Advanced Statistical Methods for Political Science

Catalog description

The goal of this class is to build upon the content of PSCI 1801 and expose students to the process by which quantitative political science research is conducted, The class will take us down three separate, but related tracks. Before engaging in any statistical analysis, we must think about why we engaging in that analysis. Thus, the first track will cover the basics of research design. Topics discussed will include what makes a good model, the art of theory building, the difference between a theory and a hypothesis, and common research designs. The second track will cover several statistical techniques that are frequently used in quantitative political science research that build upon multivariate regression analysis. Two types of statistical techniques will be emphasized. The first half of the course will focus on important statistical concepts for conducting research using survey data. Topics we cover include sampling strategies, estimating uncertainty, non-response, measurement error, and working with categorical variables. The second half of the course will focus on methods for establishing causal relationship between independent and dependent variables. Topics we will cover include the potential outcomes framework, experiments, panel data, instrumental variables, and regression discontinuity designs. Finally, we need to be able to communicate the results of our statistical analyses to interested consumers. Thus, track three will cover how we write-up the results of a statistical analysis. Students are expected to have taken PSCI 1801 or another course that covers multivariate regression analysis using R.