Help using spmcapxpns and weights in Stata

Extremely novice Stata and IPUMS user here. I downloaded 2012 - 2017 annual samples using and capped child care expenditures from the supplemental poverty measures to try and study changes in child care costs across time. First, are capped costs correct to use? And also, if there’s a Stata expert on staff, I can’t for the life of me figure out how to weight spmcapxpns properly using spmwtin Stata.

Thanks!

Wouldn’t say I’m a Stata expert but I think I know enough to help you out.

First, if you want just child care expenses, use SPMCHXPNS (https://cps.ipums.org/cps-action/vari…). The SPMCAPXPNS combines both work expenses and childcare expenses and then caps that value. The capping procedure is a but more complicated but the overriding logic is that these expenses shouldn’t exceed the reported earnings of the lowest earner in the family. You can find this in the appendix of the SPM report. (https://cps.ipums.org/cps/resources/s…)

Regarding weighting, the SPM variables are calculated at the SPM family unit level. You’ll want to bring in SPMFAMUNIT (https://cps.ipums.org/cps-action/vari…) and select one person per SPM family unit (or figure out another way of alloting the child care expenses at the individual level, if you’re trying to stick with individual-level analysis). If you do analysis at the SPM level, you’ll use SPMWT. If you do individual-level, use ASECWT.

To use the weights, I typically use:

svyset [pweight=ASECWT]

svy: mean SPMCHXPNS

This may be clearer: https://www.stata.com/manuals13/svy.pdf

Hope this helps!