Empirical Political Science has become increasingly quantitative and technical, and we have seen a profusion of “advanced quantitative methods” that appear difficult to parse for “ordinary” students, let alone be replicated by them. The aim of this course is to show you that that appearance is wrong. After this course, you should be able to a) understand what these new models are trying to do, b) replicate existing empirical works in political science and most importantly c) be able to apply these tools to your own questions. In this course, you will not only familiarize yourself with new techniques of analysis (Experimental Methods, Regression Discontinuity Designs, Instrumental Variables, Panel Data and Multilevel Models) but also more fundamental theories of causal inference (Directed Acyclic Graphs (DAGs) and the Potential Outcomes Framework) and innovations in scientific best practice (Preregistrations, Power Analyses, Simulations).

Empirically, we will focus on examples from political psychology and populism research; however, the course is explicitly structured in ways to also include your own research interests. Prior familiarity with statistical methods and tools such as R, Python, or Stata are helpful for this course, but can also be acquired during the course. The course includes both synchronous and asynchronous elements.