Relational Databases in the Social and Health Sciences: The View from Demography

October 26, 2006
Samuel Clark | University of Washington

There is an ongoing explosion in the amount and complexity of data collected by social and health scientists. Many of these are temporal data that describe time-evolving populations of individual people and their relationships between each other. Traditional training in social and health sciences does not contain a ‘data management’ course, and consequently most social and health scientists are under prepared to properly manage, manipulate, and archive the data that they collect. This has led to significant challenges and provides an opportunity to bring modern data management practices into the main stream of social and health sciences. This presentation will describe the growing use of relational databases: 1) in long-term studies of social and health issues in the developing world, 2) in building mathematical/computational models of human populations, 3) in the ‘mining’ and analysis of longitudinal population-based data, and 4) in the presentation and dissemination of those data. The presentation will end with a discussion of what may be needed to encourage social and health scientists to take more interest in data management and adopt more modern methods of managing their data.

Speaker Details

Samuel Clark is a Demographer in the Sociology Department at the University of Washington, Seattle. His work combines empirical investigation of the impacts of HIV on African populations, simulating African populations with HIV under different treatment conditions, and designing data management systems for long-term population and health research sites in the developing world. To these activities he brings training in engineering and biology at the California Institute of Technology and a PhD in Demography from the University of Pennsylvania. He has spent four of the past six years living and working in South Africa and maintains close working relationships with a range of colleagues there. His recent work investigates the impact of HIV on the elderly in Africa focusing on the creation of orphans, examines the impact of HIV on labor migrants in rural northeast South Africa, and proposes a general metadata-driven schema for a temporal relational database that is able to manage the complex inter-related histories of large numbers of human beings.