Causal Inference for Observational Data

About

The course opens with the potential outcomes framework, which provides a rigorous foundation for thinking about causality. From there, you will learn about a range of research designs and methods – including matching, instrumental variables, difference-in-differences, and regression discontinuity – examining the assumptions each requires and the questions each is best suited to answer

Course leader

Gloria Gennaro is Assistant Professor in Public Policy and Data Science in the UCL Department of Political Science. Her research focuses on the political economy of advanced democracies in Western Europe and in the U.S. using causal inference techniques, machine learning and natural language processing. She has published work on the drivers and strategy of right-wing populism and anti-immigrant sentiment, and on remedies against online hate speech. She has extensive experience teaching quantitative research methods, including causal inference and statistical learning.

Target group

Social science researchers

Course aim

The focus is on developing a clear understanding of each design: its logic, its key assumptions, and the trade-offs involved. The aim is to build the critical thinking needed to evaluate their application in published research and to adapt them to new settings.

Throughout the course, you are encouraged to connect what you learn to your own research interests. You will learn how to choose between designs based on their research question and available data, and how to tailor those designs to their own substantive needs.

Fee info