I plan to collapse individual data from ACS down to create county mean values. Is it recommended to use weights?


#1

The ACS data I downloaded from ipums includes person weights (perwt) and household weights (hhwt). I would like to use Stata’s collapse command to collapse these data down to create county-level means (using county fips codes). Is it recommended that I include a weight command (e.g., aweight=perwt), when doing so?

Thank you.


#2

Yes, in order to obtain obtain representative statistics from the weighted samples, you must apply sample weights.

For person-level analyses using a weighted sample, apply the PERWT variable. For household-level analyses using a weighted sample, weight the households using the HHWT variable.

You can read more about weighting here.