Causal Inference

Many of the theoretical claims we social scientists make are causal. Consensus democracies cause higher policy outcomes in terms of health, wealth, happiness or security. Ethnic residential segregation discourages immigrants to adapt to the host-country. Stable employment perspectives encourage researchers to work more efficiently.

More recently, the demand to explicitly test the causal implications of our theories empirically is increasing. The counterfactual approach to causality is powerful framework to think about causality. From this perspective, Joscha Legewie will present the basic ideas of various strategies to identify causal effects, such as Fixed-Effects, Difference-in-Difference and Instrumental Regression as well as Regression Discontinuity Design.

Resources:
Gelman, Andrew/Hill, Jennifer J. (2007): "Causal inference using regression on the treatment variable". In Andrew Gelman/JenniferHill: Data Analysis using Regression and Multilevel/Hierarchichal Models. Cambridge: Cambridge University Press.