Hello, I work on the 1980 Brazilian Census.
The household weight variable (WTHH) has some values equal to zero. Is this normal?
I find this weird. If I understand well, the households concerned are only useful for some person-level analysis: their household members represent part of the Brazilian population but the households themselves are not representative of anything.
Thank you for your help. The whole job you are doing guys at IPUMS is unbelievable!
For the Brazil 1980 sample, household questions were asked only of those living in “permanent private households.” You can see this in the Brazil survey text. Other dwelling classes are not included in the household variables and are given a household weight of zero. However, the class of a respondent’s dwelling does not exclude her from the person-level questions. Thus, an observation can have a household weight of zero, while still having a non-zero person weight.
I hope this helps.