Geographic granularity for birthplace

Hi folks, I’m trying to assess birthplace at the lowest geographical level of granularity. It seems that this is PUMAs, but I’d like someone with a bit more experience that this is, indeed, the case, and that some other survey tool run by the BLS/Census bureau/etc. doesn’t seem to have it available on any other lower granularity (e.g., block — for some reason I assumed census block may have this!).

So, is it actually the case that PUMAs, rather than ZTAs or other geographies, are the smallest parcels of land wherein we can get respondent birthplace data from?


In the public use micro data, available via IPUMS USA, you are correct that the lowest level of geographic identification is the PUMA in modern samples. This limitation is primarily driven by respondent confidentiality restrictions placed on public use micro data. One alternative data source is IPUMS NHGIS. This IPUMS project makes previously aggregated tables of data available and enables the identification of basic statistics down to a very low geographic level (e.g., census blocks). From the sounds of your research needs, it seems like you should be able to find the information you are looking for via IPUMS NHGIS.

1 Like

Thanks Jeff, a bit (or rather, very) late on my part, but this is super helpful!