Summary and Info
I got this book while working on an article that involved a hierarchical model with a binary dependent variable - after poking through Radenbush/Bryk and a variety of other texts that left me frustrated. Not only did this book teach me how to properly specify and estimate the model in R, I also learned a lot about interpretation and graphical means of presenting results. I don't think I've read another book that so effectively combines theoretical and practical information, while also being a relatively smooth read - the examples are clear and interesting! In addition to the extensive treatment of hierarchical models, Gelman and Hill also cover non-hierarchical OLS and ML models, plus a variety of other key stats topics. My only quibble is that the accompanying R code on Gelman's website isn't complete - but the fact that they have sample code available at all puts this far beyond most stats books. I wish I had had this book in grad school and look forward to referring to it for years to come.
More About the Author
Andrew Gelman (born February 11, 1965) is an American statistician, professor of statistics and political science, and director of the Applied Statistics Center at Columbia University.