300290 Causal Inference (MC: Soziologische Methoden - quantitativ) (S) (SoSe 2026)

In applied social science research, it is widely recognized that statistical correlations observed in non-experimental (observational) data do not necessarily imply causal relationships. Nevertheless, understanding causal effects remains essential for advancing theoretical frameworks and ensuring the practical relevance of our research. Only when the causal direction of effects—such as those between poverty and health—can be determined with sufficient certainty, are we able to design effective interventions to reduce poverty or improve health outcomes.

Building on statistical knowledge up to and including regression analysis, this seminar focuses on methods of causal inference using cross-sectional data. Key topics include matching methods (such as propensity score matching), instrumental variable approaches (IV regression), and regression discontinuity designs (RDD). The seminar offers a applied introduction to these techniques, drawing on published papers to demonstrate their implementation. In hands-on sessions, participants will apply these methods using Stata with example data; alternatively, R may be used. Beyond applied training in advanced regression techniques, the seminar aims to deepen participants' understanding of alternative research designs in sociology, ranging from classical and natural experiments to analyses of longitudinal and cross-sectional data.

Bibliography

Brady, Henry E. 2008. Causation and Explanation in Social Science. In: Box-Steffensmeier, Janet M., Brady, Henry E. und David Collier (Hrsg.). The Oxford Handbook of Political Methodology. Oxford: Oxford University Press, 217-270.

Markus Gangl. 2010. Causal Inference in Sociological Research. Annual Review of Sociology 36: 21-47.