Statistics for Public Policy II

Catalog description

This course prepares students for both evaluating and conducting quantitative analysis of public policy using regression and regression-like techniques of statistical analysis. It does so by reviewing the logic of simple and multiple regression and the inferences that can be drawn from such analysis about public policy questions. The course then reviews the detection of violations of the assumptions of the regression model (specification error, heteroskedasticity, serial correlation, collinearity, nonlinearity, nonadditivity, and measurement error), their implications for valid inference, and their correction using extensions of basic regression analysis. The course will also examine regression-like techniques for nominal and ordinal dependent variables and their statistical evaluation. Throughout the course, the several regression analysis techniques will be examined through their application to typical public policy problems. The goal of the course is to enable students to become familiar with the elements of quantitative analysis of public policy using regression analysis, to enable them to evaluate such evidence bearing on public policy decisions, and to conduct regression analysis on public policy questions, all of which are essential for professional careers in public policy.