This book describes the basis, application, and interpretation of statistics, and presents a wide range of univariate and multivariate statistical methodology. The Second Edition retains the unique feature of being written from the users' perspective; it connects statistical models and methods to investigative questions and background information, and connects statistical results with interpretations in plain English. In keeping with this approach, methods are grouped by usage rather than by commonality of statistical methodology.
From the reviews of the second edition:
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
"The author covers a wide variety of applications. Each method is presented in sufficient depth to allow the reader to understand when the method should be used...I also appreciate the author’s tendency to refer reader’s to earlier sections or chapters for more foundational materials. In conclusion, I feel that this book would be a useful resource and a good text to use in training graduate-level statistical consultants on how to think about the application of statistics from a user’s perspective."
STATISTICAL METHODS IN MEDICAL RESEARCH
"The layout of the book is quite good...In all things pertaining to format, the author has been consistent...At the end of each example in the book, the report name and name of files holding data and statistical computing programs are provided. As such, even a student of statistics can benefit very much from this book. Although the book is not that big, it contains lots of valuable information. The author has really accomplished showing that statistics should be seen as an integral component of investigation and as a tool for conducting a well-informed post-mortem of data! The author has produces a masterpiece. His book can be used as a reference and as an undergraduate text for a course in statistics for those taking statistics as a service course as well as majors. As a reference manual, it is useful because it contains statistical methods from a wide spectrum starting with univariate statistical methods to multivariate statistical methods. As a student text, it is useful because the design is user friendly and calculus has been played down to some level without loss of direction...To crown it all, the author has achieved what he purposed in his mind. His immense experience in statistical consulting is something to go by. The book itself is a masterpiece and a ‘must have’ for every practicing statistician."
"Statistical practitioners will find this book useful in many ways: as the basis of a short course ... on introductory statistical methods ... as a reference for the practitioner’s own use; and as a teaching tool to explain statistical theory and methods to our customers. ... As a textbook, this book will be especially useful as an introduction for undergraduate students who will be taking additional upper-level courses in the experimental sciences or for graduate students with varied foundations in mathematical statistics." (Jay H. Goodman, Technometrics, Vol. 44 (3), 2002)
