A further discussion on revisiting the classification of household composition among elderly people.

This discussion is a reaction to the DeVos article on the classification of households also in the current issue of Journal of Cross-Cultural Gerontology. The need for cross-national and comparative gerontological research is growing. Household composition is a compelling subject for such research, but there is a concurrent increase in the need for flexible and valid classification schemes. The DeVos classification represents an excellent step toward understanding households on a global level. Additional considerations not covered by DeVos include the gender of the coresident child, pseudo-coresidence, and living with grandchildren. These additional issues may complicate classification and, in the end, it is up to the individual researcher to choose schemes based on research questions, guiding theory, and available data. Yer, having a standard as a starting point reduces the complexity of the task of defining a classification scheme to be used comparatively across societies.