Conventional statistical methods have a very serious flaw. They routinely miss differences among groups or associations among variables that are detected by more modern techniques - even under very small departures from normality. Hundreds of journal articles have described the reasons standard techniques can be unsatisfactory, but simple, intuitive explanations are generally unavailable. Improved methods have been derived, but they are far from obvious or intuitive based on the training most researchers receive. Situations arise where even highly nonsignificant results become significant when analyzed with more modern methods. Without assuming any prior training in statistics, Part I of this book describes basic statistical principles from a point of view that makes their shortcomings intuitive and easy to understand. The emphasis is on verbal and graphical descriptions of concepts. Part II describes modern methods that address the problems covered in Part I. Using data from actual studies, many examples are included to illustrate the practical problems with conventional procedures and how more modern methods can make a substantial difference in the conclusions reached in many areas of statistical research. Rand Wilcox is a professor of psychology at the University of Southern California. He is a fellow of the Royal Statistical Society and the American Psychological Society. Dr. Wilcox currently serves as an associate editor of Computational Statistics & Data Analysis and Psychometrika. He has published over 165 articles in a wide range of statistical journals and he is the author of three other books on statistics.|¿The volume is a jewel of direct explanations and information necessary for a good understanding of analysis of data, aimed at ordinary researchers who must try to present reasonable interpretable accounts of their data or judge when to abandon a particular strategy..."|- Perceptual and Motor Skills, 2002
Summary: elementary but well-written with a modern touch using robustness Rating: 5
This is a marvelous introductory statistics book. In the first five chapters Wilcox presents the classical statistical methods with simplicity and great care. Emphasis is properly placed on the assumptions of the methods and the fact that many times important assumptions are violated in practice. This is all done in the first five chapters. Then in chapters 6 - 12 he provides alternative modern methods that can work better than the classical ones when the assumptions fail. Chapter 6 is on bootstrap. Chapter 8 covers robust measures of location. Chapter 9 covers statistical inference based on robust measures of location. Chapter 10 deals with nonparametric and robust measures of association. Chapter 11 covers robust estimates of location and Chapter 12 "Alternative Strategies" includes rank test, permutation tests and multiple comparison issues.
At the beginning of Wilcox's book there are three precious quotes that you should appreciate. 1. "If at first the idea is not absurd, then there is not hope for it" Albert Einstein. 2. "Everyone believes in the [normal] law of errors, the experimenters because they think it is a mathematical theorem, the mathematicians because they think it is an experimental fact." Henri Poincare. 3. "Each generation that discovers something from its experience must pass that on, but it must pass that on with a delicate balance of respect and disrespect, so that the race ... does not inflict its errors too rigidly on its youth, but it does pass on the accumulated wisdom plus the wisdom that it may not be wisdom." Richard Feynman.
These quotes were well chosen to fit the theme of the book and represent the wisdom of three great men from the 19th and 20th Centuries.