Graphical methods for the design of experiments (Q1306940)

1. General remarks. This is a good book because it brings about novel approaches to the use of experimental designs. The author says that ``many texts on the design of experiments focus on the analysis of experimental data, not on the creation of the design''. In general I do agree with the author, except to some of my own works. Here and there the author quotes our book [the reviewer et al., Factorial designs. (1981; Zbl 0593.62074)] to show that we were too theoretical and not cook book enough. But we did present many practical applications. As to graphic-geometrical viewpoints, the author should have read my papers, as indicated, i.e., Commun. Stat. 3, 959-968 (1974; Zbl 0294.62102); Am. Math. Monthly 86, 49 (1979; Zbl 0392.05007). 2. Specific comments. The contents of the book are as follows: 1. Introduction; 2. Planning for a designed experiment; 3. Design-plots for factorial and fractional-factorial designs; 4. Designing experiments graphically; 5. Assessing experimental designs; 6. Presenting results graphically. As one can see, the book does not cover the whole gamut of experimental designs. Clearly it caters to the author's experiences in consulting where a subclass of experimental designs were needed to solve specific problems in practice. Yet, the topics considered by the author are quite good and probably an eye-opener to many researchers in experimental designs, especially those in the theory of these topics. 3. Who can read the book. To quote the author: ``The text assumes a basic knowledge of statistics and matrices. Many of the graphical techniques are accessible without any knowledge of statistical models, and require only some familiarity with the plotting of functions and with the concept of functions and with the concept of projection from elementary mechanical drawings.'' It follows that the material can be used by a variety of audiences from factory personnel to graduate students in statistics. From reading the book I agree that this is so. 4. Concluding remarks. The author should be congratulated in bringing theory and applications of experimental designs to the attention of scientists, experimenters, and statisticians. His final version of the book would benefit collaboration with the reviewer.