Customer Reviews

Analysing Ecological Data (Statistics for Biology and Health)

5 star:		(3)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

I searched for some introductory materials on mixed modeling and additive modeling and stumbled on this book by mistake. Lucky find!

For readers who need to use various types of models for practical problems in ecology or other sciences, including linear mixed models and additive models, this text is extremely useful and highly readable. The strength of this text is the manner in which the reader is guided through differing approaches to make good statistical decisions.

The authors present a wide range of approaches for analyzing ecological data using case studies. They approach their topic as would a practicing statistician - "first do this; if you find this result, here are your options and the advantages or disadvantages of each." It's almost like having a consultant at your fingertips. They offer step by step analyses of different ecological data sets using different types of regression (linear, logistic, etc.), additive models, tree models, multivariate models, factor analysis, time series, spatial analysis, and others. Mathematics and theory are covered lightly. Even though I am not in ecology, it was easy to apply these approaches to my non-ecological data.

Since every software is different, sometimes the output can be confusing. Consequently, it's very helpful that these authors include interpretations of output and plots. They point the reader to important features that lead to appropriate statistical decision-making. Highly recommended.

I bought this book after seeing it at my University library. It gives a good

intro to several types of analysis and is well written and approachable. I really like the step by step analysis guides given (with the examples) in the second half of the book.

I bought this book for self-study, thanks to the previous reviewers input here and in another websites. I had read only the first hundred pages, but you can appreciate the consultancy and teaching experience of the book's authors, with the smooth passage from a chapter to another in a progression of more complexity in the models: you begin with the data exploration techniques, mainly graphics, and pass to linear regression, GLM, GAM and mixed effects modelling, so far I had read.

The problem with the "smoothness" of that passage is that you can think the statistical analysis are too easy, and advance to much in the book and then you find in a difficult place of the book, because you don't let rest the acquired knowledge before proceed to the next chapters (I'm talkin to you, mixed modelling chapter!)

The book barely mention it in the first chapter, but the authors have their own page [...] where they offer the data and R code used in the book. Is highly recommended to follow the book with this data and R.