This paper concentrates on the trends in peer-reviewed longitudinal panel studies under scientific direction. Household panel studies have succeeded in broadening their disciplinary scope. Numerous innovations such as questions dealing with psychological concepts, and age-specific topical modules, physical health measures, measures of cognitive capabilities, behavioral experiments have been incorporated into various panel studies or are soon to be introduced. In the UK, the household panel study Understanding Society comprising 40,000 households was launched in 2009 and recently added an 'innovation sample'; in the Netherlands, the new LISS household panel study launched in 2006 with over 5,000 households will be used for the testing of innovative measurement methods. The microdata from household panel studies like PSID (US Household Panel Study), BHPS (the predecessor of UK HLS), HILDA (Australian Panel Study), and SOEP (German Socio-Economic Panel) are in continuously high demand by the research and policy advisory community. More important than 'discovering' entirely new survey areas is 'tailoring' the details of existing survey content to new, more specific (theoretical) questions, and thus maintaining proven and widely used elements of survey content. In the years to come, 'tailoring' survey content will be the real challenge facing surveys that are integrated into the existing research infrastructure like HILDA, LISS, PSID, SHP (Swiss Panel), SOEP, and Understanding Society. We argue that, in the future, household panel studies should be designed to take the 'margins' of the life course more fully into account. Indeed, household surveys are ideally suited to gather comprehensive data on these life phases. They can be improved, on the one hand, by including specific topics about the fetal phase of life and early childhood of children born into the panel, and on the other hand, by including better information about late life and death. In the middle of the life course, improved questions on income, savings, consumption, and wealth, as well as psychological constructs will play a central role, as will specific 'event-triggered' questionnaires on central life occurrences such as marriage, divorce, and entry into and exit from unemployment. In order to substantially improve the statistical power of long-term longitudinal data, we propose an absolute minimum number of observations of about 500 persons per birth and age cohort. As of now, only the British Understanding Society will meet this target. A positive side-effect of such an enlargement is a significantly improved potential for analyses of relatively small groups within the population: for example, lone parents or specific immigrant groups. Another positive side-effect would be an improved potential for regional analyses. For example, in Germany, a cohort size of about 500 persons implies a survey sample size of about 20,000 households, which is large enough for analyses in the majority of federal states. Multidisciplinary panel studies will become even more important if they are accepted as reference datasets for specialized surveys that are independent of the original panel study (e.g., observational studies such as twin studies and laboratory or intervention studies). To enhance this important function, new types of service are needed, including advice on special surveys and possibly also data preparation for special surveys.
