Overview
The core faculty of PML run the Models and Methods (M&M) field for the political science Ph.D. program at MIT. We teach various courses in the field, both for students who choose M&M as their first or second major field and for students who simply want to acquire advanced quantitative skills for their substantive empirical research. We also offer courses at the undergraduate level.
Our graduate course offerings are centered around the four-course Quantitative Research Methods sequence, which is normally taken while a student is in their first and second years of their graduate study. The Quant Sequence is supplemented by the two Math Camps which occur towards the end of summer. The sequence is designed to get students up to speed with the cutting-edge quantitative empirical research in political science and also to prepare them to become first-rate political methodologists.
Other graduate courses include the Formal Theory class, as well as seminar classes focused on the intersections of quantitative methodology and major substative areas of political science, such as political economy and international relations.
Graduate Courses
The Math Prefesher is designed to introduce and review core mathematics and probability prerequisites that you will need to be successful in the quantitative methods courses in the Political Science department and elsewhere at MIT. In an intense one-week course, we will cover key concepts from calculus, linear algebra, probability theory, and an introduction to statistical computing. The learning will proceed through lectures, hands-on exercises, and homework. The aim of the course is to give you an opportunity to practice some of the mathematics you may have previously learned and to introduce you to areas that may be new to you so that you will be ready to enter classes that presume prior familiarity with these concepts, such as 17.800 Quantitative Research Methods I. Syllabus
17.800 Quantitative Research Methods I: Regression
Graduate level introduction to statistical methods for political science and public policy research, with a focus on linear regression. Teaches students how to apply multiple regression models as used in much of political science and public policy research. Also covers fundamentals of probability and sampling theory. Syllabus.
17.802 Quantitative Research Methods II: Causal Inference
Survey of advanced empirical tools for political science and public policy research with a focus on statistical methods for causal inference, i.e. methods designed to address research questions that concern the impact of some potential cause (e.g., an intervention, a change in institutions, economic conditions, or policies) on some outcome (e.g., vote choice, income, election results, levels of violence). Covers a variety of causal inference designs, including experiments, matching, regression, panel methods, difference-in-differences, synthetic control methods, instrumental variable estimation, regression discontinuity designs, quantile regressions, and bounds. Syllabus.
The math camp will prepare students to take Quant III and Quant IV as well as other advance classes in political methodology. The goal of the class will be to remind students of basic and intermediate mathematical concepts that are useful for Quant III and Quant IV and increase both mathematical fluency and problem solving ability. I will also try to give some programming tools that you may find useful when solving problem sets of Quant III. The prerequisites include Quant I and Quant II. Syllabus.
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. Syllabus.
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) research computing, where we introduce various techniques for automated data collection, visualization, and analysis of massive datasets; (2) statistical learning, where we provide an overview of machine learning algorithms for predictive and descriptive inference; and (3) finite mixture models (e.g., Latent Dirichlet allocation for text analysis), as well as a variety of estimation techniques such as EM Algorithm and Variational Inference. Syllabus.
17.810: Game Theory and Political Theory
This course provides an introduction to formal theoretical analysis in political science. This course is designed as a rigorous introduction to the concepts and models used to analyze political behavior in strategic contexts. The course focuses on non-cooperative game theory covering normal and extensive form games, games of incomplete information, repeated games, and bargaining. Qualified undergraduates can also take the course. Syllabus.
17.830: Empirical Methods in Political Economy
This course surveys recent methodological approaches to the study of political economy. Unlike a typical graduate-level course in political science, we will focus on a limited number of readings each week, with the goal of understanding and evaluating in detail the analytical decisions made by the authors of each study. In addition to learning about advanced methods being used in the social sciences today, the goal of the class is for students to develop an appreciation for how publishable quantitative papers are constructed, from the questions they ask to how they defend and justify the methodological choices they make. Syllabus.
17.426: Empirical Models in International Relations
This course explores statistical methods as applied to international relations, with reference to similar applications in comparative politics and other fields . We will discuss statistical approaches to analyzing various types of data used by IR scholars. We will read both methodological and applied work, familiarizing students with an array of models and critically analyzing their strengths and weaknesses. It is not intended as a substitute for Quantitative Methods I, II, and III , but as a complementary course. The goal of the course is to expose students to the range of quantitative models applied in international relations scholarship, assess the strengths and weaknesses of particular modeling choices, and to develop the ability to design empirical research projects of their own. It is strongly recommended that students have taken Quantitative Methods I prior to this course. Syllabus
17.212: Formal Approaches to American Political Institutions
This is the second in a two-course graduate sequence on American political institutions, emphasizing the concepts and methods in formal theory used to analyze domestic politics. It is organized thematically, according to strategic interactions and social problems that institutions may both solve and exacerbate, such as delegation, collective action, commitment, and preference aggregation. For each of these themes, we will learn some basic game theoretic modeling techniques; closely read a few formative papers; and apply our tools to the analysis of a wide range of specific problems in American politics, including questions about elections, political participation, polarization, representation, the internal organization of Congress and the bureaucracy, separation of powers, campaign finance, redistribution, public goods provision, and the legislative process. Syllabus
17.850: Political Science Scope and Methods (Graduate)
The world is full of compelling stories, fascinating events, and baffling puzzles. But how do these ideas translate into research? The purpose of this course is to help you move from topics of interest to research questions, and to give you the skills necessary to answer those questions with solid, well- designed empirical research. The course draws on current research in political science to introduce you to the enterprise of scientific research in politics. Specifically, the course reviews the basic principles of research design and eval uates the strengths and weaknesses of various empirical approaches. Mastering these skills —indeed internalizing them so that they become second nature —is one of the most important things that you will learn in graduate school. You will emerge from this course not only a more sophisticated consumer of the literature, but in a position to design and conduct your own independent scholarly research. Syllabus
17.878: Qualitative Methods and Fieldwork
This course is intended for political science PhD students, though we will also be drawing on sociology and anthropology. By the end of the course, students should be well-equipped to undertake their own fieldwork. Students will also be familiar with the major debates surrounding qualitative research in the discipline, and they will be able to confidently assess the design, execution, and interpretation of qualitative field research. Syllabus.
17.S953: New Methods for Causal Inference
This is a graduate-level seminar class on recent advancements in the field of statistical methods for causal inference. The purpose of this class is to provide students with experience and skills that are necessary to conduct research on methodological topics professionally. Although the class focuses on methods for causal inference, many of the research skills students will learn in the class will be transportable to methodological research in other subfields. After taking this class, students will be able to read typical articles from journals like the Journal of the American Statistical Association and Political Analysis quite comfortably. They will also be ready to embark on a methodological research project independently, particularly in the field of causal inference. Finally, they will also have built familiarity with cutting-edge causal inference methods potentially useful for their applied work. Syllabus.
17.S950: Bayesian Measurement Models
This course covers quantitative measurement from a Bayesian perspective. It focuses on the specification of measurement models linking observed data (i.e., manifest indicators) to unobserved constructs (i.e., latent variables) of interest. For estimation of these models, we will rely on the probabilistic programming language Stan, as called from R, though we will occasionally touch on other R-based methods. The goal is to get students comfortable specifying and estimating “bespoke” measurement models tailored for particular applications. The course applies this basic framework to a large range of problems and topics, including hierarchical models, factor analysis, item response theory, latent class analysis, ecological inference, network data, and text analysis. Each is covered only in enough depth to provide a sense of what a Bayesian approach to the problem might look like. The course assumes a solid grasp of generalized linear models and the theory of likelihood and Bayesian inference, so successful completion of 17.804 (Quantitative Research Methods III) or its equivalent is a prerequisite for enrollment. Syllabus.
Undergraduate Courses
17.801: Political Science Scope and Methods (Undergraduate)
This course introduces principles of empirical and theoretical analysis in political science through research projects currently conducted in the department. Introduces students to major research questions in political science - and to different ways of examining those questions. Emphasizes how this research in progress relates to larger themes, and how researchers confront obstacles to inference in political science. Includes substantial instruction and practice in writing (with revision) and oral presentations. Syllabus.
17.803: Political Science Laboratory
This class introduces undergraduate political scientists to the basic quantitative tools of political science research. The central theme that runs throughout the course will be causal inference, or how we can distinguish causation from mere association when studying complex political and social phenomena. This class emphasizes practical skills, and involves hands-on exercises, lab sessions, group work, discussion and presentation sessions along with more traditional problem sets. Throughout the semester, students will work on an original research project that involves data collection, analysis with a statistical computing language (R), and a final write-up of their findings. Syllabus.
17.811: Game Theory and Political Theory
This course provides an introduction to formal theoretical analysis in political science. This course is designed as a rigorous introduction to the concepts and models used to analyze political behavior in strategic contexts. The course focuses on non-cooperative game theory covering normal and extensive form games, games of incomplete information, repeated games, and bargaining. Qualified undergraduates can also take the course. Syllabus.
In this course, students will both learn how statistics are changing elections and how to use statis- tics to analyze political data. While the substantive focus will be on elections, the principles and methods learned in this course have broad applicability to the decision-making in a broad variety of fields. The course will be roughly divided into 4 sections organized around a different methodolog- ical topic, with an application to an electoral phenomenon. For each section, students will work with the professor on analyzing a unique dataset related to electoral politics. The first section will focus on data description and dimension reduction. The second section will involve the analysis of survey data on electoral behavior. The third section will use statistical models to predict electoral behavior using large datasets. The fourth section will focus on the design and implementation of original experiments in order to study political attitudes and behaviors. Syllabus.
17.835 Machine Learning and Data Science in Politics
Empirical studies in political science is entering a new era of “Big Data” where a diverse range of data sources have become available to researchers. Examples include network data from political campaigns, data from social media generated by individuals, campaign contribution and lobbying expenditure made by firms and individuals, and massive amount of international trade flows data. How can we take advantage of these new data sources and improve our understanding of politics? This course introduces various machine learning methods and their applications in political science research. Syllabus.