This e-mail may be of interest to the State DOTs and MPOs who have requested an ASCII
version of the CTPP Part 1 CD data from the Census Bureau, and who are using (or will be
using) SAS to analyze CTPP datasets.
The Census Bureau did a great job in providing SAS jobs for building master SAS databases.
This is *extremely* helpful!
I did some minor modifications to the programs, that I've included in a zip archive of
CTPP Part 1 SAS programs. I also have some jobs which extract my region's data, then
exports various tables into "csv" files of less than 256 columns, for purposes
of importing into spreadsheets and GIS. Essentially this is a "SAS chop shop"
approach that is an alternative to using CAT (CTPP Access Tool).
(A key addition to the SAS code that saves immensely on the SAS file size is the use of
the variable "default length=5" statement which limits variables to less than
one-half billion in value.) These SAS files are on our web site, at:
(This is using our http site, rather than our ftp site, so it should be available to some
of my colleagues who've had firewall problems accessing our ftp site.)
The SAS jobs also have some "label statements" that may be extremely useful to
non-SAS programmers, e.g.:
tab15x1 ="tab15: OCCUPATION=1, INDUSTRY=1"
tab15x2 ="tab15: OCCUPATION=1, INDUSTRY=2"
tab15x3 ="tab15: OCCUPATION=1, INDUSTRY=3"
tab15x4 ="tab15: OCCUPATION=1, INDUSTRY=4"
tab15x5 ="tab15: OCCUPATION=1, INDUSTRY=5"
tab15x6 ="tab15: OCCUPATION=1, INDUSTRY=6"
tab15x7 ="tab15: OCCUPATION=1, INDUSTRY=7"
The real effort will be to create mnemonic variable names for variables such as
"tab15x1" (perhaps "occ1ind1" ?)
Hope this helps.
Chuck Purvis, MTC