Is there such a dataset that provides a count of households by household size across different household income levels. For example, how many 1-person households make below $25,000, how many 2-people households make between $25,000 to $50,000, etc.) I don't believe this is published online, but could this be accessed using PUMS data?
Yes, the Public Use Microdata Sample (PUMS) data from the U.S. Census Bureau can be used to analyze the distribution of household sizes across various income levels. By leveraging PUMS data, you can derive insights into the number of households based on size and income brackets, allowing for a detailed socioeconomic analysis.
Note that you cannot currently crosswalk the ACS PUMAs from 2010 to 2020, so you cannot derive these results at the PUMA level. But perhaps there's another way to do this that I'm not aware of...
Yes, you can. Geocorr can create crosswalks between 2020 and 2010 PUMAs, and I'm sure NHGIS has them, too.
Interesting -- so if the analyst is summing the person weight in the microdata file, you would use the individual weighted allocation factor to crosswalk 2010 to 2020?
For example, if I summed the individual weights (PWGPT) for PUMA10/12 060101 and got ~10k folks, I would allocate those folks to the new PUMA20/22 based on the individual weighted allocation factor (.97, .01. 004, etc.)?
And same with housing weights (WGPT) and housing allocation factor?