Hi there,
Can someone explain to me what the deal is with the county level data not being available in recent years? Why did they stop imputting that information correctly?
The main reason is due to confidentiality restrictions. Some counties don’t have a large enough population to warrant geographic identification. Instead the PUMA (public use microdata area) is the lowest level geographic area in public use data since 1950. It is sometimes possible to identify counties post-1950 by imputing based on other lower-level geographic identifiers. Please see this spreadsheet for a summary of when specific counties are identifiable.
Great thank you Jeff, very helpful answer. And I see that many of the PUMAs do indeed try and follow county lines, so that is helpful.