Aimed at practising biologists, especially graduate students and researchers in ecology, this revised and expanded 3rd edition continues to explore cause-effect relationships through a series of robust statistical methods. Every chapter has been updated, and two brand-new chapters cover statistical power, Akaike information criterion statistics and equivalent models, and piecewise structural equation modelling with implicit latent variables. A new R package (pwSEM) is included to assist with the latter. The book offers advanced coverage of essential topics, including d-separation tests and path analysis, and equips biologists with the tools needed to carry out analyses in the open-source R statistical environment. Writing in a conversational style that minimises technical jargon, Shipley offers an accessible text that assumes only a very basic knowledge of introductory statistics, incorporating real-world examples that allow readers to make connections between biological phenomena and the underlying statistical concepts.
Muu info
A guide for biologists describing statistical methods for testing causal hypotheses using R and their application to biological examples.
Preface;
1. Cause from correlation?;
2. From cause to correlation and
back;
3. Sewall Wright, path analysis and d-separation;
4. Covariance-based
SEM without explicit latent variables;
5. Statistical power, AIC statistics
and equivalent models;
6. Piecewise SEM with implicit latent variables;
7.
Modelling explicit latent variables in covariance-based SEM;
8. Multigroup
and multilevel structural equation models;
9. Exploratory structural
equations modelling;
10. A cheat sheet of important R functions; References;
Index.
Bill Shipley is Associate Professor within the Department of Biology at Université de Sherbrooke, Canada. He is the author of From Plant Traits to Vegetation Structure: Chance and Selection in the Assembly of Ecological Communities (2012) and continues to make significant contributions to statistical methodology in ecology.
