asad <[email protected]> has posted a problem
on data on husbands and wives,
my understanding of which is as follows:
1. Data are held on individuals in households.
2. Some individuals are children or unmarried
adults.
3. Married adults can be husbands or wives.
4. Husbands may have one or more wives.
5. Wives have just one husband.
6. A household may contain more than one husband.
7. A spouse identifier is the identifier of her
husband in the case of a wife, and the identifier
of one of his wives in the case of a husband.
Spouse identifier allocation seemingly was arbitrary in the
case of polygynous husbands.
The aim is to identify groups of husbands and wives.
Once we have done that, we want to calculate
the (mean) age of the spouse of each married
adult.
