Geo error in 1870 full count census

Hi IPUMS team,

I’m looking at the 1870 Census data and found a couple of counties with data but no matches in the gisjoin variable for the shapefile. Could you please provide me the names and icp codes for the following gisjoin codes









You can find the names of these counties in this document. However, you should note that these counties have no matches in the 1870 shapefiles because none of them actually existed in that year. With nearly all of these codes, I suspect the issue is likely due to miscoding in the microdata such as premature identification of counties that weren’t yet legally instituted or that were only ever in informal use. I’m not certain if this is the case for these codes, but such issues are a reality when working with the historical microdata.

Thanks Ivan.

No worries, I know how working with historical data is. You are doing a great job.



Hi Ivan,

two weeks ago you linked the county availability over time in this document

Do you have this as excel document?



I haven’t been able to track down the original source for that file. A similar alternative would be to get the time series table for Total Population, 1790-2020, for counties, through the NHGIS Data Finder. That will include all NHGIS codes used for any county in any decennial census’s summary data, along with population data for any year in which the county was reported.