ACS Data Users Group

I'm looking at the official poverty level and comparing to the supplemental poverty level using MDAT. I understand the basic differences between the two measures. I'm not fully understanding why there would be significant differences in the number of people "Not in Poverty Universe." Any ideas?

UPDATE: actually, I might not be accessing the supplemental data properly. I'm using the CPS Annual Social and Economic (March) Supplement (202303) with the "Perlis" variable.

Unrelated individuals under 15. They are excluded from official poverty but included in the supplemental measure.

https://www.census.gov/newsroom/blogs/random-samplings/2023/09/difference-supplemental-and-official-poverty-measures.html

Thank you, Matthew!! The documentation is very helpful too.

Dear Lorna,

Here is some additional documentation that might be useful.

Here is some documentation on how the income and expense items in the SPM PUMS file are computed.

https://www2.census.gov/programs-surveys/supplemental-poverty-measure/technical-documentation/spm_techdoc.pdf

As you noted the additional data collection in from the CPS ASEC.

Here are some additional considerations.

In some cases there are more people below the supplemental poverty threshold than the "ordinary" threshold in other cases the relationship is reversed. The "regular" poverty threshold is based on total family income, the family size and the number of children. On the other hand the supplemental poverty measure is base on a different measure of income (government benefits are counted) and a computation of expenses. Look at the SPM PUMS codebook for the variables and the exact computation. The ordinary measure does not account for expenses for example housing costs and food costs. Both of these vary by geography. Also the official poverty is based on the family (related individuals) There may be non-family members in the household.

Here is a good summary:

https://www.ssa.gov/policy/docs/ssb/v75n3/v75n3p55.html

Dave

Another good less technical resource from the Congressional Research Service is::

crsreports.congress.gov/.../R45031

You should also consider going directly to the SPM PUMS SAS file (2021) and merge on SERIALNO & SPORDER against the 2021 1 year PUMS. There isn't an exact match as the regular PUMS data includes some records for vacant households.