Methods for International Relations (Undergraduate)
This course is designed to introduce students to the style of analytic thinking required for research in the social sciences; the concepts and procedures used in the conduct of empirical political science research; and the use of computers for analysis of quantitative social science data. In short, this course teaches a set of skills that are essential for both understanding the research you will encounter in classes, and being able to produce high-quality original research of your own. We will cover the principles of the scientific method as applied to the study of international relations, emphasizing an approach to understanding politics that uses generalizing theory and testable hypotheses. The first part of the course addresses critical issues in the design of empirical tests of theories about political phenomena, including sample selection, concept definition and measurement, and types of data collection. The remainder of the course focuses on a variety of techniques for analyzing quantitative political data, from simple descriptive statistics and graphs, to tabular data analysis, tests of bivariate association, multiple linear regression models, and models for measurement. This is an applied course that draws on real political applications and research examples from international relations. The course strongly favours practice (e.g., choice of appropriate statistical procedure, diagnostics, interpretation) over theory (mathematical derivations and proofs). You do not need any more math background than high school algebra for this course, and you will not be expected or required to memorize any mathematical formulas. Instead, youwill learn real, practical skills in using the free statistical software R, identifying which approaches to takefor different kinds of problems, and interpreting the sometimes conflicting and confusing results reported in academic journals.
Week 1: What is science? Is social science really science? Where is science going?
Research questions, theories, concepts, and hypotheses
Research design: non-experimental large-N designs
Descriptive statistics, visualizing data, measurement validity and reliability
Confidence intervals, statistical inference, and hypothesis testing
Confidence intervals, statistical inference, and simulations
Causal Inference Using Regression on the Treatment Variable
Linear modeling: ordinary least squares
Interpreting and communicating regression estimates
The multiple linear regression model, model fit, and replication
The multiple linear regression model, missing data, and selection
Using Models for Measurement
More Data Visualization and Presenting Results and Findings