Do weights adjust for non-independence of children within households?
I wanted to know whether the weights provided for the youth self-completion data adjust for the problem of non-independence due to different children coming from the same household. Thank you.
Updated by Olena Kaminska over 10 years ago
Thank you for your question. Weights do not adjust for so-called clustering or nesting within households. They adjust for unequal selection probability and for nonresponse. If you are using svyset command to adjust for complex sample design you should use household or psu as a cluster, in addition to a weight and strata. We suggest using psu as a cluster. Because a number of households are nested within psu, adjusting for clustering within psu will automatically adjust for clustering within households. The cluster and strata can be found in a_hhsamp.dta .