Mapping transit inequality

May 8, 2013 7:18 AM

Dan Grover and Mike Belfrage have mapped transit inequality in the Bay Area after reading a New Yorker piece on the New York City subway (previously). The ways in which a widening income gap are changing the demography of San Francisco have been widely reported of late (previously, previously). The project's code is available if you'd like to try mapping your own city.
posted by liketitanic (25 comments total)

This is awesome, and relevant to my interests/profession. Thanks!
posted by filthy light thief at 7:28 AM on May 8, 2013

This is another relevant post (and an essay I enjoyed).
posted by selfnoise at 7:30 AM on May 8, 2013

There's one for Chicago too (and this followup on the Sedgwick stop).
posted by enn at 7:45 AM on May 8, 2013 [1 favorite]

This is very cool! It should be noted that this is based on the single census tract in which the stop is located-- which can be quite a small area. (As the creators hinted at in their methodology blurb.)

I also would amend the OP's framing-- this isn't really a map of transit inequality per se at all, but a map of income inequality using transit as a spatial organizing framework. If anything, high proximity to transit is being used as a control variable; the project tells you less about transit inequality than it otherwise might. (Which doesn't necessarily reduce its overall value.)
posted by threeants at 7:47 AM on May 8, 2013 [6 favorites]

Certainly not to rag on the creators, because this is neat, but I'm sort of wondering what the instructive takeaway is here. For me, I feel like the most this visualization really tells you is "income levels vary significantly and are associated with place in some way". I certainly can't object to people thinking about cities and urban equity and doing cool pretty things with that, but I think if people end up building off this model for other cities they could push it a bit in some really useful ways!
posted by threeants at 8:14 AM on May 8, 2013

This is very cool! It should be noted that this is based on the single census tract in which the stop is located-- which can be quite a small area. (As the creators hinted at in their methodology blurb.)

The issue with using census tracts is noticeable on the Fremont-Richmond line. I don't think Downtown Berkeley and Ashby are that different, except fewer people live in the same census tract as the Downtown Berkeley BART. See also anything going through Civic Center/Powell/Montgomery/Embarcadero. I mean, one expensive high rise near Embarcadero or Montgomery probably takes care of that. That tells you something about the immediate vicinity of the station, but mostly just that few people live there.

So I guess count me in with threeants in not being sure what we're meant to learn from this.

Also, I think the Ashby BART's parking lot is the size of my current census tract.
posted by hoyland at 8:25 AM on May 8, 2013 [1 favorite]

It is also a very ahistorical way of looking at things. Chicago's Elevated Transit system evolved from multiple private commercial enterprises. In some cases it created those neighbourhoods rather than being created for those neighbourhoods.

I think a more telling analysis would be to look at bus routes and scheduling which are more fluid and changeable and thus a more direct reflection of politics and influence. Living in the fairly affluent Lakeview I am always struck by just how much more pleasant (and how rich and white) the 151 into the Loop is compared to how I find buses around the rest of the city. It is a straight shot single line ride and I never have to wait very long for a bus. Many other routes I ride seem to have much fewer buses and long waits at transfer points.

I'd like to see things like actual time to the Loop and some stats like average and median transit speeds included some measure for waits for each neighbourhood.
posted by srboisvert at 8:29 AM on May 8, 2013

So I guess count me in with threeants in not being sure what we're meant to learn from this.

Well the gist of the article in the (widely reported) link was that middle class neighbourhoods are disappearing leaving only rich and poor. Which if true would make those graphs look a lot more like roller coasters than they seem to.

Of course there is a much easier and accurate way of doing this without all of the math and code - just count the number of check-cashing locations you can see when you emerge from each station.
posted by three blind mice at 8:49 AM on May 8, 2013

So what does this tell me besides that the rich live where they will and the poor live where they must?
posted by ocschwar at 9:11 AM on May 8, 2013

Well the gist of the article in the (widely reported) link was that middle class neighbourhoods are disappearing leaving only rich and poor. Which if true would make those graphs look a lot more like roller coasters than they seem to.

My issue is that I don't see that we should expect these graphs to speak meaningfully to the distribution of income. It looks like the MUNI ones might due to the density of stops (unless those plateaus are due to stops being in the same census tract), but one wouldn't expect to be able to conclude much from the BART graphs because so many stations are going to be in a census tract not necessarily representative of the people who use the station, not because poor people have to go to a station in a rich neighbourhood, but because some of those tracts are going to have few residents.

Of course there is a much easier and accurate way of doing this without all of the math and code - just count the number of check-cashing locations you can see when you emerge from each station.

Before the station was rebuilt, it used to look like the North/Clybourn stop was a check-cashing place.
posted by hoyland at 9:13 AM on May 8, 2013 [2 favorites]

It's interesting that poorer areas still have access to rapid transit. A while ago, there was an article pointing out that in Toronto, the areas with the best access to rapid transit (mainly subways) tended all to be richer than the more distant areas that were reliant on buses and streetcars alone.
posted by jb at 9:20 AM on May 8, 2013 [1 favorite]

One thing this doesn't show is that East Contra Costa Country (Antioch, Brentwood,) which have been paying into BART since day one, got screwed out of it's promised expansions. They went toward richer areas instead.

East Contra Costa is now at least slated to get a crappy e-BART light rail connector, but is still hosed by more wealthy communities.
posted by cccorlew at 10:23 AM on May 8, 2013

So glad this is being discussed.
posted by happysocks at 10:25 AM on May 8, 2013

Another striking map of where the rich people live vs. everybody else in the bay area. Link.
posted by Space Coyote at 10:40 AM on May 8, 2013 [1 favorite]

I think a more telling analysis would be to look at bus routes and scheduling which are more fluid and changeable and thus a more direct reflection of politics and influence.

There are a few bus routes on there on a pull-down menu by the MUNI lines. No schedule info, though. And it sadly lacks my favorite SF bus, the 33 Stanyan, which makes that crazy ridiculous left off of Market onto Clayton and infuriatingly (for a tourist like myself) goes south on Mission before heading north toward the^* inner Richmond.

(^*: I obviously do not live in the Bay Area. I never remember which parts of town require an article and which don't. Nor do I know if Arguello is far enough west to truly be in the Richmond or if it's whatever that last bit of town before Golden Gate Park is. Sorry about that. They're just all squares here.)
posted by maryr at 10:57 AM on May 8, 2013

I found the New Yorker version a bit frustrating because the line graph kind of breaks the spatial-ness of it. You can't cross-reference it with the map very well, especially if you aren't a local. I tried an alternate approach.
posted by veltman at 11:12 AM on May 8, 2013 [5 favorites]

It's interesting that poorer areas still have access to rapid transit.

This interests me too. There are at least a small handful of places along most US subway lines that benefit from an extremely convenient inner-urban rail stop and yet remain (blessedly or stubbornly, depending on your orientation) ungentrified to a large extent. I'm a planner but not too clued in to the literature on gentrification; I'd be really interested to be made aware of dominant theories on what factors make a geographically convenient neighborhood just too resistant for gentrification forces to flip.
posted by threeants at 1:50 PM on May 8, 2013

Oh, that's really cool, veltman. The whole point of the original seems like it should be to use the transit lines as a linear geographic framework that people are familiar with, and yours adds the ability to tap into that. The original might as well be any random series of census tracts, as far as the viewer is concerned.
posted by threeants at 1:53 PM on May 8, 2013

It's interesting that poorer areas still have access to rapid transit. A while ago, there was an article pointing out that in Toronto, the areas with the best access to rapid transit (mainly subways) tended all to be richer than the more distant areas that were reliant on buses and streetcars alone.

SF will probably become more like Toronto in that respect soon. Muni and the TTC subway are both very sparse systems for the cities they cover.

The poorest areas on Muni seem to be 3rd Street on the K/T (the extension to 3rd Street opened just a few years ago), and the part of the Tenderloin near Market Street. Both are rapidly becoming more expensive.
posted by Pruitt-Igoe at 1:54 PM on May 8, 2013

Hmm, also the N near 19th Ave & Holloway Ave. SFSU students I guess?
posted by Pruitt-Igoe at 2:02 PM on May 8, 2013

veltman: nice work!
posted by Pruitt-Igoe at 2:05 PM on May 8, 2013

The poorest areas on Muni seem to be 3rd Street on the K/T (the extension to 3rd Street opened just a few years ago)

I think this is a case where linking a neighborhood to public transit made its housing stock gain market value.
posted by foxy_hedgehog at 5:05 PM on May 8, 2013

When I read "transit inequality", I thought this would be a map of which areas are well-served by public transit, versus which ones require you to have a car (and perhaps correlated with average incomes, proximity to grocery stores and healthy food options, etc...)

Can someone do that map next, please?
posted by otherthings_ at 5:19 PM on May 8, 2013

Another striking map of where the rich people live vs. everybody else in the bay area.

I would say those maps reflect spatial issues as much as anything. The average Whole Foods is 38,000 square feet, while Walmart averages 197,000 square feet. When Whole Foods moved to Downtown Oakland, they took over an old car dealership in a historic building. That's not something Walmart could do. Walmart needs big space, with big parking lots.

We're not rich. Neither are most of my Downtown Oakland neighbors. Yet we can walk to Whole Foods in ten minutes. In fact, the East Bay Whole Foods, for the most part, are in lower income census tracts than the Walmarts.
posted by oneirodynia at 6:07 PM on May 8, 2013

I think this is a case where linking a neighborhood to public transit made its housing stock gain market value.

I agree.
posted by Pruitt-Igoe at 6:26 PM on May 8, 2013