Stereology for Statisticians sets out the principles of stereology from a statistical viewpoint, focusing on both basic theory and practical implications. This book discusses ways to effectively communicate statistical issues to clients, draws attention to common methodological errors, and provides references to essential literature. The first full text on design-based stereology, it opens with a review of classical and modern stereology, followed by a treatment of mathematical foundations. It then presents core techniques. The final chapters discuss implementing techniques in practical sampling designs, summarize understanding of the variance of stereological estimators, and describe open problems for further research.
Overall, I enjoyed reading this book very much. I like the authors' intuitive explanations and careful insights into why one path or approach is taken over another. And most of all, I am impressed with the wealth of information on the subject and the plethora of references listed at the end of the book. In conclusion, I feel that Stereology for Statisticians is an enjoyable addition and it certainly contains many interesting methods and examples from various important application areas. -Stergios B. Fotopoulos, Technometrics, August 2010 Overall the level of the book is nicely balanced between explaining ideas with reference to the main issues in scientific use of stereology and giving clear mathematical derivations of the results. The diagrams are excellent, and the reader to visualize how three-dimensional structures will appear in two-dimensional slices . . . this book succeeds in what it aims to do: to provide a comprehensive introduction and reference to stereology for statisticians. It will be a valuable addition to this reviewer's bookshelf. -Graham Horgan, Rowett Research Institute, Biometrics, December 2006 Scientists at large, with or without a direct interest in stereology, should own a copy of this book, (and they should also encourage their students to get one), if only because sampling ideas pervade modern science. -Luis M. Cruz-Orive, University of Cantabria, Statistics in Medicine, 2005, Vol. 24