New subject: ESRI article on ACS errors

28 Jul 2008

Michael:

Short answer: very well.

Long answer: 

The wording on the ACS vehicle availability question is identical to
the Census 2000 question. It's placement in the ACS is between the two
housing utility questions (plumbing/kitchen/telephone facilities, and
home heating), so these are all the "easy" housing questions before one
has to answer the "difficult" housing questions (e.g., how much is spent
on x, y, and z....)

We've done some county and regional level tabulations of our region's
auto ownership patterns, from 1960 to 2006, and the trends do make good
sense at both the regional and county level. There are some
ups-and-downs in the county-level zero vehicle shares, comparing 2000 to
2005 to 2006, but these may not be statistically significant. (We
published this data, released in I believe September 2007, for a
November 2007 regional workshop, url:
http://www.mtc.ca.gov/maps_and_data/datamart/census/)

Our regional share of zero-vehicle households has declined from 19.7%
in 1960, to 15.8% in 1970; 12.2% in 1980; 10.5% in 1990; 10.0% in 2000;
and then 9.5% in 2005, and 9.2% in 2006. This is for the nine-county San
Francisco Bay Area.

For San Francisco City, the zero-vehicle household share has decreased
from 42.1% in 1960; 39.6% in 1970; 34.6% in 1980; 30.7% in 1990; 28.6%
in 2000; and then with the ACS: 31.3% in 2005, but dropping back to
28.6% in 2006 (same as 2000).

So, we're anxiously awaiting the 2007 annual ACS data that's scheduled
for released next month and September, and then the 3-year period
estimates (2005-2007) expected this December. The challenge will be how
to analyze and report all of this data in a timely manner, all the while
trying to do our "regular jobs."....The other challenge: do we just
report the estimates WITHOUT their standard errors necessary in
understanding the year-to-year, or period-to-period difference; or do we
take the EXTRA time needed to report the estimates WITH the standard
errors (adding, or perhaps "delaying" the reporting of the results by
say 3 to 6 months?

Chuck

**************************************************************
Charles L. Purvis, AICP
Principal Transportation Planner/Analyst
Metropolitan Transportation Commission
101 Eighth Street
Oakland, CA 94607-4700
(510) 817-5755 (office)
(510) 817-7848 (fax)
cpurvis(a)mtc.ca.gov (e-mail)
www: http://www.mtc.ca.gov/
**************************************************************

...
 >> Michael.Cline(a)utsa.edu 07/28/08 7:38 AM
>>> Chuck (or others),

In your opinion, how well do you think ACS is estimating Zero Vehicle
Households?  (or conversely household vehicle ownership?)

Michael E. Cline
Research Associate
Institute for Demographic & Socioeconomic Research
The University of Texas at San Antonio
1 UTSA Circle
BB 4.06.10
San Antonio, TX 78249-0704
(210)458-6537 f(210)458-6541
michael.cline(a)utsa.edu 
http://idser.utsa.edu 

RE: [CTPP] ESRI article on ACS errors