# An R Companion to Applied Regression

- John Fox - McMaster University, Canada
- Sanford Weisberg - University of Minnesota, USA

**Other Titles in:**

Regression & Correlation | Research Methods in Psychology | Sociological Research Methods

**An R Companion to Applied Regression**is a broad introduction to the R statistical computing environment in the context of applied regression analysis. John Fox and Sanford Weisberg provide a step-by-step guide to using the free statistical software R, an emphasis on integrating statistical computing in R with the practice of data analysis, coverage of generalized linear models, and substantial web-based support materials.

The **Third Edition **has been reorganized and includes a new chapter on mixed-effects models, new and updated data sets, and a de-emphasis on statistical programming, while retaining a general introduction to basic R programming. The authors have substantially updated both the car and effects packages for R for this edition, introducing additional capabilities and making the software more consistent and easier to use. They also advocate an everyday data-analysis workflow that encourages reproducible research. To this end, they provide coverage of RStudio, an interactive development environment for R that allows readers to organize and document their work in a simple and intuitive fashion, and then easily share their results with others. Also included is coverage of R Markdown, showing how to create documents that mix R commands with explanatory text.

Projects in RStudio |

R Basics |

Fixing Errors and Getting Help |

Organizing Your Work in R and RStudio |

An Extended Illustration |

R Functions for Basic Statistics |

Generic Functions and Their Methods* |

Data Input |

Managing Data |

Working With Data Frames |

Matrices, Arrays, and Lists |

Dates and Times |

Character Data |

Large Data Sets in R* |

Complementary Reading and References |

Examining Distributions |

Examining Relationships |

Examining Multivariate Data |

Transforming Data |

Point Labeling and Identication |

Scatterplot Smoothing |

Complementary Reading and References |

The Linear Model |

Linear Least-Squares Regression |

Predictor Effect Plots |

Polynomial Regression and Regression Splines |

Factors in Linear Models |

Linear Models with Interactions |

More on Factors |

Too Many Regressors* |

The Arguments of the lm Function |

Complementary Reading and References |

Coefficient Standard Errors |

Confidence Intervals |

Testing Hypotheses About Regression Coefficients |

Complementary Reading and References |

The Structure of GLMs |

The glm() Function in R |

GLMs for Binary-Response Data |

Binomial Data |

Poisson GLMs for Count Data |

Loglinear Models for Contingency Tables |

Multinomial Response Data |

Nested Dichotomies |

The Proportional-Odds Model |

Extensions |

Arguments to glm() |

Fitting GLMs by Iterated Weighted Least-Squares* |

Complementary Reading and References |

Background: The Linear Model Revisited |

Linear Mixed-Effects Models |

Generalized Linear Mixed Models |

Complementary Reading |

Residuals |

Basic Diagnostic Plots |

Unusual Data |

Transformations After Fitting a Regression Model |

Non-Constant Error Variance |

Diagnostics for Generalized Linear Models |

Diagnostics for Mixed-Effects Models |

Collinearity and Variance-Inflation Factors |

Additional Regression Diagnostics |

Complementary Reading and References |

A General Approach to R Graphics |

Putting It Together: Local Linear Regression |

Other R Graphics Packages |

Complementary Reading and References |

Why Learn to Program in R? |

Defining Functions: Preliminary Examples |

Working With Matrices* |

Conditionals, Loops, and Recursion |

Avoiding Loops |

Optimization Problems* |

Monte-Carlo Simulations* |

Debugging R Code* |

Object-Oriented Programming in R* |

Writing Statistical-Modeling Functions in R* |

Organizing Code for R Functions |

Complementary Reading and References |

### Supplements

An **accompanying website** for the book found at study.sagepub.com/RCompanion provides:

- R scripts for examples by chapter
- Data files used in the book
- The car package (Companion to Applied Regression), an accompanying software for regression diagnostics and other regression-related tasks
- Other resources to help students get the most out of the text

“**An R Companion to Applied Regression **continues to provide the most comprehensive and user-friendly guide to estimating, interpreting, and presenting results from regression models in R.”

**University of California, Davis**

“This is the best book I’ve read for teaching the modern practice of regression. By going deeply into both R and applied regression, it manages to use each topic to motivate and illustrate the other. The whole is much greater than sum of the parts because each thread so effectively reinforces the other. There are many nice surprises in this new edition. R Studio and markdown are used to encourage a reproducible workflow. There’s an excellent and accessible chapter on mixed and longitudinal data that expands the reach of regression methods to the much more complex data structures typical of current practice. Like its predecessors, this edition is a model of clear, thoughtful exposition. It’s an outstanding contribution to the teaching and practice of regression.”

**York University**

“This is an impressive update to a book I have long admired. The authors have brought the description of how to do data analysis and plots of Applied Regression related data to a modern and more comprehensive level.”

**York University**