To continue from a couple of post ago, I want to look into why walking spiked do much for those making under $10k between ages 40-44.  This is an esoteric point, but it feeds into debates about the behavior of the poor and the provision of infrastructure for the poor.   I had to write this post before I could feel better about other posts.  Hopefully this will not read like I’m trying to clear a logjam, even though  I clearly am.

To revisit, there is the graph of average walk trip length by age for the poorest and wealthiest survey respondents in the 2009 National Household Travel Survey (NHTS).  Remarkably similar, except for the spike among poor 45-49 year olds.  What gives?


To look at this, I wanted to see if it was an anomaly of the NHTS itself, by comparing the walk trip lengths  for 1995, 2001,and 2009.


Well there’s the smoking gun.  It’s work and family trips on foot that are elevate, with an average trip distance of over 2 miles (40 minutes at walking speed).  What’s fishy about this now is the rise in walking trip length overall.  So I wanted to see of the poor were taking more walking trips, or just longer ones.  If a poor person has to take longer trips to get to what they need, that’s different than if they need to take more walking trips because they prefer to.  Their preference can be one of cost savings, unavailability of traffic, transit or bicycle, or other reasons.  Many would call this deprivation, but I’m not willing to put a value judgement on it.


Again, we see a noticeable rise in walk trip numbers between 1995 and 2009.  This is dramatic, but suspicious.  It seems too radical a rise to be real.  Just like me, the NHTS always seems to times its studies with recessions, so some of this could be the artifact of recent recessions.  1995 (1991), 2001 (2001), and 2009 (2009) were all within recent memory of recessions, and recessions would have impacted the poor the hardest.  This is a possible explanation for this rise.

Another possibility I that the data itself is flawed.  NHTS data extraction tool is very convenient, and very explicit in what it provides.  When I see missing bins for different trip purposes, as I do with NHTS data I get worried tat the data is not reporting as it should.  There are “undetermined” and “other” bins I don’t look at because I don’t know what they represent.  There are trips for certain groups that only have the 1-2 mile trip reporting, with no “Less than 1 mile” reporting.  There are even trip groups that only show 4-5 miles with nothing else.  These are alarm bells.

To look at the validity of the data, I took the above graph and simply expressed it as a percent, to see if the mix of purposes had changed.  For work, not much, but social and school trips were eating into family.  Is this a change in the society of the poor, or an artifact of bad data.  This has stuck in my craw but not so much I;m going to hold up other posts for it.


Another is that walking really is rising among the poor as well as other classes.  That’d be nice, but I am no longer sure that the NHTS is definitive evidence of this.

There’s an unresolved and unsurveyed* debate about the effect of bike lanes on gentrification.  The poor should be better served by bike lanes connecting them to the city, as bikes are much more affordable than cars.  Bikes, however, are limited in range, and do not provide as much metropolitan opportunity to jobs as carts do in traffic.  The real knot about bike lanes and the poor is that many of the poor do not trust traffic to value their lives.  Of course, I can’t find data on it, yet, but blacks are reputed to be especially loath to bike in the traffic way.  To see how real this perception is, for blacks and for planners, I’d need to see per mile risk of being hit while biking for blacks versus whites.  I’d like to find trip purpose and time data for those collisions.

* Like so many things around the emotionally loaded term “gentrification”