Creating and Utilising Synthetic Population Data: Examples, Innovations and Pitfalls

Through their ability to fill important data gaps, synthetic populations have become well-established resources in research spanning a wide range of population geography aligned disciplines. By providing readily available data on key life domains and for entire populations, the role of synthetic populations is ever growing – for example within the context of modelling policy questions around public budgets, urban planning, climate mitigation, or health inequalities. Nevertheless, many approaches to creating synthetic populations present important limitations, impacting their robustness and utility for policy and research. In this talk I will outline the creation and utility of synthetic population data, covering important innovations such as the nesting of household and individual level structures, validation, approaches to sharing datasets, and undertaking applied research based on these datasets.