Focusing on mixed-effects models, this book offers a comprehensive guide to analysing experiments across diverse fields, including behavioural, agricultural, and medical sciences. The text opens with a traditional analysis of variance and then ranges from linear fixed-effects models to generalised linear mixed-effects models. It covers the most common experimental designs, such as factorial, hierarchical, between-subject, within-subject, cross-over, two-factor mixed, and split-plot designs, before studying analysis of covariance, models with group-specific error variances and models for repeated-measures analysis. Frequently drawing on real-life experiments, the book offers 69 examples and 134 exercises. Readers are supported with digital supplements, comprising the solutions to exercises, the datasets and R code and SAS code for all examples requiring software computation. This is an essential resource for students, practitioners conducting experiments and applied statisticians wishing to use mixed-effects models for the analysis of experiments.

‘This book is full of real examples and their data sets, and it is very clearly explained. The first chapter does a wonderful job of introducing the main ideas while using hardly any mathematical notation. Later chapters show the challenges of the different standard ways of writing the same model.’

R. A. Bailey - University of St Andrews

‘This book bridges the gap between statistical models for standard parallel-group designed experiments and models for more complex designs yielding clustered or correlated data. It is therefore an indispensable resource for any applied statistician faced with the analysis of clustered data from designed experiments.’

Geert Verbeke - KU Leuven

‘This is an excellent resource for researchers getting started with the design of experiments or observational studies and the analysis of real data arising from them, with lots of good examples and very lucid illustrations of key concepts, terms and methods.’

Hans-Peter Piepho - University of Hohenheim

‘Forkman clearly explains how experimental design should guide mixed-effects modeling, starting from the practical distinction between fixed treatment effects and random design effects. Through carefully developed examples spanning standard and complex designs, the book helps students and applied researchers specify models, check assumptions, and interpret treatment comparisons.’

Paulo Canas Rodrigues - Federal University of Bahia