Advanced Quantitative Methods in Political Science: Causal Inference, Panel Data and Multilevel Modeling

This course provides students with a comprehensive overview of advanced quantitative applications in political science, including methods of causal inference, panel data analysis and multilevel modelling. The key goal is to provide students with both a strong theoretical background in these methods and the knowledge to apply them in future research projects. The course builds upon skills acquired in foundational methods courses (Methods I and Methods II) and does not presuppose knowledge of mathematics or probability theory beyond what is introduced there. It consists of four different parts: (1) mathematical foundations, (2) theory and application of causal inference, (3) theory and application of panel data analysis and (4) theory and application of multilevel modelling. In the first part of the course, because a thorough understanding of advanced quantitative methods requires a strong mathematical foundation, we begin with a review of the concepts in probability theory and regression analysis that are most directly relevant to us. In the second part of the class, we discuss the mathematical theory and application of key causal inference tools, including matching, regression discontinuity designs, instrumental variables and differences-in-differences. In the third part of the class, we consider different types of panel data analysis, including fixed effects models, random effects models, panel models with instrumental variables and dynamic panel models. In the fourth and final part of the class, we introduce multilevel modelling, including multilevel linear regression, multilevel logistic regression and multilevel generalised models. During class, students will participate in a series of replication exercises through which they will learn how to apply these tools themselves. By the end of the course, students will formulate their own research question and present a research design that makes use of one of the advanced methods they have learned about.

After having participated actively in the course, the student will be able to:

• Understand how probability theory relates to advanced quantitative methods in political science, and apply concepts from the former to the latter.
• Describe the logic behind key tools of causal inference; and answer research questions by using these tools in applied research.
• Explain different panel data structures and employ a variety of tools to analyse these data.
• Recognise when data has a multilevel component, and make use of appropriate multilevel models to examine them.
• Formulate their own research question, select an appropriate statistical tool, and describe how the tool can be applied to answer the research question.