Modern Regression Analysis

Excellent course that covers a lot of material in the two short weeks of the Methods School. — participant from Australia

This course provides the tools needed for the quantitative testing of theories. Participants will learn the logic and central assumptions underlying the multivariate ordinary least squares regression model, but the course also covers such advanced topics as the analysis of time series and pooled time series data and of limited dependent variable models. The emphasis is on making the transitions between theory, model specification, and result presentation as seamless as possible. Participants will get hands-on experience with applying the techniques covered in this course to their own datasets.

This course provides the foundation for more advanced quantitative methods courses, such as Panel Data Analysis. Participants without a minimal background in statistics should consider the introduction to Applied Data Analysis instead.

Dates

This two-week, 35-hour course runs Monday-Friday, July 1-12, 2024. The course is scheduled for 9:00 am-12:30 pm.

Classroom Location

Faculty of Arts and Social Science, AS1 02-09

Image preview

Instructor

Guy D. Whitten, Texas A&M University
Katsunori Seki, Nagoya University

Detailed Description

This modern regression analysis course teaches participants the tools that they need to test their theories and produce presentations of their results for publication in the top journals in Political Science, International Relations, Public Policy, Economics, and other social science disciplines. Participants are strongly encouraged to bring their own data sets so that they can get the hands-on experience provided by this course and have the opportunity to apply the covered techniques directly to their own research.

The course is divided into three parts. The first part involves a thorough presentation of the logic and the central assumptions underlying the multivariate ordinary least squares regression model. The second part focuses on issues that researchers typically encounter as they attempt to test their theories in a regression framework. The third part focuses on application and extension of the concepts covered in the first two parts to time series data, pooled time series data, and models of limited dependent variables.

This is a hands-on course, meaning that a major goal is to have participants learn about techniques by putting them to work with statistical software. To facilitate this, we combine lectures on each topic with study group and lab sessions. In these sessions participants have the option of working with their own data or working with data provided by the instructor. The main statistical software programs used are Stata and R. Prior experience with these programs will be helpful, but is not assumed or required.

Prerequisites

The course requires only a minimal background in statistics and mathematics. However, participants without any prior knowledge should consider taking a more basic of quantitative methods course instead (cf. Applied Data Analysis).

Requirements

Participants are expected to have access to an internet-connected computer. Access to data, temporary licenses for the course software, and installation support will be provided by the Methods School.

Core Readings

Kellstedt, Paul M., and Guy D. Whitten. 2018. The Fundamentals of Political Science Research. 3^rd edition. New York, NY: Cambridge University Press.