Applications of Regression for Categorical Outcomes Using R

This book covers the main models within the GLM (i.e., logistic, Poisson, negative binomial, ordinal, and multinomial). For each model, estimations, interpretations, model fit, diagnostics, and how to convey results graphically are provided. There is a focus on graphic displays of results as these are a core strength of using R for statistical analysis. Many in the social sciences are transitioning away from using Stata, SPSS and SAS, to using R, and this book uses statistical models which are relevant to the social sciences. Social Science Applications of Regression for Categorical Outcomes Using R will be useful for graduate students in the social sciences who are looking to expand their statistical knowledge, and for Quantitative social scientists due to it’s ability to act as a practitioners guide. Key Features: Applied- in the sense that we will provide code that others can easily adapt Flexible- R is basically just a fancy calculator. Our programs will enable users to derive quantities that they can use in their work Timely- many in the social sciences are currently transitioning to R or are learning it now. Our book will be a useful resource Versatile- we will write functions into an R package that can be applied to all of the regression models we will cover in the book Aesthetically pleasing- one advantage of R relative to other software packages is that graphs are fully customizable. We will leverage this feature to yield high-end graphical displays of results Affordability- R is free. R packages are free. There is no need to purchase site licenses or updates.

1. Introduction 2. Introduction to R Studio and Packages 3. Overview of OLS Regression and Introduction to the General Linear Model 4. Describing Categorical Variables and Some Useful Tests of Association 5. Regression for Binary Outcomes 6. Regression for Binary Outcomes – Moderation and Squared Terms 7. Regression for Ordinal Outcomes 8. Regression for Nominal Outcomes 9. Regression for Count Outcomes 10. Additional Outcome Types 11. Special Topics: Comparing Between Models and Missing Data

David Melamed is a Professor of Sociology and Translational Data Analytics at The Ohio State University. His research interests include the emergence of stratification, cooperation and segregation in dynamical systems, and statistics and methodology. Since 2019 he has been co-Editor of Sociological Methodology. Long Doan is an Associate Professor of Sociology at the University of Maryland, College Park. His research examines how various social psychological processes like identity, intergroup competition, and bias help to explain the emergence and persistence of social stratification. He focuses on inequalities based on sexuality, gender, and race.