 Alan Agresti has a reputation for being a fantastic textbook author. He is probably best known for categorical data analysis. Having read categorical data analysis and foundation of linear and generalized linear models, my opinion is that the latter is his crowning achievement. Now, this is a little like comparing apples and oranges because the topics are pretty different. I have so many positive feelings about this book. I love the Matrix approach to many of his derivations. I like the full notes section at the end of each of his chapters, taking time to explore some topics further and allowing myself to get lost and some of his notes paid dividends. It's a good way to become generally informed with statistics overall. I would say his end of chapter notes are comparable in quality to what would be found in computer age statistical inference. I think the exercises are useful. At the end of the book, he has a list of exercises for the reader to work out in R. I think that this is a good way to review the entire book. It is as if he is inviting the reader to have one last skim of the book and apply what was learned with data sets he provides. I highly recommend Foundations of Linear and Generalized Linear Models by Alan Agresti.