Contemporary Bayesian and frequentist statistical research methods for natural resource scientists /

This work aims to provide readers (with background knowledge of introductory statistics) the basic tools of data analysis and inference in an easy-to-understand style. This is an excellent book presenting difficult statistical ideals by using data obtained from real-life situations. Stauffer (Humboldt State Univ.), an experienced forestry and wildlife management statistician, is aware of the frequent misuse of statistics in the natural resource sciences, and he tries to rectify it here. The book has eight chapters in all. Chapter 1 introduces three case studies illustrating the need for newer contemporary methods of statistical analysis and includes a brief introduction to the frequentist software S-Plus and R; chapter 8 summarizes all the techniques in chapters 2-7. Chapters 2-4 introduce the basic concepts of Bayesian statistical analysis and inference as an alternative approach to the traditional frequentist approach. Chapters 5-7 include alternative strategies to model selection and inference; generalized linear models, concentrating mostly on logistic regression models; and mixed-effects models in R and S-Plus. The book includes numerous illustrations that show the strengths and limitations of the frequentist and Bayesian approaches. Valuable for advanced undergraduate or beginning graduate-level coursework, and as a reference book for researchers. Summing Up: Recommended. Upper-division undergraduates through researchers/faculty. D. V. Chopra Wichita State University