Arbeitsbereich

Bevölkerungsdynamik und Nachhaltiges Wohlbefinden

Auf einen Blick Projekte Publikationen Team

Forschungsfeld

Family and the Life Course

The research area Family and the Life Course investigates how family structures, intergenerational relationships, and individual life trajectories evolve under conditions of demographic change, social inequality, and technological transformation. It brings together demographic theory, computational methods, and new data sources to advance our understanding of how lives unfold within extended families over time.

A central feature of this research area is its commitment to studying the life course as a dynamic and interconnected process. Rather than treating events such as employment transitions, migration, caregiving, or fertility as isolated phenomena, the research emphasizes how these experiences accumulate, interact, and shape long-term outcomes for individuals and families. This perspective is reflected in projects that examine trajectories of employment and health, the consequences of mobility for mental well-being, and the distribution of care and resources across generations. By integrating multiple domains of life, this research reveals how inequalities emerge, persist, and potentially widen over time.

Methodologically, this research area advances the use of computational approaches in demography. An area of innovation is the application of machine learning, in particular sequence modeling techniques inspired by transformer architectures, to predict and interpret life course trajectories. These approaches move beyond traditional descriptive analyses by modeling entire sequences of states, quantifying uncertainty in future pathways, and enabling counterfactual “what if” scenarios. This allows researchers not only to identify groups at risk of adverse outcomes, such as employment instability, but also to explore how alternative policy interventions might alter life trajectories.

A second distinctive theme is the development and application of computationally intensive kinship microsimulation. By reconstructing kinship networks across time and space, these models provide unprecedented insight into how individuals gain and lose family members throughout their lives. This includes the study of bereavement, orphanhood, and the overlap of generations; phenomena that are fundamental to understanding the availability of social support and the distribution of vulnerability and dependency within populations. Importantly, these methods enable analysis even in contexts where survey data are incomplete or unavailable, thereby extending demographic research to historically and geographically diverse settings.

This research area also advances the measurement and understanding of fertility, particularly in data-deficient contexts. By developing Bayesian modeling frameworks and combining survey data with novel sources, such as social media- and network-based data collections, we aim to generate fertility indicators for contexts for which we would not otherwise have relevant or timely information. These approaches address longstanding challenges related to missing data, small sample sizes, and limited vital registration systems, while also expanding the scope of fertility research to include understudied populations.

Across all projects, there is a strong emphasis on methodological innovation coupled with substantive relevance. The integration of formal demographic methods with Bayesian statistics, computationally intensive approaches, and our own primary data collections allows for richer representations of individual and population dynamics, with implications for the sustainability of intergenerational support systems and the well-being of vulnerable groups.

Projekte dieses Forschungsfelds