Thesis: Estimating population with high spatial resolution to complement censuses in data-scarce contexts
Supervisors: Professor Ridhi Kashyap and Dr Douglas Leasure
Edith is a DPhil student at the Department of Sociology. Previously, she was part of the Spatial Statistical Population Modelling team in the WorldPop Research Group at the University of Southampton, developing Bayesian statistical models and applying machine learning approaches to produce high resolution population estimates supporting initiatives of the Bill and Melinda Gates Foundation and the United Nations Population Fund. She applied her skills in the academic (Worldpop, University of Southampton), in the humanitarian (UN World Food Program) and in the governmental (Etalab, France) sectors.
In her thesis, Edith will investigate predicting population in data-scarce contexts. A first data-scarce context is geographical and consists in areas impacted by conflicts. A great case study is Sahelian countries where she studies the impact of the multiple recently released built-up maps on population modelling. The second data-scarce context is temporal and consists in intercensal years, where she will focus on Colombia, using administrative records and digital traces to nowcast census count. Finally, she is interested in the matters of concern arising from this new co-production of official statistics between statisticians from government, research institutes and international institutions.
Research Interests: Edith’s research interests lie in the articulation of population, geography and Bayesian statistics through the lens of new data sources such as satellite imagery and digital traces.