The Centre for Microdata Methods and Practice (CeMMAP) is a world centre for research into the methods that reveal the nature of human decision processes using observational data. Its work is widely read, highly respected and influential amongst practitioners.
Faced with limited resources in an uncertain world, societies need guidance to attain good outcomes in the future. Governments need to know what works and what does not. Firms need to know the most effective strategies for succeeding in a dynamic marketplace. Individuals need to know what risks they face, what opportunities exist. Social and economic policy, and indeed individual decisions, cannot be undertaken successfully without high quality information and robust reliable analysis.
Research areas are divided between methods and practice. Choose one below to see the current strands of research.
The research below looks at methods for analysing microdata.
Understanding individuals’ responses to changes in environment and to changes in risk and uncertainty requires structural analysis of sequences of individuals’ decisions and outcomes. Models currently used for analysis of longitudinal data employ such strong restrictions that our understanding of important aspects of behaviour is very limited. Research is needed to enable analysis using models with essential non-linearities, models with multiple sources of heterogeneity and models suitable for application in complex dynamic decision environments where people’s computational capabilities may be limited.
The knowledge that can be obtained from empirical research in social science ultimately rests on the quantity and quality of measurements of social and economic phenomena. Our research on measurement is complementary to our research on identification.
Research on identification underpins all social science research and lies at the basis of all of CeMMAP’s endeavours.
While the study of identification is about what can be learned from economic models when applied to data, the study of estimation, inference and testing is concerned with, inter alia: the measurement of identified quantities using real data; summarising the precision of estimates; quantifying how well the data support particular hypotheses of interest; and quantifying the degree of support that the data give to the restrictions imposed by the model being employed.
This strand of the Centre’s research focusses on advancing computational social science.