The existing trip_start_time field in the Chattanooga data was calculated by sorting each trip by stop_sequence and getting the arrival_time of the earliest stop in each trip. In most cases, this was where stop_sequence = 1. However, there were some trips where stop_sequence = 1 did not exist, so the arrival_time of stop_sequence = 2 was taken (or stop_sequence = 3, if stop_sequence = 2 also did not exist).
Possible alternatives:
-
Pull the arrival_time of the first stop in each trip directly from GTFS (use gtfs.csv in Teams under WeGO-Data > data-used-for-analysis > Nashville GTFS (static and realtime) > gtfs.csv)
-
There’s a field in the original RideCheck data called 'TRIP_START_TIME' which could be used instead, but many rows are missing a proper start time (i.e., TRIP_START_TIME = '1/1/00' instead of '1/1/00 18:30'). We will need to find out these missing stop times.
The existing trip_start_time field in the Chattanooga data was calculated by sorting each trip by stop_sequence and getting the arrival_time of the earliest stop in each trip. In most cases, this was where stop_sequence = 1. However, there were some trips where stop_sequence = 1 did not exist, so the arrival_time of stop_sequence = 2 was taken (or stop_sequence = 3, if stop_sequence = 2 also did not exist).
Possible alternatives:
Pull the arrival_time of the first stop in each trip directly from GTFS (use gtfs.csv in Teams under WeGO-Data > data-used-for-analysis > Nashville GTFS (static and realtime) > gtfs.csv)
There’s a field in the original RideCheck data called 'TRIP_START_TIME' which could be used instead, but many rows are missing a proper start time (i.e., TRIP_START_TIME = '1/1/00' instead of '1/1/00 18:30'). We will need to find out these missing stop times.