shield sodapop

PUMS 1990 and 2000 Samples


The Population Research Institute (PRI) is currently evaluating the SODAPOP service. The ability to extract a data set containing a subset of variables will not be supported after September 15, 2016.

The data collections in SODAPOP will still be available from the PRI. If you should have concerns about this change to the SODAPOP service or need to access some of the data, contact the PRI's Computational and Spatial Analysis (CSA) core at

STIPULATIONS FOR USE: Access Restricted to Penn State researchers and members of the Association of Population Centers (APC). The files prepared below were obtained from ICPSR. ICPSR does not allow redistribution to third party users beyond Penn State or the APC (permission to redistribute within the APC granted by ICPSR).

DESCRIPTION 1990: The 1990 Public Use Microdata Sample (PUMS) contains household and person records for a sample of housing units that received the "long form" of the 1990 Census questionnaire. Data items include the full range of population and housing information collected in the 1990 Census, including 500 occupation categories, age by single years up to 90, and wages in dollars up to $140,000. Each person identified in the sample has an associated household record, containing information on household characteristics such as type of household and family income.

The Public Use Microdata Samples from the 1980 Census contain individual and household level information form the "long-form" questionnaires distributed to a sample of the population enumerated in the Census. Each of the PUMS files contains two types of records: "household" records and "person" records.

DESCRIPTION 2000: PUMS files have state-level Census 2000 data containing individual records of the characteristics for a 1 percent sample and 5 percent sample of people and housing units. The PUMS files contain geographic units called super-Public Use Microdata Areas (super-PUMAs), a new geographic entity for Census 2000. The state files, which may contain one or more super-PUMAs, include geographic equivalency files that show the relationship between the super-PUMA and standard Census 2000 geographic concepts (e.g., counties, etc.). The super-PUMAs are made up of a Public Use Microdata Area (PUMA) or group of contiguous PUMAs (each PUMA must have a minimum of 100,000 population). PUMAs are only identified on the 5-percent files and not on the 1-percent files.

Download PUMS 1990 and 2000 File

R Users -- Code for downloading and working with the PUMS in R is available on the Analyze Survey Data for Free website 


Census Bureau PUMS Web Page
Integrated Public Use Microdata Series Web Site
United States Census Bureau Census 2000 Gateway
American Community Survey (ACS) PUMS 2000-2002 Page

PUMS files on SodaPop

The PUMS data has been reshaped from its original hierarchical format into a rectangular person-level file with household data attached to each person.

All empty households have been deleted.

In order to uniquely identify a person in the PUMS, PRI has created a variable ID which is a combination of the state FIP code, SERIALNO and a sequential person number within the household for each person record.  

Using the full US 5pct Sample:

Do not attempt to access the full US 5pct Sample table in an interactive session. It is simply too large. A good strategy is to go through a trial extract on a single state -- selecting variables and establishing selection criteria. When you are satisfied with your code, save it and change the SAS table name to point to the US file --

This is the 1993 re-release version of the PUMS data and was downloaded on 2/12/98 to PRI. It includes some changes from the earlier version involving the Texas file and the variables POWSTATE, MIGPUMA, PSA, MIGSTATE, POWPUMA.