Should I use individual weights (wtper) in order to compute per capita household income?



I want to compute per capita household income for each household in my Brazilian sample. I know that wthh should be used for household-level analysis and wtper for individual-level analysis but I have a doubt in this case.

Should I take into account the individual weight variable wtper, which differs across household members? Or should I not use it, and thus implicitly assume that all household members represent the same number of individuals (ie. have the same weight) within their household?

Thanks a lot for your help!


How do I get the number of households and average household size- IPUMS-I

Within any household, you should treat the persons as equally weighted. If you were to take into account different person weights when calculating your per-capita household income variable, then you would be altering the actual characteristics of a person’s household (i.e. the income distribution/amount within the household). Simply summing household income and dividing by the number of persons in the household will preserve the correct household-level information. This will then allow you to use the weight variables in order to generate accurate national estimates.

I hope this helps.



It’s very helpful. Thank you!