Recently there have been a number of questions about when to delineate TAZs for the NEXT CTPP.  We are working with the Census Bureau for TAZ submission for the next CTPP.  They have not yet provided us with a cost estimate for the software.   It will be a GIS-based approach, as the TAZ-UP process was in CTPP2000. We had a meeting  on July 25, and the schedule for TAZ submission into TIGER is most likely going to begin January/Feb 2009.   In the last CTPP Status Report  http://www.fhwa.dot.gov/ctpp/sr0307.htm , we had slightly earlier dates (I think we said Fall of 2008).  By starting a few months later, preliminary (not final) 2010 Census tract and block group boundaries will be available as references for TAZ definition.

Many regional models are moving toward smaller and smaller TAZs.  In one case, we heard that TAZs were planned to have an average of 120 households.  Because the ACS sample size is so much smaller than the decennial census "long form" this might result in only 6-10 completed surveys for housing units, even after 5 years of sampling.  We are concerned that the CB might require TAZ to have the same population thresholds as "block groups".  The CB issued a Fed Reg notice about raising the population threshold for block groups from 600 to 1200.  At today's meeting, the CB said they had gotten a lot of comments that people preferred the CB to keep the threshold at 600.    TAZs that are submitted into TIGER for CTPP might need to aggregated your model zones,  for example,  two of your model TAZ = 1 Census TAZ, otherwise you might risk a lot of data suppression based on disclosure avoidance rules by the Census Bureau's Disclosure Review Board (DRB).  We have a meeting set up for August 23 or 24 with the DRB which we hope will provide more insight on how they will consider the next CTPP.  We are planning to discuss synthetic data approaches which may result in the ability to provide small TAZ summaries.   The Census Bureau is concerned that releasing data for very small geographic units results in data with very large margins of error.  They would prefer that larger geographic zones be tabulated to reduce the margin of error. 

Finally, just as a reminder—small area tabulation (like tracts and TAZ) from ACS is restricted to 5-year accumulation of sample records. 

Elaine Murakami

FHWA Office of Planning

206-220-4460