I smoothed the pattern of TNP trips’ time-series plot to a more pleasant level. Surprisingly the TNP trips show a small spike on May 30, which might be used by the protesters moving to their destination. Also, during May 31 and June 1, the number of trips dropped dramatically. Since I previously guessed many TNP trips during the protest period were used for gathering at the Chicago downtown, I created a map showing average daily trips per community area, by dropoff location, from May 28 to June 1.
Interactive Map
Also, this research generated an interactive map showing all the trips between May 24 and June 6, by using Folium.
Conclusion
The author’s current data analytic skill is not able to generate a map linking pickup and dropoff points from the dataset, if such a map could be produced, it should be able to show a more descriptive pattern of the ride-hailing trips in Chicago. The privacy protection while important and meaningful, also limited the dataset’s ability to perform detailed analysis.
Data Source
City of Chicago, Boundaries — Community Areas (current)
Chicago Department of Business Affairs & Consumer Protection, Transportation Network Providers — Trips
Reference
[1] George Floyd protests in Chicago, https://en.wikipedia.org/wiki/George_Floyd_protests_in_Chicago?oldformat=true
[2] Chicago Tribune, https://www.chicagotribune.com/news/breaking/ct-loop-protests-property-damage-20200530-ncvxjtwglnfoddlyj4yuhd6lwm-story.html
