Applied Nonparametric Statistical Methods

This new edition of Peter Sprent's highly successful introduction to nonparametric methods retains many features acclaimed by reviewers of the first edition. Basics are simply and carefully explained using many realistic but simple examples. The text has been completely revised to highlight the way modern computer software enables us to carry out exact tests based on permutation distributions where one had previously to resort to asymptotic results in situations where these are now known to be misleading. New methods are introduced in this edition and relationships between inference procedures that may at first sight appear to be unrelated are clearly demonstrated. More attention is given to the analysis of censored data and the sections on analysis of counts and categorical data have been greatly extended. This additional material is relevant to applications in fields as diverse as medicine, the social sciences, industry and market research. This book should be of interest to research workers in industry, pharmaceutical firms, and all who use basic statistical methods. It should also be a useful text for undergraduate students taking university and polytechnic courses.