Abstract
The paper focuses on the measurement problems of family structure. In the case of family structure identification, there is reasonable suspicion that the applied measurement will erroneously put some blended families into standard categories (Gennetian 2005, Ginther and Pollak 2004).
The source of bias is the detection of kinship only between parent and child without control the relationship between siblings. The problem is illustrated on example of blended families with half siblings and step siblings living in the same household.
Given the current variability in family structure, it would be a mistake to assume a priori that siblings living in the same household have shared both biological parents. The Information about kinship between siblings eliminates the misclassification of blended family as a biological family.
Further advantage is the ability to show the history of changes in family structure.