I am working on mapping the districts to city/town using the census tiger database for 2020. unfortunately, I could not find any relational keys to map the congressional districts to cities/towns.
Another alternative would be to use the 2020 Census Block Assignment Files and identify the city/town and district relationship. www.census.gov/.../block-assignment-files.html
I have reviewed one of the files and it seems this is what I was looking for. thank you.
Personally I didn't see any accuracy problems, though it took me a few tries to get just the right proportion variables set so I saw what I wanted to see as a proportion: city X is y percent of CD Z by…
You could do it with mapping -- overlay towns/cities on CD shapes and use geoprocessing functions to find the pieces (there ARE pieces, they don't align to city boundaries necessarily) of districts in each city/town and vice versa. A shortcut would be the University of Missouri's GeoCorr engine which will write you a report including estimated population in each piece for the current Congress and fairly recent but not current population -- this 2018 version DOES include 2019 Congressional shapes which are the current ones mcdc.missouri.edu/.../geocorr2018.html
tried that but could not get accurate results.
How do you know the Geocorr results aren't "accurate"? Do you have something to compare it to?
Personally I didn't see any accuracy problems, though it took me a few tries to get just the right proportion variables set so I saw what I wanted to see as a proportion: city X is y percent of CD Z by population or area etc. The ones I checked did check out, though I did wonder about a few really tiny population numbers like Joliet IL having 3.05 people in CD 16, but that could be an artifact of many things like parkland or a bit of near-vacant property that gets thrown in