Hi: This might be a straightforward question, but I’m curious about what options are out there. I’m looking for a PUMA-to-sub-county division crosswalk with an allocation factor. I’ve already downloaded one from Geocorr, but I’m wondering if there are any alternative sources or methods that others recommend. Thanks! - Greg
If all you’re looking for is a crosswalk between PUMAs and county subdivisions (which is what I assume you mean by “subcounty divisions”), then Geocorr is a great source, and I don’t know of another.
But I suspect what you may really be looking for is guidance on “small area estimation”, i.e., is there a good way to associate microdata with county subdivisions and/or generate estimates for county subdivisions from microdata. That’s a more complicated question. You can certainly use the Geocorr allocation factors for this, but that approach can be highly inaccurate. A simple allocation model assumes that the characteristics of a county subdivision within a PUMA are the same as the whole PUMA’s, e.g., that the county subdivision’s population has the same racial composition, income distribution, etc., as the PUMA as a whole. But a single county subdivision may contain only a small fraction of a PUMA’s entire population, or it may contain parts of multiple PUMA’s populations, and the specific population characteristics within such a county subdivision may differ significantly from those of the entire containing PUMA(s).
First, I’d say that there’s no simple way to produce consistently reliable estimates in such settings. That said, here are some examples of past research that have estimated tract-level characteristics from PUMA-level characteristics, and you might look into ways to extend this approach to county subdivisions:
- Nagle, Buttenfield, Leyk, & Spielman. 2014. “Dasymetric Modeling and Uncertainty”. Annals of the Association of American Geographers. DOI: 10.1080/00045608.2013.843439
- Tucillo, Stewart, Rose, Trombley, Moehl, Nagle, & Budhendra. 2023. “UrbanPop: A spatial microsimulation framework for exploring demographic influences on human dynamics” DOI: 10.1016/j.apgeog.2022.102844
Thank you for the reply!