Computational/quantitative methods are increasingly utilized in political science and policy research. Do political leaders matter? Does canvassing change people’s political attitudes? How can you find election fraud? This course will teach students computational skills for addressing these questions using quantitative data. The course will require students to work on a number of hands-on exercises in programming language R with real-world data. The goal is to provide students with basic knowledge on how to implement data analysis as well as how to visualize their results to effectively convey their evidence-based arguments.
This is the first course to provide the foundation in statistics that students will need in the Political Science graduate methods sequence. Topics covered include probability theory, sampling distributions, sampling theory, confidence intervals, hypothesis testing, and strategies of data analysis. The course assumes students has taken Political Science Math Camp or equivalent. A previous background in statistics is desirable but not required for taking the course.
This course is designed to introduce students to modern Bayesian data analysis with emphasis on applications in social sciences. While the course is largely an applied course, it is intended to provide modeling and computational tools to its students so that they will be able to develop new applied models for analyzing original data in future research. Topics covered include Bayesian regression models, item response theory models, topic models, sampling methods, approximate Bayesian inference, and Bayesian nonparametrics. Prerequisites are POLSCI 599 and 699, or familiarity with basic mathematical statistics and regression analysis equivalent to these courses.