Skip to main content
ARS Home » Pacific West Area » Maricopa, Arizona » U.S. Arid Land Agricultural Research Center » Plant Physiology and Genetics Research » Research » Publications at this Location » Publication #224813

Title: Book Review of Introduction to Mixed Modeling -- Beyond Regression and Analysis of Variance by N. Galwey

Author
item White, Jeffrey

Submitted to: Field Crops Research
Publication Type: Trade Journal
Publication Acceptance Date: 5/15/2008
Publication Date: 8/1/2008
Citation: White, J.W. 2008. Book Review of Introduction to Mixed Modeling -- Beyond Regression and Analysis of Variance by N. Galwey. Field Crops Research. 366 pp.

Interpretive Summary: Agricultural research usually involves numerical data that include “noise” due to effects of environment and measurement error. Statistical analyses are conducted in order to extract coherent and reliable information and of course, peer reviewed papers almost always include such analyses. Regression and analysis of variance have long been applied to agricultural experiments. In recent years, however, mixed modeling based on restricted maximum likelihood (REML) estimation has emerged as a superior alternative. The basic approach for REML has long been understood, but only with recent advances in computing power has REML become practicable for routine use by scientists. The book Introduction to Mixed Modelling – Beyond Regression and Analysis of Variance by N. Galwey (2007; John Wiley & Sons) is targeted for researchers who wish to learn about mixed models and REML. A key attraction of the book is that the writing style and structure are especially suitable for scientists who are familiar with regression and ANOVA but wish to analyze their data using mixed models. This book review concludes that the book is well written and especially suited to researchers who need to learn about the new techniques.

Technical Abstract: Regression and analysis of variance have long been applied to agricultural experiments. In recent years, however, mixed modeling based on restricted maximum likelihood (REML) estimation has emerged as a superior alternative. The basic approach for REML was developed decades ago, but only with recent advances in computing power has REML become practicable for routine use. The book Introduction to Mixed Modelling – Beyond Regression and Analysis of Variance by N. Galwey (2007; John Wiley & Sons) is targeted for researchers who wish to learn about mixed models and REML. The book’s ten chapters start with a review of regression and then gradually build up to increasingly complex applications of mixed models. Numerous examples are discussed using software code and listings from the commercial package GenStat and the open-source package R. Downloadable files are provided for the datasets, code and listings; SAS examples are also given although not discussed in the text. The writing style and structure are especially suitable for scientists who are familiar with regression and ANOVA. This book review concludes that the book is well written and especially suited to researchers who need to learn about the new techniques.