4 ms·
No, the opposite is true. SF population increases by 20% because of the net increase due to commuters. Table 2 on page 11: http://www.census.gov/hhes/commutin
by muzz 13y ago
No, the opposite is true. SF population increases by 20% because of the net increase due to commuters. Table 2 on page 11:
http://www.census.gov/hhes/commuting/files/ACS/Commuter%20Adjusted%20Population%20Paper.pdf http://www.census.gov/hhes/commuting/files/ACS/Commuter%20Ad...
- SamReidHughes 13y agoThat's not the number the parent is talking about. The question is how many people are commuting out of the city, not the difference between in and out.
- muzz 13y agoSure, but his claim is still wrong. Even nearby San Jose has a much higher proportion of reverse commuters than San Francisco. http://www.census.gov/hhes/commuting/files/ACS/Table3.xls http://www.census.gov/hhes/commuting/files/ACS/Table3.xls
- tomkarlo 13y agoThe proportion of reverse commuters in a region has nothing to do with how many commute in each day. It's just reverse commuters divided by residents. As far as the increase due to commuters, 20% is very low for any area with corporate offices and not just residential neighborhoods. Manhattan grows 95%. In the Bay Area, Mountain View has 75,000 residents, and Google alone brings 25,000 people in each day. Palo Alto (the next town up the 101) also grows ~80%.
- muzz 13y agoYour original claim is still incorrect. Even nearby San Jose has a higher proportion of reverse commuters than San Francisco, much higher: http://www.census.gov/hhes/commuting/files/ACS/Table3.xls http://www.census.gov/hhes/commuting/files/ACS/Table3.xls