I've worked with the data from 2017-2021. The trip info files from before 2020 included the following fields: 0. trip_id
- start_time
- end_time
- bikeid
- tripduration
- from_station_id
- from_station_name
- to_station_id
- to_station_name
- usertype
- gender
- birthyear
The station names and IDs are not terribly helpful in terms of mapping, etc., so in the DataMunging notebook, I documented how to merge them with station data, to end up with the following fields:
0 trip_id 1 start_time 2 end_time 3 bikeid 4 tripduration 5 from_station_id 6 from_station_name 7 to_station_id 8 to_station_name 9 usertype 10 gender 11 birthyear 12 start_lat 13 start_lng 14 end_lat 15 end_lng
Starting in 2020, DIVVY dropped (a) trip duration, (b) bikeid, (c) gender, and (d) birthyear. Losing c and d are unfortunate, as they allow some demographic analysis of who is using the Divvy system. It renamed the (a) station fields, (b) the time fields, and (c) usertype became "member_casual". It started recording (a) the sort of bike used ('ridable_type') and (b) lat/long itself.
0 ride_id 1 rideable_type 2 started_at 3 ended_at 4 start_station_name 5 start_station_id 6 end_station_name 7 end_station_id 8 start_lat 9 start_lng 10 end_lat 11 end_lng 12 member_casual