Multilevel Statistical Models (Kendall's Library of Statistics) - Hardcover

Goldstein, Harvey

Hardcover

ISBN 10: 0340595299 ISBN 13: 9780340595299

Publisher: Hodder Education Publishers, 1995

Synopsis

In the mid 1980s a number of researchers began to pursue systematic approaches to the statistical modelling and analysis of hierarchically structured data. Aitkin's early work on teaching styles and his subsequent classic work with Longford initiated a series of developments that by the early 1990s had resulted in a core of established techniques and software. The methods are now finding wide applications in education, epidemiology, geography, child development, and sociology. This new edition aims to integrate existing methodological developments, provide examples, and explain new developments, especially in discrete response data, times series models, random cross classifications, errors of measurement, missing data, and nonlinear models.

About this title

From the Publisher

Goldstein integrates existing statistical methodologies within a consistent terminology and avoids undue statistical complexity. He explains a number of new developments, particularly in the areas of discrete response data, random cross classifications, time series and nonlinear models, measurement errors and missing data. Numerous examples and diagrams illustrate the applications of techniques and methodological derivations are included in the appendices.

From the Back Cover

It is now generally recognised in many areas of the social, medical and other sciences that statistical data typically have complex hierarchical or multilevel structures in which individuals are grouped together in communities or institutions. This grouping affects their behaviour and multilevel modelling is now the accepted statistical technique for the analysis of this type of data. An understanding of these methods is vital for researchers in fields such as education, epidemiology, geography, child growth and social surveys, among others. This new edition brings the book fully up to date, explaining important new developments such as the use of Markov Chain Monte Carlo methods, bootstrapping and mulitvariate models. The book has been completely restructured for this third edition and extra space has been given to discussion of key issues such as missing data, measurement errors and multivariate models. Real-life examples are used throughout to illustrate clearly the theoretical concepts.

Edition details

• Publisher: Hodder Education Publishers
• Publication date: 1995
• ISBN 10: 0340595299
• ISBN 13: 9780340595299
• Binding: Hardcover
• Language: English
• Edition number: 2
• Number of pages: 192