Description

PUMA identifies the Public Use Microdata Area (PUMA) where the housing unit was located. PUMAs are the smallest geographic units identified in Public Use Microdata Samples since 1990 for both the decennial censuses and the ACS/PRCS. To help maintain the confidentiality of respondents in public use microdata, every PUMA must, by design, have at least 100,000 residents at the time of the corresponding census.

IPUMS has also created PUMAs for the 1960 5% sample, which each correspond approximately to one or more 2000 PUMAs using a minimum population threshold of 50,000.

Note: PUMA codes are state-dependent. The PUMA codes for 1990 and later samples are unique only within each state, so a single code may identify different PUMAs in different states. To uniquely identify all PUMAs, it is necessary to combine the PUMA variable with a state identifier (STATEFIP or STATEICP). The IPUMS-defined 1960 PUMA codes are unique across all states and do not have this requirement.

The Census Bureau updates PUMA definitions with each decennial census, so PUMA codes may or may not identify consistent areas over longer periods of time. See the Comparability section for information about which PUMAs are used in each sample. See the Codes section for information about each set of PUMA definitions.

Note: In some multi-year samples, PUMA definitions vary or are allocated In multi-year ACS/PRCS samples that span a PUMA definition change, the identified PUMAs vary based on the survey year when each respondent was interviewed (as given by MULTYEAR). This occurs in samples that span the change from 2000 PUMAs (survey years 2000-2011) to 2010 PUMAs (survey years 2012-2021) and in the 2022 5-year ACS/PRCS sample (which uses 2010 PUMAs for survey years 2018-2021 and 2020 PUMAs in 2022). With the 2023 5-year ACS/PRCS sample, the Census Bureau began reporting only 2020 PUMAs for all survey years, which they achieved by assigning pre-2022 respondents a 2020 PUMA based on their 2010 PUMAs of residence and the known relationships between 2010 and 2020 PUMAs. The Bureau has provided no details on their PUMA allocation method, but IPUMS has confirmed that it produces plausible population totals for the assigned 2020 PUMAs.

In 1980 samples, "county groups" (CNTYGP98) are functionally similar to PUMAs with the same minimum population threshold of 100,000. In fact, "county group" is a misnomer because 1980 county groups, like PUMAs, typically subdivide large-population counties into smaller areas (though 1980 county groups still do not subdivide cities within counties as PUMAs do). In contrast, the county groups in 1970 samples (CNTYGP97) nearly all correspond to groups of one or more counties with a minimum population threshold of 250,000. In 1940 and 1950 samples, the identified geographic areas most like PUMAs are State Economic Areas (SEA).

PUMA is a 5-digit numeric variable identifying the Public Use Microdata Area (PUMA) where the housing unit was located.

In all samples but 1960, PUMA codes are state-dependent. To uniquely identify all PUMAs in 1990 or later samples, the codes must be read in combination with one of the STATE variables: STATEFIP or STATEICP.

As explained in the Description and Comparability sections, PUMA definitions are updated every ten years and therefore vary across samples and within some multi-year samples. IPUMS provides maps, composition files, and boundary files describing each set of PUMA definitions through the links below.

PUMA Resource Pages
2020 PUMAs: 2022-2031 ACS/PRCS
2010 PUMAs: 2010 Census and 2012-2021 ACS/PRCS
2000 PUMAs: 2000 Census and 2005-2011 ACS/PRCS
1990 PUMAs: 1990 5% State and 1% Metro samples
1990 PUMAs crossing state lines: 1% Metro sample
1960 PUMAs and Mini-PUMAs: 1960 5% sample

Comparability

This variable is not comparable across all years, but it is comparable for any set of samples that use the same PUMA definitions, as summarized below.
In coordination with State Data Centers, the Census Bureau updates PUMA definitions every ten years to maintain PUMA populations within an accepted range (at least 100,000 and generally less than 200,000), to satisfy the evolving goals of each state's stakeholders, and occasionally to implement new standards. Samples from decennial censuses always use PUMA definitions corresponding with that census. (E.g., the 2010 census 10% sample uses 2010 PUMAs.) In contrast, ACS/PRCS samples typically incorporate new PUMAs two years after a decennial census, resulting in these correspondences between ACS/PRCS samples and PUMA definitions:

• 2022-forward ACS/PRCS: 2020 PUMAs

• 2012-2021 ACS/PRCS: 2010 PUMAs

• 2005-2011 ACS/PRCS: 2000 PUMAs*

*The PUMA definitions are the same for the 2000 5% sample and all the 2005-2011 ACS/PRCS samples with one notable exception in Louisiana: due to population displacement following Hurricane Katrina, three PUMAs (01801, 01802, and 01905) were combined into code 77777 for the 2006-2011 1-year samples and for all cases in the 2005-2007 3-year file and 2005-2009 5-year file. The original 2000 PUMAs in this area no longer had sufficient population to be identified separately.

For 1990, there are distinct PUMA definitions for different types of samples. In the 1990 5% State sample, as in later samples, none of the PUMAs cross state lines. For the 1990 1% Metro sample, PUMAs generally follow the boundaries of central cities, Metropolitan Statistical Areas, Primary Metropolitan Statistical Areas, or non-metropolitan places (see METAREA for definitions of these terms), and they may cross state lines to maintain correspondence with metropolitan areas. Where a PUMA crosses state lines, the state is not identifiable for households in the PUMA, and STATEFIP and STATEICP provide a code of 99.

For greater comparability, IPUMS USA provides two other variables that identify "ConsPUMAs": minimally aggregated sets of PUMAs (or 1980 county groups) with consistent extents across time. CONSPUMA identifies ConsPUMAs for 1980-2011 samples, and CPUMA0010 identifies ConsPUMAs for 2000-2021 samples.

IPUMS designed the 1960 PUMAs to each correspond approximately to a set of one or more 2000 PUMAs. The 1960 PUMA resource page provides a relationship file identifying which 2000 PUMAs correspond to each 1960 PUMA.

Universe

Availability

• 2023: All samples
• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: All samples
• 2010: All samples
• 2009: All samples
• 2008: All samples
• 2007: All samples
• 2006: All samples
• 2005: All samples
• 2004: --
• 2003: --
• 2002: --
• 2001: --
• 2000: 5%; 1% unwt
• 1990: 5% state; 1% metro; 1% unwt; 3% elderly
• 1980: --
• 1970: --
• 1960: 5%
• 1950: --
• 1940: --
• 1930: --
• 1920: --
• 1910: --
• 1900: --
• 1880: --
• 1870: --
• 1860: --
• 1850: --

• 2023: All samples
• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: All samples
• 2010: All samples
• 2009: All samples
• 2008: All samples
• 2007: All samples
• 2006: All samples
• 2005: All samples
• 2000: PR 5%
• 1990: All samples
• 1980: --
• 1970: --
• 1930: --
• 1920: --
• 1910: --

Flags

Editing Procedure

There is no editing procedure available for this variable.