This repository contains my entry to the Divvy Data Challenge. As a Divvy member and civic data enthusiast, I enjoyed digging into the dataset. My entry is an interactive map that shows trip statistics at the station-level upon hovering over a station.
I constructed the visualization using D3.js for the interactive charts and MapBox for the map tiles. Station-level statistics are grouped by the from_station_name. In my exploration, I found that most stations averaged an equal number of inbound trips and outbound trips, though the pattern is time-of-day specific in many stations as you can see by hovering over a station located near a commuter rail terminal such as Union Station.
Analysis and preparation of the data was completed using Python and the pandas library. In the notebooks directory of this repository, you'll find several IPython notebooks that show the code used to generate the CSV files used in the visualization and some of the observations below. To use the IPython notebooks, download this repository and a copy of the Divvy data which is larger than the recommended filesize on GitHub. Place unzipped Divvy directory into the data directory, and then launch an IPython notebook instance from inside the notebooks directory.
In my exploration of the dataset, I discovered some interesting facts about how people are using the system.
- Total Trips: 760,000 or more than 4,000 per day
- Median Subscriber Age: 33 years
- Median Estimated Trip Distance: 1.4 miles, I used the Manhattan Distance function to approximate distance between two points
- Longest trip: 19 miles, from Lincoln and Eastwood to 55th and Shore Dr in an impressive 41 minutes
- Trip Duration 75th percentile: 22.5 minutes, well under the 30 minute time limit
I've only scratched the surface of this extremely rich dataset. I'd love to spend more time working with it or discuss further, send me a message or submit a pull request if you have any other ideas or want to contribute.