Block designs: a randomization approach. Vol. 1: Analysis (Q5926509)

1. Before I present an abbreviated review of this volume of a proposed two-volume book by the authors, it was a great surprise to me that the authors did not mention the last paper of a pioneer in this subject. This is the paper by \textit{O. Kempthorne} in the book, Ch.E. McCulloch et al. (eds.), Statistical Design: Theory and Practice, Procedings of a Conference in Honor of W. T. Federer, Cornell University, Ithaca, 43-88 (1986). The authors should give me credit in their book for pointing this out. 2. The first volume deals with analysis aspects and consists of five chapters, while the second volume is concerned with design aspects and consists of continuing five chapters and some appendices. Instead of listing the chapters as usual, I will present the authors' words in their preface: ``The book is composed of two volumes. Volume I is devoted to the analysis of experiments in block designs. Volume II concerns the constructions of block designs. This order has been determined by the fact that one cannot understand the advantages of that or other designs without knowing the model on which the analysis is based. The outline of the two volumes of the monograph is presented in the introduction chapter (Chapter 1). The reader is advised to read the whole chapter carefully, as it gives the main ideas on which the whole exposition relies. Chapter 2 comprises the main terminological notions and concepts used throughout the monograph. Readers more familiar with the subject can, perhaps, skip the first two sections of this chapter, but it is advisable to read the remaining two, to which references are frequently made in other chapters. The main part of Volume I is Chapter 3, presenting a randomization model for a general block design and its implications for the analysis. It offers ideas and methods essential for any further study of block designs within the randomization approach. Implications for considering properties of these designs and their classifications are discussed in Chapter 4. Those who are interested in extensions of the results presented in Chapter 3 to more advanced design structures are advised to read Chapter 5, devoted to nested block designs, to resolvable block designs in particular. It is hoped that readers finding the theory presented in Volume I interesting will also like to read Volume II, now under final preparation. The book is aimed at an advanced audience, including students at the post-graduate level and research workers interested in designing and analyzing experiments with the full understanding of the principles. This monograph does not pretend to give an exhaustive exposition of the theory of experimental design within the randomization approach, but it is hoped that at least it gives some perspectives for further research on the subject. In the course of preparing the book, the help of several colleagues has been appreciated. In particular, thanks are due to Dr. Pawel Krajewski for his help in performing computations, and to Dr. Idzi Siatkowski and Dr. Takashi Seo for their assistance in the technical preparation of the manuscript. Last but not least, the authors are most grateful to Dr. John Kimmel, the editor, and to several reviewers for their many instructive comments and suggestions, and inspiring encouragement.'' 3. I believe that both volumes will be basically combined as a mathematical statistics book dealing with randomization aspects of the analysis and design of block designs. The first volume is terse and few people concerned with practicalities will read this and the subsequent second volume. In fact, the first volume reminds me of a Ph.D. thesis type of approach already. In fact, more of the type of writing as done by Kempthorne as mentioned above, should be incorporated to motivate theoretical and practical workers in design and analysis of randomized experiments. I wish to see what the second volume will look like before I form a complete opinion of the proposed book. So far, I wish the authors good luck in terms of acceptance by the statistical community.