I'm using CPS to calculate LFP for a subset of workers. I think I use stata code pweight=wtfinl.


I downloaded 12 years of individual, monthly data for all 51 states along with demographic information into Stata. I’m using CPS to regress predict state-month labor force participation for a subset of individuals (probably females with less than college education).

I’ve used your online tutorials (very helpful!) to figure everything out so far except for the weighting. Weighting is covered in the tutorials, but it doesn’t get at my question perfectly and it’s very important I get this right.

I am 100% sure I have my regression right, but am only 80% confident on the weighting…

reg Y X1 X2 if X3=value, [pweight=WTFINL]

I’m hoping you can confirm that WTFINL is the proper weight. And that it’s a sampling weight (thus, the ‘pweight’ command).

Thank you for any guidance you can provide!!


You are correct, WTFINL is the proper weight when using basic monthly IPUMS CPS data. Additionally, WTFINL is a sampling weight that adjusts for the following factors: failure to obtain an interview; sampling within large sample units; adjustments to the known distribution of the entire population according to age, sex, and race; and allotting a weight of zero to populations not sampled in other monthly surveys (i.e., persons in the Hispanic oversample and members of the armed forces in ASEC samples).

I hope this helps.