Mixed Effects Models and Extensions in Ecology with R BYWalker

by Walker

Building on the successful Analysing Ecological Data (2007) by Zuur, Ieno and Smith, the authors now provide an expanded introduction to using regression and its extensions in analysing ecological data. As with the earlier book, real data sets from postgraduate ecological studies or research projects are used throughout. The first part of the book is a largely non-mathematical introduction to linear mixed effects modelling, GLM and GAM, zero inflated models, GEE, GLMM and GAMM. The second part provides ten case studies that range from koalas to deep sea research. These chapters provide an invaluable insight into analysing complex ecological datasets, including comparisons of different approaches to the same problem. By matching ecological questions and data structure to a case study, these chapters provide an excellent starting point to analysing your own data. Data and R code from all chapters are available from www.highstat.com.

Genres: Science, Nonfiction, Textbooks, Biology

Hardcover

First published March 12, 2009

Reviews

[Tomek · Rating 3 out of 5]

Although I have yet to find such a comprehensive overview of modeling ecological data in R (which is precisely what I needed in my research), I found the presentation wanting. While the authors sometimes clearly explain their thought processes and logic behind making certain decisions, others are left to the reader to divine or are explained later, which can cause a good deal of frustration up to that point. Also, while it is advertised as "non-mathematical" it certainly assumes a good deal of familiarity with statistical notation, which I feel many applied ecologists do not. While, I do think this is a useful resource for ecologists looking to do higher level statistical analysis, but I feel it would be best paired with a course to clear up any questions and uncertainties.

[Karina · Rating 5 out of 5]

This is definitely not a book for beginners, but is an extremely valuable resource for postgraduates or research scientists attempting to analyse biological data that do not conform to the basic assumptions of the simple statistical analyses that are usually taught during undergraduate degrees. Most real biological datasets do not meet these assumptions and researchers are left floundering for solutions. This book covers many of the typical issues and provides clear r code and explanations to attempt their application to real data. It is a dense read and reliance on the index to skip to relevant sections is required, but all of the information is there or suggestions for further reading provided. An incredibly valuable piece of work!

[Jenna Black · Rating 5 out of 5]

i'm pretty sure this book saved my honours project... bless ya

[Sophie · Rating 5 out of 5]

they don't tell you that when you go into environmental science you are actually about to be balls deep in statistics and code. so like fyi.

[Barbara · Rating 3 out of 5]

This is a pretty good text if you are already familiar and comfortable with regression modelling, including GLM and GAM, probability and error distributions and heteroscedastic data. I first read this book when some of these things were new to me, and found it very confusing. Revisiting it now, it seems much clearer and has a lot of useful content, but I would not recommend it for beginners.

[Sofia · Rating 4 out of 5]

This book really gives a great insight for working with mixed effects models in R! Although, I would like to see more insight for using the lme4 package over the nlme package.