Teaching

MIT

• 17.802: Quantitative Research Methods II
  This is the second course in the quantitative research methods sequence at the MIT Political Science department. The goal of the four-course sequence is to teach you how to understand and confidently apply a variety of statistical methods and research designs that are essential for political science research. Building on the first course (17.800), which covered probability, statistics, and linear regression analysis, this second class provides a survey of more advanced empirical tools, with a particular focus on causal inference. We cover a variety of research designs and statistical methods for causal inference, including experiments, estimation under conditional ignorability, sensitivity analysis, instrumental variable estimation, regression discontinuity designs, difference-in-differences, panel methods, and synthetic control methods. We will analyze the strengths and weaknesses of these methods. Applications are drawn from various fields, including political science, public policy, economics, and sociology.

• 17.S953: Recent Developments in Political Methodology
  This course covers a wide range of recent methodological developments in political methodology, statistics and computer science. We first cover (1) machine learning and AI for causal inference, (2) generative AI and large language models for the social sciences, and (3) external validity. We then cover more topics depending on students’ interests. Potential topics include causal inference with network and spatial data, causal inference with texts, proximal causal learning, and modern experimental designs. The main goal of this course is to help students learn topics of their interest by combining broad surveys (covering many topics) and teaching modern statistical theories (fundamental to understanding and contributing to recent methodological innovations). Throughout the semester, students will collaborate to present discussion papers. Students will also write a final research paper that either proposes methodological extensions or applies a method learned in this course to an empirical problem of their substantive interest.

Columbia University

• POLS 4700: Mathematics and Statistics for Political Science
  This course presents basic mathematical and statistical concepts that are essential for formal and quantitative analysis in political science research. It prepares students for the graduate-level sequence on formal models and quantitative political methodology offered in the department. The first half of the course will cover basic mathematics, such as calculus and linear algebra. The second half of the course will focus on probability theory and statistics. We will rigorously cover the topics that are directly relevant to formal and quantitative analysis in political science such that students can build both intuitions and technical skills. There is no prerequisite. The course is aimed for both students with little exposure to mathematics and those who have taken some courses but wish to gain a more solid foundation.

• POLS 4722: Statistical Theory and Causal Inference
  This course is the second course in the graduate-level sequence on quantitative political methodology offered in the Department of Political Science. Students will learn (1) a framework and methodologies for making causal inferences from experimental and observational data, and (2) statistical theories essential for causal inference. Topics include randomized experiments, estimation under ignorability, instrumental variables, regression discontinuity, difference-in- differences, and causal inference with panel data. We also cover statistical theories, such as theories of ordinary least squares and maximum likelihood estimation, by connecting them to causal inference methods. This course builds on the materials covered in POLS 4700 and 4720 or their equivalent (i.e., probability, statistics, linear regression, logistic regression, and knowledge of statistical computing environment R).

• POLS 4726: Topics in Political Methodology
  This course is the fourth course in the graduate-level sequence on quantitative political methodology offered in the Department of Political Science. Students will learn a variety of advanced topics in political methodology, such as machine learning, recent measurement methods (e.g., ideal point estimation, text analysis, list experiment, and conjoint experiment), network analysis, and causal inference with network and spatial data. Students will collaborate to present discussion papers throughout the semester. The main goal of this course is to help students to write a final paper that applies or develops advanced statistical methods. This course builds on the materials covered in POLS 4700, 4720, 4722, and 4724, or their equivalent courses (i.e., probability, statistics, linear regression, logistic regression, causal inference with observational and experimental data, and knowledge of statistical computing environment R).