Multilevel modeling using R (Q2875816)

The book under review presents on a total of 216 pages an introduction into multilevel modelling using \texttt{R} package and is organized into 9 chapters. It addresses to readers with little statistical and \texttt{R} package background knowledge. Each chapter of the book ends with a brief summary.NEWLINENEWLINEChapter 1 introduces the basic concepts of simple linear regression models as well as the implementation in \texttt{R}. Chapter 2 is devoted to methods handling multilevel data structures (intraclass correlation, random intercept, random slope, centering) and basics of maximum likelihood parameter estimation (restricted maximum likelihood, two-level and three-level multilevel linear models). Fitting multilevel models for different situations is illustrated in the following three chapters using the two \texttt{R} packages \texttt{lme4} and \texttt{nlme}: fitting two-level models in Chapter 3, fitting models with three and more levels in Chapter 4 and longitudinal data analysis in Chapter 5. After a brief description of the graphical presentation of multilevel data in Chapter 6, Chapter 7 concentrates on basic aspects of general linear models (logistic regression for dichotomous or ordinal outcome, models for count data). Chapter 8 focuses on the extension of the general linear model to multilevel models. The last chapter deals with Bayesian multilevel modelling. The volume ends with an appendix on basic introduction to \texttt{R} code.NEWLINENEWLINEIn summary, the book under review is written for readers with little background in statistics and \texttt{R} rather seeking a first orientation than an in-depth understanding in methodology.