By census block-grouped ridership distributions, we mean for each census block, we check the proportions of usage of each mode of transportation.
Can serve as a decent starting analyses, that can be modular enough to replace with improved analysis such as replacing boarding locations with predicted home locations.
Works well if we have access to a census table that gives aggregated ridership preferences of people by income level.
We can compare lift card users with ridership preferences of people with 'low-income'.
By census block-grouped ridership distributions, we mean for each census block, we check the proportions of usage of each mode of transportation.