Linear Regression Models

This book includes chapters on specifying the correct linear regression model, adjusting for measurement error, understanding the effects of influential observations, and using multilevel data.

John P. Hoffmann is a professor of sociology at Brigham Young University. He holds a PhD in Criminology from the State University of New York at Albany and a Masters of Public Health (MPH) from Emory University. He has worked at the U.S. Centers for Disease Control and Prevention (CDC) and the National Opinion Research Center (NORC) of the University of Chicago; and taught at Hokkaido University and the University of South Carolina. Hoffmann is the author of more than 100 journal articles and book chapters and 10 books on applied statistics, criminology, and the sociology of religion.

1. Introduction 2. Review of Elementary Statistical Concepts  3. Simple Linear Regression Models  4. Multiple Linear Regression Models  5. The ANOVA Table and Goodness-of-Fit Statistics  6. Comparing Linear Regression Models  7. Indicator Variables in Linear Regression Models  8. Independence  9. Homoscedasticity  10. Collinearity and Multicollinearity  11. Normality, Linearity, and Interaction Effects  12. Model Specification  13. Measurement Errors  14. Influential Observations: Leverage Points and Outliers  15. Multilevel Linear Regression Models  16. A Brief Introduction to Logistic Regression  17. Conclusions  Appendix A: Data Management  Appendix B: Using Simulations to Examine Assumptions of Linear Regression Models  Appendix C: Formulas  Appendix C:  User-Written R Packages Employed in Examples