In an E-mail message, Eric Kostello wrote:
The problem is that there is only one CFCC per TIGER line, and
precedence is given to tranportation classifications. You will
need to examine the ZIP codes on both sides of each line to see
if they are different. If they are, that is a boundary.Eric Kostello
kostello@carson.math.ucla.eduOliver Weatherbee writes:
>
> I am having trouble extracting zipcode boundaries from the tiger census
> line files. I extract the files matching the cfcc code F50 which is supposed
> to correspond to the zip code boundaries but I get a total of 0 lines read
> out. I don't think its a scripting/software problem because it works for
> other cfcc codes (F10, F73, etc.).
>
> Oliver Weatherbee
> oliver@earthview.cms.udel.edu
I've come across a similar experience recently in Illinois.
I tried extracting zip code boundaries for Cook county, Illinois (Chicago)
by CFCC F50 but came up with nothing. I then tried extracting zip code
boundaries through comparison of left and right code numbers ( a la extracting
census tracts) but got only a few boundary lines and a bunch of discontinuous
line segments. I used the following awk script to do this:
awk 'substr($0,107,5)!=substr($0,112,5)'
At this point I think the problem may be that zip code boundary info is only
available where address ranges occur. Although I thought this was only a
problem in rural areas, it may also occur in urban areas as well. I've been
using 1990 Tiger/Line Files but I know the 1992 files were supposed to have
a lot of added information. I put a call into the Census Bureau today and
expect to hear back on Monday. Will keep you posted.
--
-------------------------
Paul Loechl p-loechl@cecer.army.mil
USA CERL (217)352-6511x668