Multilevel Statistical Models

This text describes techniques developed in the early 1990s which introduce a more systematic approach to the statistical modelling and analysis of hierarchically structured data. The book discusses, and expands on, the core set of established techniques, experience and software packages, which are becoming more commonly used in areas such as education, epidemiology, geography, child growth and household surveys. The second edition aims to integrate existing methodological developments, within a consistent terminology and notation, and avoids undue statistical complexity. Methodological derivations are provided in the appendices and examples and diagrams are used to illustrate the applications of techniques. It explains a number of new developments, especially in the areas of discrete response data, time series models, random cross classifications, errors of measurement, missing data and nonlinear models. This book is suitable for postgraduate courses and as a general reference tool and as such is extensively referenced.