Graduate Courses

17.800: Quantitative Research Methods I: Probability, Statistics and Regression Analysis

This is the first course in the four-course sequence on quantitative research methods in the Ph.D. program at the MIT political science department. This course provides a graduate-level introduction to regression models, along with the basic principles of probability and statistics which are essential for understanding how regression works. Regression models are routinely used in political science, policy research, and other disciplines in social science. The principles learned in this course also provide a foundation for the general understanding of quantitative political methodology.

I co-taught this course most recently in Fall 2014 with In Song Kim. The syllabus can be downloaded here; other course materials are available for class participants on the course website.

 

17.802: Quantitative Research Methods II: Causal Inference

This is the second course in the quantitative research methods sequence at the MIT political science department. 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, matching, regression, panel methods, difference-in-differences, synthetic control methods, instrumental variable estimation, regression discontinuity designs, causal mediation analysis, nonparametric bounds, and sensitivity analysis. We will analyze the strengths and weaknesses of these methods. Applications are drawn from various fields including political science, public policy, economics, and sociology.

I am teaching this course in Spring 2016. The syllabus can be downloaded here; other course materials will be made available for class participants on the course website.

 

17.804: Quantitative Research Methods III: Generalized Linear Models and Extensions

This course is the third course in the quantitative research methods sequence at the MIT political science department. Building on the first two courses of the sequence (17.800 and 17.802), this class covers advanced statistical tools for empirical analysis in modern political science. Our focus in this course will be on techniques for model-based inference, including various regression models for cross-section data (e.g., binary outcome models, discrete choice models, sample selection models, event count models, survival outcome models, etc.) as well as grouped data (e.g., mixed effects models and hierarchical models). This complements the methods for design-based inference primarily covered in the previous course of the sequence. This course also covers basics of the fundamental statistical principles underlying these models (e.g., maximum likelihood theory, theory of generalized linear models, Bayesian statistics) as well as a variety of estimation techniques (e.g., numerical optimization, bootstrap, Markov chain Monte Carlo). The ultimate goal of this course is to provide students with adequate methodological skills for conducting cutting-edge empirical research in their own fields of substantive interest.

I taught this course most recently in Fall 2015. The syllabus can be downloaded here; other course materials are available for class participants on the course website.

 

17.806: Quantitative Research Methods IV: Advanced Topics

This course is the fourth and final course in the quantitative methods sequence at the MIT political science department. The course covers various advanced topics in applied statistics, including those that have only recently been developed in the methodological literature and are yet to be widely applied in political science. The topics for this year are organized into three broad areas: (1) Advanced causal inference, where we build on the basic materials covered earlier in the sequence (17.802) and study more advanced topics; (2) statistical learning, where we provide an overview of machine learning, one of the most active subfields in applied statistics in the past decade; and (3) Bayesian inference and statistical computing, where we extend the model-based inference techniques covered in the previous course of the sequence (17.804) and study more technically sophisticated materials as well as applications in political science.

I co-taught this course in Spring 2013 with Danny Hidalgo. The syllabus can be downloaded here; other course materials are available for class participants on the course website.

 

Undergraduate Courses

17.871: Political Science Laboratory

The purpose of this class is to introduce undergraduate political scientists to the basic quantitative tools of political science research. In particular, this class explores the key statistical and computational research tools that social scientists use to frame and answer empirical questions. When students finish this subject successfully, they will be able to conduct quantitative research, be better able to read critically much of the professional literature in political science and other statistically-based fields, and have an employable skill. A particular focus of the class will be on the issue of causal inference. The political world is composed of a web of cause-and-effect relationships that are entangled and intertwined. The complex nature of our world makes our life as political scientists tough and challenging, even compared to those of rocket scientists and nuclear physicists. The central theme that runs throughout the course will be: How can we tell causation from mere association? The answer lies in good research designs and appropriate statistical tools, as students will learn by the end of the semester.

I am teaching this course in Spring 2016. The syllabus can be downloaded here; other course materials are available for class participants on the course website.