|
Status and Holdings
Summary File 3 was released during August and September of 2002. See the Bureau's Summary File 3 page
for a summary with related links. The MCDC has complete SF3 data (the "A" files) for the states of Missouri, Illinois, Kansas,
Minnesota, Michigan and Delaware. We also have the National ("B") file which has summaries for areas as small as ZCTA (ZIP code), MCD and place (of any size). (See the Summary Level Sequence Chart, p.3, for a detailed list of the geographic units available on these files.)
Finally, in the spring of 2004 we decided that we wanted our CAPS
(Circular Area Profiling System) web application to work with very small geography for anywhere in the US. So
we obtained and converted a complete set of Summary Level 090 data for every state in the U.S. These are the smallest
geographic entities on SF3. We store all these in a special subdirectory, xxsl090. Since these are
sub-census tract level summaries we have only P and H tables for them.
Accessing Summary File 3 Data
There are a number of ways to approach access to this rather daunting collection of data. Some are a lot easier than others, and some provide a lot more power and control but have learning curves associated with them. Much of this document will deal with one of the more challenging ways to access it via uexplore. But here are some other, generally much easier, ways to approach SF3. Unless you are just a very casual user of census data all of these approaches are worth checking out.
- The MCDC's SF3 Population Profiles web application give you quick, menu-driven access to standard profile reports ("dp3_2k" reports) extracted from Summary File 3. These reports contain links to more detailed tabular displays.
- The MCDC was proud to be a key participant in the national State Data Centers' cooperative project to create software for generating what we think are the best demographic profiles available anywhere based on SF3 data. Be sure to check out MCDC's Detailed Profiles collection.
- Use the American Fact Finder application at the Census Bureau. There are
many ways to access these data there. A good way to start is by clicking on the 2000 Summary File 3
link in the
Data Sets section. AFF provides access to detailed tables, "quick tables" extracts, data maps, metadata with detailed information about the tables, etc. There is an online tutorial to help you get started with this software.
- Use the uexplore utility to access SF3 datasets in the MCDC's sf32000 or (recommended as the better choice for the novice user)
sf32000x data directories. For a more detailed discussion of using uexplore to access the full sf32000 datasets, see below. For more detail on exploring the sf32000x extract sets see the Readme.html file in that subdirectory.
Technical Documentation
The Census Bureau has provided extensive technical documentation in the form of a 1249-page pdf file. The complete document can be accessed at the Bureau's web site, or you can access the MCDC version where it has been partitioned so that each chapter is its own separate pdf file, with an index page for easier access.
For persons using the uexplore application to access the datasets as stored in the MCDC's data archive, the best codebook files to use are in the Varlabs subdirectory (see below.)
Related Filetypes
sf32000x: Standard Extract
This is the MCDC's standard extraction of just over 200 key data items from the 16,000+ cells of tabular data available in the full SF3 summary records. Many users of Summary File 3 data will find that less is more, and that accessing the extract is easier and more rewarding than dealing with the full set of detailed tables. See the Readme.html file for a full explanation.
sf3prof: Census Bureau's Demographic Profiles (DP-1 thru DP-4)
In May of 2002, several months prior to the release of the full SF3 files, the Census Bureau released these Demographic Profile products. Very much in the same spirit as the MCDC's sf32000x standard extract and the corresponding dp3_2k profile reports, these are extracts of the complete set of tables contained on the SF3 files. They are available only for governmental units. So, no data for census tracts, block groups or
ZIP codes; just for the country (i.e. U.S. totals), states, counties, and places (cities). Avaiable for the entire country as machine-readable database files or as nicely formatted reports in pdf format.
The MCDC has obtained the complete set of data files for these profiles and has created a separate sf3prof filetype which you can access via our uexplore utility using
the URL http://mcdc2.missouri.edu/cgi-bin/uexplore?/pub/data/sf3prof.
We have also created a custom set of menus to facilitate easy access to the pdf files for the entire United States.
Purists will note that this product has 4 distinct parts - DP-1, DP-2, DP-3 and DP-4 - and that one of those parts, DP-1, is really not related to SF3. It is related to the complete count tabulation Summary File 1. It was issued as a separate product in 2001 but then rereleased as a companion to the other 3 DP-x products in 2002.
Accessing the sf32000 Data Via Uexplore
Do NOT even consider using uexplore as a way to create your own complete SF3 data library. It will not work! Use it to extract specific tables for specific geographic areas. You will not be able to convert entire datasets because they are simply too large for this application.
Before beginning to explore the data here you should consider two possible alternative approaches:
- Do I really need the tabular detail of the full sf32000 datasets, or would I do better to explore either the sf32000x standard extract datasets? (Or, for certain geographic levels, possibly the sf3prof filetype.)
- For access to complete tables the Census Bureau's American Fact Finder application is much more user friendly and provides better access to metadata (especially for first-time users). Once in AFF, you'll want to choose the "2000 Summary File 3" option within the "Data Sets" section (just below "Basic Facts"). Note that once you have selected your geography and tables you'll need to use the "Print/Download" option on the menu bar in order to retrieve an extract.
AFF is probably the easiest way to search for specific tables by subject matter. Once you have identified the tables you are interested in, you may find that uexplore is somewhat faster or delivers the data in a more useful format than AFF. To use uexplore effectively against these datasets you will need to know about our dataset naming conventions, our table variable naming conventions, and about filtering using SumLev and (when extracting data for the state or counties) Geocomp variables.
The fastest-loading and simplest "codebook" files to assist you in using these datasets are the 4 modules in the Varlabs subdirectory . There are 4 modules - for P, PCT, H and HCT tables. Here is an example from the hlabels.txt dataset:
/* H10. HISPANIC OR LATINO HOUSEHOLDER BY RACE OF HOUSEHOLDER [17] */
/* Universe: Occupied housing units */
H10i1='Total:' /* H010001 */
H10i2=' Not Hispanic or Latino householder:' /* H010002 */
H10i3=' Householder who is White alone' /* H010003 */
H10i4=' Householder who is Black or African American alone' /* H010004 */
H10i5=' Householder who is American Indian and Alaska Native alone' /* H010005 */
H10i6=' Householder who is Asian alone' /* H010006 */
H10i7=' Householder who is Native Hawaiian and Other Pacific Islander alone' /* H010007 */
H10i8=' Householder who is Some other race alone' /* H010008 */
H10i9=' Householder who is Two or more races' /* H010009 */
H10i10=' Hispanic or Latino householder:' /* H010010 */
H10i11=' Householder who is White alone' /* H010011 */
H10i12=' Householder who is Black or African American alone' /* H010012 */
H10i13=' Householder who is American Indian and Alaska Native alone' /* H010013 */
H10i14=' Householder who is Asian alone' /* H010014 */
H10i15=' Householder who is Native Hawaiian and Other Pacific Islander alone' /* H010015 */
H10i16=' Householder who is Some other race alone' /* H010016 */
H10i17=' Householder who is Two or more races' /* H010017 */
This documents table H10 and shows which variables contain exactly what information. It tells you, for example, that the variable containing the count of non-hispanic African American (alone) households is h10i4. Note the variable naming convention used by MCDC -- the values to the left of the "=" signs in the label statements. The data dictionary name used by the Census Bureau is shown at the far right, enclosed in "/*" and "*/" ; so what we call h10i4, they refer to as h010004. The latter are the names you will see if you search the Bureau data dictionary (534-page pdf file).
Click here to go to uexplore the sf32000 directory. (Don't bother clicking on the Readme.html file - you are already looking at it.) Or, save yourself a step and go directly to the Datasets.html page in that directory. It is by far the best way to navigate the very complex waters of Summary File 3.
|
