Hi all
I was looking for data on children under 18 in poverty by race, at the county level.
I found one set of tables, b17020a – b17020i, that show people in poverty by age. Each table is for one race. So b17020d is for Asian alone. B17020b is for Black or African American alone.
But these tables have three age groups for children:under 66 to 1112 to 17
Is there any table or set of tables that show number or percent of children, under 18, as a single age group, in poverty by race?
Thanks
Gene ShackmanOffice of Public Health PracticeNew York State Department of HealthAlbany NY
I'm new to the ACS but just spent some time looking at county-level data, it's hard, and age grouping are not consistent across tables. In case you didn't know. You can easily pull just the fields you want (even if they're in separate tables). As long as they're the same type of table. The max is 50 fields (I've pulled over 40).
Part 1
https://api.census.gov/data/2021/acs/acs1?get=
Part 2 (field names separated by a comma
B19055_001E
Part 3
,NAME&for=county:*
=
https://api.census.gov/data/2021/acs/acs1?get=B19055_001E,B19055_002E,NAME&for=county:*
The API formats are here;
https://www.census.gov/data/developers/data-sets/acs-1year.html
I hope this helps.
Dear Gene,
You need to sum/combine across the 3 age categories that you want. The calculation (including MoE - margin of error) is in https://www.census.gov/content/dam/Census/library/publications/2018/acs/acs_general_handbook_2018_ch08.pdf
For a percent of children under 18 to need to get the denominator by adding across the 6 categories above/below age under 6, 6 to 11, 12 to 17.
The MoE calculation for percents in is the above reference. You can probably do the calculations using Excel formulas. I use the R statistics package..
Dave Dorer
You might be looking for small area income and poverty estimates www.census.gov/.../saipe.html
Hi Tim. Thanks for the suggestion. I don't think SAIPE has race data, does it? At least the interactive tool doesn't show race. www.census.gov/.../
You’re right!
It would be great if it did have race.
One thing about SAIPE estimates they use an SAE (small area estimation) model to computes the SAIPE poverty variable The "small areas" are county and school district. For the SAIPE estimates the Census uses ACS data in combination with administrative data https://www.census.gov/programs-surveys/saipe/guidance/model-input-data.html The administrate data includes IRS data SNAP benefit data, and Current Population Survey (CPS) data Details and public use files are here https://www.census.gov/data/datasets/time-series/demo/saipe/model-tables.html and https://www.census.gov/programs-surveys/saipe/technical-documentation/methodology.html The details of the county model are here https://www.census.gov/programs-surveys/saipe/technical-documentation/methodology/counties-states/county-level.html By contrast the ACS "B" detail tables are "direct estimates."
In any case, the SAIPE, estimates may not be comparable to the ACS tables variable Above/Below Poverty Level, which I believe, relies on items from the ACS form alone. (Please correct me if needed). There is another recent thread about how the ACS calculates their poverty. See that for more details.
Yup, sounds right.