ABSTRACT
As we saw in Part I, panels are excellently suited for causal modelling of individual and household behaviour. Keeping track of individuals and households during the life cycle enables researchers to study in depth the occurrence and duration of life events and the underlying causal mechanisms. The modelling can be carried out more accurately if information is available over a longer time period. In the case of socioeconomic panels such as the PSID (Panel Survey of Income Dynamics) in the USA, the SOEP (Sozial-Ökonomisch Panel) in Germany, the British Household Panel Study (BHPS) in Britain and the Dutch SEP (Sociaal-Ekonomisch Panel), a number of topics have been dealt with or are under study. The potential for panel research is high and what particularly has been learned from the experiences with American panel studies is that panels provide datasets of high quality permitting advanced methodological and substantive research. Since the start of the Dutch panel in 1984, a number of issues have been addressed in publications of both The Netherlands Central Bureau of Statistics and Tilburg University (see Berghman et al., 1988, 1990). Among these are labour market mobility, job search modelling, income dynamics, demographic changes and household formation (marriage, divorce and separation, birth, death, children leaving home, migration, etc.), consumption and saving behaviour, dynamic analysis of indebtedness, sociocultural changes (changes in attitudes), housing mobility, transition and duration analyses of social welfare and social security programmes (social assistance, unemployment, disability) and research on the social and economic position of various social groups such as the poor, the disabled, the unemployed and the elderly.
