ASTK18021U Core Subject: Advanced quantitative methods in the study of political behavior

This seminar is part of the specialization in Political Behavior and Applied Quantitative Methods. It covers the applied approaches to quantitative methods that are commonly used in the study of political behavior. The course is designed to interact closely with Political Behavior—the specialization's substantive course—and incorporates its texts and themes. We will examine some of the seminal literature on political behavior and ask: How have the authors arrived at their results? Can we reproduce them? And how sensitive are they methodological choices? The course places particular emphasis on students' ability to independently apply the methods introduced in the course to real-world data.

The course revolves around three themes. The first, Brush-up (classes 1-4), re-introduces central aspects of Methods 2, up to and including regression analysis, and introduces the R statistical programming language.

For the second theme, Causal Inference (classes 5-11), we will introduce a paradigm for how to think about causality, and examine the primary methods for estimating causal effects with experimental and observational data. This theme arises from the fact that the social sciences have recently undergone a "credibility revolution" that places methods for the estimation of causal effects center-stage. In this tradition, the course will introduce research designs and statistical techniques for causal inference. Particular emphasis is placed on research design, because well-crafted designs can often eliminate the need for more advanced statistical methods that can rely on implausible assumptions.

For the third theme, Data Science (classes 12-14), we will introduce techniques for collecting and analyzing data from non-traditional sources, such as those from unstructured online data (e.g. text).

Tentative schedule

1. R Basics I

2. R Basics II

3. OLS

4. R Advanced & Data Visualization

5. Introduction to Causal Inference

6. Experiments I

7. Experiments II

8. Instrumental Variables

9. Panel Data

10. Difference in Differences

11. Regression Discontinuity Designs

12. Web-scraping

13. Text as data

14. Social Data Science