Analysis of Ordinal Categorical Data

Statistical science now has its first coordinated manual of methods for analyzing ordered categorical data. The last decade has seen a dramatic rise in the development of methods for cross-classification tables having ordered categories. Volumes of material concerning specialized methods for ordinal variables have appeared scattered throughout the technical literature, but Analysis of Ordinal Categorical Data is the first book to present professional statisticians with a coherent view of them. These methods are being more and more frequently applied in such diverse fields as sociology, public health, ecology, marketing, and pharmacy. Researchers are finding that ordinal methods make possible simpler description of data, and also permit more powerful inferences about crucial questions of association linkages. This book holds the key to such applications, discussing specialized models that, unlike standard methods for nominal categorical data, efficiently use the information on ordering. There is no more organized and contemporary treatment of the various ways to analyze ordered categorical data. The book begins with an introduction to basic descriptive and inferential methods for categorical data, and then gives thorough coverage of the most current developments, such as the loglinear and logit models for ordinal data. The book places special emphasis on interpretation and application of methods, and it contains an integrated comparison of the available strategies for analyzing ordinal data. This is a case study work, with illuminating examples taken from across the wide spectrum of ordinal categorical applications. It features data sets concerning the influence of racial characteristics on the death penalty verdict for homicides; political ideology and party affiliation; the implications and effects of an operation for ulcers; and more. Analysis of Ordinal Categorical Data also explains in full how to implement methods by using existing statistical computer packages such as SAS, SPSS^x, and GLIM. This completes an invaluable resource that serious professional statisticians and researchers will need for future analysis in the social, political, behavioral, and biomedical fields.