ACS Data Users Group

Hi All,

I'm currently working on a project for a client in which I'm analyzing demographic shifts (population, race/ethnicity, age, housing, etc) by state and congressional district, comparing 2010 to 2020 data. I collected the 2010 ACS data several months ago and wasn't paying attention enough to anticipate the (understandable) data collection issues with ACS. Given what folks might know about the experimental data and decennial census data, I'm curious to know if there's a case to be made that I should pivot to using redistricting data, instead of using ACS experimental for 2020 data.

Thanks!

Derek

Given that you're comparing two decennial periods, I'd be inclined to compare 2010 Census to 2020 Census. Decennial might not have all of the variables of interest, but solves the 2020 data availability problem.

Thanks, Beth! One quick follow up. Does the decennial report on population by age? I didn't see that, but wanted to confirm.

In my demographics analysis, I'm reporting on total pop, by age and race/ethnicity. By state and by congressional district. So if decennial doesn't include age, and I compare two decennial periods for total pop and race/ethnicity, I might need to use 2019 if ACS experimental. That seem right?

The only 2020 decennial data yet released, the redistricting data, include separate counts for population age 18 and over. You can subtract these counts from the total counts to get population under 18. These age breakdowns are available by race and Hispanic origin.

Future releases of 2020 decennial data, scheduled for 2022, will include complete age breakdowns.

Besides the 2019 ACS data, you could also get 2020 estimates of state-level population by age through the Census Population Estimates Program, but that doesn't include data for congressional districts.