[In reading this, recall that mobility means "how far you can go" or "how much area you can cover" in a given time. "Accessibility" or "access" means "how many economic, social, and recreational opportunites that you can reach" in a given time.]
"[The U.S. Federal Transit Administration (FTA)] believes improvements to both access and mobility are key features of a good transit investment. FTA agrees a measure that defines accessibility instead of mobility might be a better representation of the kind of benefits transit projects are intended to produce. As noted, however, it has proven very difficult to measure. Although it is relatively easy to specify a measure such as number of jobs within a specified travel time of a single location, creating a broader corridor or regional measure including calculations to and from multiple locations is more difficult and complex. FTA believes a measure focusing on project ridership will indirectly address access improvements since more people will ride a project that has enhanced access to jobs or other important activity centers. Focusing on the way a transit project can enhance an individual’s ability to get places, rather than just travel faster, is a desirable outcome of the evaluation process. FTA intends to continue to explore how best to do so."
The FTA's Notice of Proposed Rule Making [pdf] that
proposes to shift the criteria for funding
new transit projects from travel time to ridership,
a move that Socrates* had some questions about.
Hat tip to Susan Pantell for reminding me
of this passage.
This is indeed hopeful. I'll lay out a fuller argument on how this agenda might move forward in a coming post.
Question: When FTA refers to the difficulty of aggregating accessibility measures for everyone in a region, do you think they're referring to a logical problem (i.e. the stated task is logically or philophically incoherent), or a data availability problem, or some other kind of problem? It certainly shouldn't be a processing power problem anymore.
* To anyone who suggests that I'm being grandiose in assigning my own thoughts to Socrates, I can only reply that (a) the dialogue in question is broadly consistent with Socratic method, which is Socrates's primary legacy, and (b) Plato made quite a successful career of ascribing his own ideas to Socrates, including many that were not at all consistent with Socratic method, and he doesn't seem to have come to much moral/karmic harm. As long as a fiction is obviously a fiction, it's not lying, it's metaphor.
WalkScore is essentially accessibility minus the jobs right? I mean it can be tweaked to give different weights to different things… SO I think it is a philosophical challenge – what weighting you give everything (is a line near a job paying 15K worth more than one near a job paying 150k? How about to a big grocery store with a big parking lot versus one without a parking lot etc etc etc)
My conjecture: It’s the Self-Esteem Thing.
If you ride transit in California, or work for an agency here, you know all about it.
You likely cannot get usable data because there are many agencies — particularly in L.A. and the Bay Area — and each of them have their own mission, metrics and organizational structure. This means you have many separate systems with no or little means of integration.
That would be OK if we were talking about something like computers, but you add a human element and you now overlay human and cultural aspects to the systems issue.
Even in professional capacities, humans still bring in good and bad personality traits into their jobs. Even though they are supposed to be paid to know better, humans don’t always act better. Employees don’t put jealousies, fears, resentments and wants in a paid capacity.
If enough employees do this, it defines an agency’s culture.
Common cultural disputes are over turf, AMP (authority, method and process — or Who is in charge, What is to be done and How is it supposed to be planned and maintained) and David v. Goliath (think BART versus Bay Area bus systems or L.A. Metro versus the munis).
The FTA knows that if it wants this kind of data, it will have to be the grown-up in the room and tell every agency what it’s supposed to do, and by doing so would have to take responsibility for the agencies.
Correction to my post above: The sentence should read: “Employees don’t put jealousies, fears, resentments and wants aside in a paid capacity.”
I suspect there’s at least partly a philosophical problem here. I think the problem is one of aggregation. You can say what it means for a transit line to improve one person’s access to jobs, education, healthcare, etc, and you might even be able to develop a measure of how much a transit line improves an individual’s access to all these resources. But even once you’ve done that, there’s a question of which transit improvement improves more people’s access more – does it help more to bring 1000 people from a very low access rating to a moderate one, or bring 1000 people from a moderate one to a very high one, or to bring 500 people from a very low one to a very high one? You could use a Rawlsian maximin idea (in which case we want to improve 1000 of the worst off to medium access) or an average value idea (which could go to any of these three, depending on how the numbers work out) or something else. It’s not clear that one measure is more justified than another.