2 Dec
1998
2 Dec
'98
4:23 p.m.
____________________________ Forward Header ________________________________
Subject: 2000 CTPP TAZ-UP
Author: Don Burrell <DBURRELL(a)oki.org> at SMTP-GATEWAY Date: 11/25/98
5:12 PM
The Journey-to-work and FHWA folks are currently working on
a program to prepare the CTPP TAZ geography in advance of the census,
specifically by next summer. This is to be done using a program called
TAZ-UP, TIGER/Line 99 and ArcView GIS software.
Today, I shipped the results of the 2000 Census Statistical Areas Review
process for 2 of the Cincinnati area counties to the Detroit regional
office. This work, which is underway throughout the country, has identified
census tract and block group boundary changes to be used for the 2000
census.
Having just now put 2 + 2 together, my question is whether or not the Bureau
will have our new local recommendations for census boundary changes in the
TIGER/Line 99 file that is to be shipped to us for the TAZ-UP work? OKI
plans to revise many of our TAZ boundaries to conform to the 2000 census
geography. It would be helpful to have the 2000 census geography shown.
Don Burrell
OH-KY-IN Regional Council of Govmts.
____________________________________________________________________________
Response from P. Salopek
Don:
I could not answer this question on my own, so I spoke with Bob LaMacchia in
Geography Division of the Census Bureau. The first point is that for the
TAZ-UP program, TIGER/Line 98 will be the base, not T/L 99. T/L 98 will NOT
include the local recommendations for block boundaries that you just
submitted. Bob L. says that the verification phase for both programs will
occur at the same time. This means that the T/L 99 that I guess
participants in both activities will receive to review will contain both the
new TAZ boundaries and the new block boundaries (collection blocks, not
tabulation blocks). One point to remember is that in defining TAZs for 2000
you will not be constrained to build them from blocks, since blocks won't be
defined yet. You will in fact have a wider array of lines/features to use
as TAZ boundaries than has been available in the past. We think this is an
improvement and should allow closer approximation to true TAZ boundaries.
--Phil
7 Dec
7 Dec
8:35 a.m.
OKI
plans to revise many of our TAZ boundaries to
conform to the 2000 census
geography. It would be helpful to have the 2000 census geography shown.
Don Burrell
OH-KY-IN Regional Council of Govmts.
This and several other comments I've been reading from MPO staffs in the
past several weeks have left me a bit baffled. I offer the following
comments in the hope that they'll be helpful to some of the agencies on
this e-mail list that are about to review their traffic zone geography.
1) The only valid criteria for traffic zone delineation is access
patterns to the transportation network being modeled. Census geography
is a constraint, not a criteria. Assuming that you have a reasonably
detailed network to model, you'll need in some areas some pretty small
zones, geographically speaking. (If this leads to zones with
populations so small that the Census is not comfortable reporting data
for confidentiality reasons, we can work around that-we just need to
know where that level is.)
2) Historical continuity? That's what census tracts are for. And
without quoting Henry Ford, in a country where people move on average
every 5 years, history won't count for much in an area as small as a
TAZ.
3) Statistical confidence (in data from very small areas)? I assume
we've all had our Stats 101 and know what we're getting into with small
sample populations (to say nothing of non-response and mis-response).
But by going with smaller zones, your erosion of statistical confidence
is more than made up for, in my view, by more accurate coding of network
access.
Feel free to contact with any questions or comments.
Sam Granato, Cedar Rapids MPO
9515
days inactive
9520
days old
1 comments
2 participants
participants (2)
-
phillip_a_salopek＠ccmail.census.gov -
Sam Granato