Status:Closed    Asked:Apr 06, 2018 - 08:16 AM

Match IPUMS-CPS with (I)PUMS USA census data 1970-2010 at PUMA/county level

Dear all,

I am looking for a way to match the CPS files 1994 - 2010 to the census files at the PUMA / county level to (collapsed at the PUMA/county level) census data (starting 1970).

As I far as I understand, there have been changes in the PUMA definition over time.

Thus, my questions are:

1) Is there a mapping from historical to current PUMA definitions?

2) Is there a mapping from IPUMS to CPS PUMAs or is this trivial and uses the same unique identifier?

Looking forward to your response.



Do you have the same question? Follow this Question

Staff Answer

You are correct in your understanding that PUMA boundaries have changed over time (you can see the comparability here). IPUMS USA PUMA boundary information and crosswalks can be found here. However, there may be a larger concern with your attempt to match IPUMS USA to IPUMS CPS, which is that PUMAs are not identified in the CPS. These are the available geographic variables in IPUMS CPS.

Here are a few alternatives that may be helpful: (1) you could try to merge on county or at the metropolitan area, however, not all counties/metropolitan areas are defined in IPUMS USA or IPUMS CPS or (2) you could explore IPUMS NHGIS, which provides summary level statistics for a variety of geographic levels. Which will be appropriate will depend on what you are trying to accomplish.


Apr 06, 2018 - 03:00 PM

Report it


Match 1860-1920 census data to county subdivisions in 2010 TIGER shapefile
Which is the lowest level of geography in the ACS PUMS 5 year samples: PUMA o...
2010 5-year PUMA definition
crosswalk county FIPS to PUMA 1% for 2000 and 2010 based PUMA
Login   |   Register

Recently Active Members

View More »

Share |