0

I’m working with a dataset that crosses 2020 with data from 2015-2022. What is the best practice for geocoding within a dataset that spans two census' and multiple ACS years.

Should I be benchmarking to look at a single year snapshot for each ACS product/year or should I be benchmarking on the decennial census alone for Pre 2020 (2010) and post 2020?

I know that boundaries change between decennial census’ and when going to connect to tract level ACS data I will want to have the correct tract for correct 5 year estimates I’m going to get.

For those who are curious my code for 2021 results is here though likely not required for answering the question.

Thanks, Rez

mresner
  • 3
  • 3
  • Are you asking us to advise on how to do this research? Or are you asking for finding different APIs or how to connect/use specific APIs? – r2evans Aug 29 '23 at 23:16
  • The question is unclear, what do you mean by benchmarking? Can you provide an example? Also, have you looked at the [`tidycensus`](https://walker-data.com/tidycensus/) package? – bretauv Aug 30 '23 at 12:27
  • Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. – Community Aug 30 '23 at 22:02

1 Answers1

0

US Census got back with me after two weeks:

"American Community Survey (ACS) data uses boundaries from the last year of the 5-year estimate. Therefore, the 2015-2022 dataset will use 2020 tract boundaries (tracts are only updated every 10 years). It will therefore be consistent with your 2020 Decennial Census data."

mresner
  • 3
  • 3