Each redistricting dataset merges the electoral data the SWDB collected and processed over the preceding decade with the most current census data (PL94-171). The result is a census block level dataset that allows for longitudinal analysis of electoral data over time on the same unit of analysis. Electoral data consist of the Statements of Vote (SOV) and Statements of Registration (SOR) for each statewide election. These data are collected from the Registrars of Voters for each of the 58 California counties with each election.
The SWDB collects the Statement of Vote and the Statement of Registration along with various geography files from each of the 58 counties for every statewide election. The Statement of Vote is a precinct level dataset and precincts in California change frequently between elections. The goal of the SWDB is to make election data available that can be compared over time, on the same unit of analysis – a precinct, a census block or a census tract.
The alert eyes (and computer processing skills) of one of the many state
Department of Transportation folks processing the UA Census 2000 TIGER/Line
files has discovered that the 1990 Urban/Rural code on Record Type S in the
UA Census 2000 TIGER/Line files is in error. We have determined that this
affects over 1000 counties, and are now preparing a notice for our web site.
We also are discussing how to best resolve this situation. The error
overstates the extent of the 1990 urban classification in the affected
counties -- some polygons are classified as 1990 urban when they should have
been 1990 rural. The 2000 UA/UC codes and Urban/Rural codes are correct, as
is the 1990 UA code.
Until we determine a solution, the correct 1990 Urban/Rural code is located
on the Census 2000 TIGER/Line files on Record Type A. For both versions
(the Census 2000 and the UA Census 2000 TIGER/Line files), the CENID/POLYID
codes are the same, so you can use the CENID/POLYID as a match key across
these two versions.
Once we have a solution determined, we will let you know.
U. S. Census Bureau
The Census Bureau has recreated the 1990 Census Tract Outline maps and has made them
available on the web. You may access the 1990 Census Tract Outline maps directly from the ftp site at:
Once they have updated the Map Products page, they will be available from that
It is important to note that these PDF maps are re-creations of the 1990 Census
Tract/BNA outline maps and differ from the original 1990 Census Tract Outline Maps.
Similar to the 2000 Census Tract outline maps, these maps are county-based and created
for all 1990 Counties/County Equivalents in the United States. All maps utilize 1990
geography; however, display features are based on what was used for Census 2000. These
maps show the boundaries and numbers of the 1990 census tracts/BNAs as well as the named
features underlying the boundaries. They also show the boundaries, names and codes for
1990 American Indian/Alaska Native areas, counties, county subdivisions, and places. The
scale of the maps will be optimized to keep the number of map sheets for each area to a
minimum, but the scale and number of sheets will vary by the area size of the county and
the complexity of the census tracts.
File naming convention:
State - stSS_statename (e.g. st12_Florida/);
County - SSCCC_countyname (e.g. 12099_PalmBeach/ st12_Florida/);
Sheet number/File - 90TSSCCC_(sheet number).pdf (e.g. 90T12009_001.pdf).
A PDF with the 2002 TIGER/Line format can be accessed here. As previously mentioned, there
are major structural changes in this layout. We have eliminated some record types (3
and 9) and added others (B,E,T,U). We have rearranged the location of the data fields.
We have added a large number of reserved fields on some of the record types to account
for fields we expect to use in 2003 or later -- fields that will accommodate, for
example, the new MA delineations expected from OMB next year based on their new
core-based statistical areas. Fields are also set aside to accommodate information we
expect to have in files that have been processed through the MAF/TIGER Accuracy
Also, please note that we have added a permanent zero-cell identification number to each
node (TZID). The TZID is similar in concept to the TLID and appears on RT I and the new
RT T. The possible range for the TZID, with the highest value used for that range,
appears in RT R.
One of the major difficulties data users had last decade was the mixing of "current"
state and county codes with the census tract and block numbers which are uniquely
identified only by the census state and county codes. For this reason, RT 1 continues
to provide Census 2000 codes even though the distribution unit will be current county.
Thus, it is possible that to fully match all the Census 2000 block numbers in a Census
2000 county a user will need to reference multiple "current" 2002 TIGER/Line files. An
example of this will be Broomfield County, CO which was created last year (after Census
2000) from parts of four other Colorado counties.
We expect that the 2002 TIGER/Line files will contain a small number of counties with
improved road coordinates based upon work that we have been doing in the past year.
Tentatively, we expect these counties to be included (list is subject to change):
10001 Kent, DE
10003 New Castle, DE
10005 Sussex, DE
12117 Seminole, FL
24015 Cecil, MD
24039 Somerset, MD
24041 Talbot, MD
24510 Baltimore City, MD
39011 Auglaize, OH
39033 Crawford, OH
39039 Defiance, OH
39041 Delaware, OH
39051 Fulton, OH
39069 Henry, OH
39109 Miami, OH
39137 Putnam, OH
39139 Richland, OH
39161 Van Wert, OH
39171 Williams, OH
46065 Hughes, SD
If you have any questions, please let me know. We are working on the full
documentation. We expect the 2002 TIGER/Line files to be available the last calendar quarter of this year.
We would love to help. Please leave us a message and we will get right back to you: