POLI 450A: Political Methodology I (Regression)
Yiqing Xu, Stanford University
Lecture slides, worksheets, short videos, R code and data for the first course in Stanford’s political science PhD methods sequence. The course also suits advanced undergraduates in economics, political science, and other social sciences. The course covers point and interval estimation, hypothesis testing, simple and multiple regression, and robust and clustered standard errors. The R scripts in the repository draw the figures and print the numbers on the slides from public data. The syllabus lists the topics, readings, and course policies, including the policy on generative AI.
The slides have one page per slide. To present with the step-by-step reveals, build that version from the LaTeX source. The worksheets are the in-class exercises. Each lecture title opens the lecture’s folder on GitHub: the LaTeX source, the figures, the R scripts that draw them, and the data.
Each lecture also has a video of five to twelve minutes, all in one YouTube playlist. The videos are designed to be conceptual and intuitive. Each presents the main ideas, one or two real examples, and the numbers from the lecture’s R code; the details of the mathematical derivations are covered in class.
Pedagogy
The videos pin down the conceptual issues, so that class time can go to the worksheet questions, the mathematics, and deeper insights. In class, we watch each video in segments, one part at a time, so that each concept is tied directly to its mathematical derivation and to examples. Between segments, we pause to settle conceptual questions, turn to the slides for the derivations, and work through worksheet questions with the students at the whiteboard. Outside class, students can watch a video before the lecture as a preview and again afterwards as a review.
Prerequisites
Linear algebra and calculus at the level of a political science math camp, and some undergraduate probability, statistics and statistical computing. The course uses R.