Without looking it up, I think it's also related to adjacency to a significant metro. But, yeah, the US census uses a binary classification that makes a lot of people assume "urban" means a big walkable city when, in reality, it often includes very dispersed exurbs (including places many would consider basically rural) that are never going to be serviced by public transit among other things.
So a lot of people tend to translate 80% urban into 80% cities which is manifestly not true, and even less dense cities.
> Consistent with previous decennial censuses, changes were made to criteria classifying urban areas following the 2020 Census. Key changes to the Census Bureau’s urban area concept and criteria include:
> The use of housing unit density instead of solely population density. The minimum population threshold to qualify as urban increased from 2,500 to 5,000 or a minimum housing unit threshold of 2,000 housing units.
> The jump distance was reduced from 2.5 miles to 1.5 miles for 2020. Jump distance is the distance along roads used to connect high-density urban territories surrounded by rural territory.
> No longer distinguishing between urbanized areas and urban clusters. All qualifying areas are designated urban areas.
We agree (i think), i'm just quoting the census bureau document.
I hadn't looked in a while and, yeah, the definition seems to have switched a bit though the overall result seems to be fairly similar. The bottom line is that "urban" in the census has a lot broader definition than what a lot of folks think of as urban colloquially.
So a lot of people tend to translate 80% urban into 80% cities which is manifestly not true, and even less dense cities.