Causality plays an essential role in the social sciences whenever we wish to find out whether a particular intervention actually changes an outcome of interest. However, causality is difficult to study. It has long been known in the social sciences that correlation does not imply causation. Especially when working with observational data, researchers miss crucial information for making causal interpretations of statistical associations.
This module explores research designs and modern methods of causal inference for statistical research. This module will explore under which circumstances and subject to which assumptions, researchers can interpret estimated associations as causal with substantially higher confidence.
Amongst other things, the module will explore theories of causality and the potential outcome framework; the fundamental problem of causal inference and its possible solutions; ‘selection on observables’ identification strategies; and selection on unobservables’ identification strategies.
A module handbook and course material will be made available in April 2019.