Redistricting Files P.L. 94-171
The redistricting files or Public Law 94-171 files were created by the Census Bureau to give states the information they need to redraw legislative boundaries. They also provide data users with a first look at Census 2000 results. The information on the files includes counts of the total population and counts by race, Hispanic origin, and the same counts for the voting age population (18+). The counts are repeated for different levels of geography. For instance, the total population and one measure of racial diversity for selected geographies in Michigan are as follows:
| Summary level | Total Population |
Number identifying
with 2+ races |
|---|---|---|
| State [040] | ||
| Michigan | 9,938,444 | 192,416 |
| County [050] | ||
| Washtenaw | 322,895 | 8,293 |
| Wayne | 2,061,162 | 51,269 |
| Wexford | 30,484 | 333 |
| County-County Subdivision [060] | ||
| Ann Arbor City, Washtenaw County | 114,024 | 3,480 |
| Ann Arbor township, Washtenaw County | 4,720 | 89 |
| Augusta township, Washtenaw county | 4,813 | 64 |
| Bridgewater township, Washtenaw county | 1,646 | 9 |
| Dexter township, Washtenaw county | 5,248 | 56 |
| County-Place [160] | ||
| Detroit city | 951,270 | 22,041 |
| Dexter village | 2,338 | 33 |
| Dimondale village | 1,342 | 16 |
| State-County-Census tract [140] | ||
| Washtenaw county, tract 4038 | 2,884 | 167 |
| Washtenaw county, tract 4053 | 4,615 | 46 |
| Washtenaw county, tract 4105 | 2,515 | 214 |
The original release of these data by the Census Bureau was in three parts (geography, Tables 1 and 2, and Tables 3 and 4). We have merged the three parts into one record per level of geography and written the files using a fixed format. This works best for those who are used to using a statistical package. For those who prefer to use spreadsheets, we suggest the Census Bureau's site.
The redistricting files are quite large for two reasons. The first is that the information is repeated over and over for each distinct unit of geography (as shown in Table 1 above). For instance, there will be a record for each county in a state, for each place within a county, for each census tract within a county, for each block group within a tract, etc. A relatively unpopulous state can have many records if it is a state with many units of geography. The attached table shows the sizes of these files by state.
The second reason that these files are large and so much larger than they were in 1990 and 1980 is due to the significant changes in the measurement of race. In the past, respondents could only identify with a single race and the number of races reported in these tables was limited to White, Black, Asian, Native American and Aleut, and some other race. For the 2000 census, respondents could identify with multiple races and the number of possible races has increased to six categories: White, Black, Asian, American Indian and Alaska Native, Native Hawaiian and other Pacific Islander, and some other race. Allowing for all possible multiple combinations of the six races results in 63 categories. Because the race tables are also tabulated by Hispanic/not Hispanic, the number of categories grows to 126.
The fixed format files for each state, D.C. and Puerto Rico are available from our anonymous ftp site . Also available for downloading is a data dictionary and code for reading the data into SAS.
Because these files are so large, users may run into problems if they read all records and all variables. The best way to use these data is to keep only the records that are appropriate for the geography you are interested in. For instance, if one was interested in counties for the state of Michigan, the following example shows how to read a record in, determine if it is for a county (keep) or not (delete). It also may help if one does not try to make a permanent data set until one has reduced the size of the file (note the use of data_null in the sample code).
For users who would prefer to deal with a reduced number of races and somewhat smaller files, check out the extracts we have created that provide several reasonable choices for summarizing the 126 race categories. For further discussion of race coding in 2000 go to Coding of Race in 2000 in our subject area. Even though these are smaller files, one might still have problems with their size unless one employs the same strategies described above to minimize the need for SAS temp space and to read the data effciently.
PSC blog
Recent resources, events, news
New Publications
Bingenheimer & Geronimus, "Behavior & HIV"
Wildeman, "Imprisonment & Infant Mortality," PSC Research Report
Next Brown Bag
Mon, Nov 9
John Bound
Stratification in US Higher Education
For live stream
LINK HERE
W A R N I N G
If you are reading this, it may be that you are using rather old web browsing software that does not support modern international Web technology standards. For a better experience of the Web and this site in particular, please upgrade your web browser software today. The following are good choices: Firefox 2; Opera 9; Safari 3.