PUMAs are redefined every ten years in conjunction with the decennial census. This document describes the PUMAs as they were defined
for use with the 2000 census. For 2010 we shall get a new set of PUMA definitions. We do not anticipate that these new entities will differ radically from their predecessors but there are going to be some notable changes. The differences will be not just in the specific
geographic boundaries, but in the guidelines used to create them. Two of the more important changes are:
We first saw the new 2010 PUMA's (aka "2012 PUMAs" especially in connection with the American Community Survey where data for these entities
was first reported for vintage year 2012) in the spring-summer of 2012. They will be used in the 2010 census Public Use MicroSample (PUMS) files
(if they ever get generated - at the moment it appears that no such product will be created) as well as the ACS PUMS files (starting with vintage 2012).
They were first used in ACS products starting with vintage 2012 (released in calendar year 2013). See our alternate page dealing with the
2010 PUMAs, referenced at the top of this page.
- Incorporated city boundaries may no longer be used to define PUMAs -- only continguous counties and/or census tracts may be used.
- The local agencies that define the PUMAs will be asked to assign mnemonic names to the PUMAs.
The Census Bureau has created a (2010) Public Use Microdata Areas (PUMAs) web page. This page summarizes the (proposed, for now) criteria to be used in defining these entities and contains links to pages with more detail. A Reference Information section provides general links to various related sites, including one to a PUMA tutorial module .
Some Basic Facts About PUMAs
Using Maps to See Which PUMAs Are Where
You can access the collection of pdf file base maps accessible from the Bureau's web site at http://www.census.gov/geo/www/maps/puma5pct.htm. From this index page choose your state. The key to using these maps is to remember that the 5% PUMAs nest within the Super-PUMAs and these pdf files are organized as follows:
Note that these maps are designed to be printed on 8.5 x 11" paper, in landscape mode. A color printer is best, since the boundaries are color-coded.
- The first page of the pdf file is the state overview map showing the Super-PUMAs.
- Depending on the state, there will be none, one or more inset maps showing Super-PUMAs for metropolitan areas of the state.
- Following the inset map page(s) are a series of pages, one per Super-PUMA, showing the 5% PUMAs within that super-PUMA. The maps also show relevant place and county boundaries to help you see what geographic areas correspond to the PUMAs.
Go to the 3rd page of the Colorado map file. This map shows the 5% PUMA areas within the 1st Super-PUMA for Colorado. You can see from this page that PUMA 00101 for Colorado is comprised of a series of rural counties in the extreme northwestern corner of the state and going across the nor therm border over to Larimer county. We see that the PUMA is made up of 5 complete counties (Moffat, Rio Blanco, Garfield, Routt and Jackson) as well as parts of Mesa and Larimer. It may take a little time staring at the map to actually see this, and some people seem to never see it. But for most people and most PUMAs the maps are a pretty good tool for seeing where a PUMA is located. But sometimes, it can get more than a bit tricky trying to decipher the county and PUMA boundaries on these maps to make sure you know exactly what areas a PUMA covers. That is where the the tool we'll talk about next comes in.
Using MABLE/Geocorr to Relate PUMAs to Other Geographic Codes
If you not familiar with this geographic utility application we suggest you start by looking at the explanation and example from the MCDC Quick Tour page . There is also a powerpoint tutorial (linked to from the MABLE/Geocorr2k page) that should help you get started.
What geocorr does is create reports (and/or comma-delimited files) showing how different geographic layers correspond to one another. A good example relevant to the current topic involves using the application to generate a report showing how PUMAs relate to counties in the state of Colorado. To do this, invoke the application and fill out the form as follows:
- Choose Colorado as the state to process.
- From the "Select 1 or more "SOURCE" Geocode(s)" select list choose PUMA for 5 Pct Samples (2000) .
- From the "Select 1 or more "TARGET" Geocode(s)" select list choose County (2000) .
- Skip down a short ways to the Output Options section, and check the box labeled "Generate 2nd allocation factor (AFACT2): portion of target geocodes in source geocodes". This means that our report will not only show us what portion of the PUMA population resided in the county in 2000, but also what portion of the county population resided within the PUMA.
- Accept defaults (ignore) the rest of the options. Find the first "Run Request" button you can and click it to invoke the geocorr program.
It should take about 5-10 seconds to process this request. In your browser a page will be generated summarizing the results and providing hyperlinks to the 2 Output Files. If you click on the Listing (report format) link you should see a report, the first few lines of which should look like this:
puma5 to county to
Total Pop, county puma5
2000 alloc alloc
puma5 County cntyname census factor factor
00101 08045 Garfield CO 43791 0.438 1.000
08057 Jackson CO 1577 0.016 1.000
08069 Larimer CO 12749 0.127 0.051
08077 Mesa CO 3112 0.031 0.027
08081 Moffat CO 13184 0.132 1.000
08103 Rio Blanco CO 5986 0.060 1.000
08107 Routt CO 19690 0.197 1.000
These are just the lines of the report dealing with the first value of the "source" geocode, which happens to be the Colorado PUMA we were just looking at on the pdf map page. Each line of the report represents the intersection of this geographic area with a "target" geocode -- a county. The first line of the report tells us that the intersection of the PUMA with Garfield county had 43,791 persons living in it according to the 2000 census. The first "alloc factor" column has a value of 0.438, which is telling is what portion of the PUMA's total population is represented by this intersection. So almost 44% of this PUMA's population is in Garfield county. The last column, titled "county to puma5 alloc factor" is the allocation factor going the other way, the one we only get because we checked that special option ("Generate 2nd allocation factor (AFACT2)") on the input form. The value of 1.000 tells us that the entire county of Garfield is (was) contained in this PUMA. In fact, as you scan the 7 lines of the report you see there are 5 counties listed that have a value of 1.000 in this column, indicating that they fall entirely within the PUMA. These are the 5 counties we already mentioned above, based on our perusal of the map. The two other counties that we noted were only partially within the PUMA (Larimer and Mesa) also are listed. We learn from the report that only about 5% (.051) of the population of Larimer county is in this PUMA, while less than 3% (.027) of Mesa county's population is in PUMA 00101.
This example shows us how to relate the PUMA areas to counties. We can just as easily (by changing our selection in the Target Geocodes select list) get comparable reports for other geographic levels such as places (cities), congressional districts, CBSA's (metropolitan and micropolitan statistical areas), Urbanized Areas/Urban Clusters, etc.
Special Report Shows What Counties and What Cities Comprise PUMAs
The MCDC has created a custom report (in pdf format) showing all the 5% PUMAs in the United States (an alternative report for just Missouri is also available). The report is sorted by State, Super PUMA and 5% PUMA. It shows the 2000 census total population for each PUMA and two geographic lists:
- A list of all the counties (or equivalents) that are contained (all or partly) within the PUMA. When the county is only partly contained within the PUMA a percentage figure is displayed (in parentheses following the county name).
- A list of place (city) names showing which of these areas fall within the PUMA. Only places of 2500 or more population (per the 2000 census) appear in these lists. An "Other" category indicates that the PUMA includes territory that was not within a place. Once again, a percentage is displayed to indicate what portion of the place's population (as of the 2000 census) was included in the PUMA. If a place has no percentage shown, it indicates that the entire place is contained within the PUMA.
Summary Data at the PUMA Level
Data at this level are being published annually as part of the American Community Survey. Tables can be accessed via American FactFinder.
The Census Bureau did not publish any summary data for these units based on the 2000 census. You will not be able to find a PUMA summary level on either sf1 or sf3 from the 2000 census. However, it is possible to aggregate data at the split block group (090 summary level code) to create such summaries. The Missouri Census Data Center has done this for 5% PUMAs for the entire country for sf3 data from the 2000 Census.
These data are available in the data archive as well as in the format of our standard 2000 census demographic profile reports (dp3_2k series). To access these go to the main menu page at http://census.missouri.edu/census2000/ and choose PUMAs. Filter by a state and then to a PUMA within that state. You can access the data set with data for all 5% PUMAs for the entire U.S. via Dexter using the link that appears at the bottom-left of any dp3_2k profile report page (Extract Data Via Dexter).