This book is intended as an introduction to multilevel analysis for students and applied
researchers. The term ‘multilevel’ refers to a hierarchical or nested data structure,
usually subjects within organizational groups, but the nesting may also consist of
repeated measures within subjects, or respondents within clusters, as in cluster sampling. The expression multilevel modelis used as a generic term for all models for nested
data. Multilevel analysisis used to examine relations between variables measured at
different levels of the multilevel data structure. This book presents two types of multilevel models in detail: the multilevel regression model and the multilevel structural
equation model.
Multilevel modeling used to be only for specialists. However, in the past decade,
multilevel analysis software has become available that is both powerful and accessible.
In addition, several books have been published, including the first edition of this book.
There is a continuing surge of interest in multilevel analysis, as evidenced by the
appearance of several reviews and monographs, applications in different fields ranging
from psychology and sociology, to education and medicine, and a thriving Internet
discussion list with more than 1400 subscribers. The view of ‘multilevel analysis’ applying to individuals nested within groups has changed to a view that multilevel models
and analysis software offer a very flexible way to model complex data. Thus, multilevel
modeling has contributed to the analysis of traditional individuals within groups data,
repeated measures and longitudinal data, sociometric modeling, twin studies, metaanalysis and analysis of cluster randomized trials.