2020 ACS 5 Year estimates

Hello.

I'm looking for some suggestions, guidance or advice. I'm working on creating a language access plan for the city, I work in. In my draft I had used the latest information on LEPs from the 2019 ACS 5-Year Estimate. My question is, does it make sense to update my info with the recently release information from 2020 ACS 5-Year estimate or does it make more sense for me to wait until 2024 when it will be more comparable? I'm also thinking that if I stay with numbers from the 2019 ACS 5-year estimate, it will be more accurate since it was before the pandemic hit. Any thoughts or ideas on how I can take the best route for this? I'm in a good spot right now because we are just starting to work on language access services hence I want to make sure I use the best data.  

Thank you so much.

Liliam

Parents
  • Hi Liliam -

    In a typical year there is not much change from one 5-year ACS estimate to the next because 4 of the 5 years are the same (data for 2016, 2017, 2018, and 2019 show up in both the 2019 5-year and 2020 5-year). In addition the number of households who responded to the ACS was considerably lower in 2020, so the margins of error on each estimate are quite a bit larger than usual (see here: https://www.census.gov/programs-surveys/acs/technical-documentation/user-notes/2022-04.html). And small language groups already tend to have large margins of error. For that reason, if I were doing LEP data collection I would check the reliability of the 2020 estimates before making a decision. Either way it's ok to stick with the 2019 5-year.

    One other non-ACS thought: Have you also gotten language data from your local school districts? That may be a helpful resource in addition to the ACS.

    Hope that helps.

    -Beth

Reply
  • Hi Liliam -

    In a typical year there is not much change from one 5-year ACS estimate to the next because 4 of the 5 years are the same (data for 2016, 2017, 2018, and 2019 show up in both the 2019 5-year and 2020 5-year). In addition the number of households who responded to the ACS was considerably lower in 2020, so the margins of error on each estimate are quite a bit larger than usual (see here: https://www.census.gov/programs-surveys/acs/technical-documentation/user-notes/2022-04.html). And small language groups already tend to have large margins of error. For that reason, if I were doing LEP data collection I would check the reliability of the 2020 estimates before making a decision. Either way it's ok to stick with the 2019 5-year.

    One other non-ACS thought: Have you also gotten language data from your local school districts? That may be a helpful resource in addition to the ACS.

    Hope that helps.

    -Beth

Children
  • Good morning, Beth

    Thank you for your insight. That makes sense to me. I did notice that for the data I needed went from 9.4 percent to 8.7 and deduced it was because of the lower response. I did looked at the school districts numbers and it has been very helpful. Thank you!