A culmination of the author's many years of consulting and teaching, Design and Analysis of Experiments with SAS provides practical guidance on the computer analysis of experimental data. It connects the objectives of research to the type of experimental design required, describes the actual process of creating the design and collecting the data, s
Table of Contents:
Introduction. Completely Randomized Designs with One Factor. Factorial Designs. Randomized Block Designs. Designs to Study Variances. Fractional Factorial Designs. Incomplete and Confounded Block Designs. Split-Plot Designs. Crossover and Repeated Measures Designs. Response Surface Designs. Mixture Experiments. Robust Parameter Design Experiments. Experimental Strategies for Increasing Knowledge. Bibliography. Index.
About the Author :
John Lawson is a professor in the Department of Statistics at Brigham Young University.
Review :
This book deserves to be seriously considered as an external reference for a beginning, one semester course in experimental design. It is a handy, reference showing how to get things done quickly using SAS. …
—Francis Giesbrecht, Biometrics, December 2011
The scope of the material coverage is one of the strengths of this book. The author brings a wealth of industrial, consulting, and teaching experience to the book, and adds his great writing style, which keeps the reader paying attention throughout the book. … one of the biggest strengths of the book is the breadth of coverage of topics and examples relevant to different disciplines. … The level of detail with respect to SAS codes in this text is excellent. … I would say that this book fills a gap in guiding the design and analyses of experiments using SAS, and would be well suited as a text for students in applied disciplines who had some linear algebra and statistics background. This book is also an outstanding reference for design and analysis of experiments using SAS. In this book I can find SAS codes for virtually all problems I had to solve in applications, and I will be reaching for this book time and time again in the future.
—Alla Sikorskii, The American Statistician, August 2011
The design and analysis of experiments is a fundamental part of statistics, and this book gives a comprehensive treatment of this broad topic. … this book focuses on linking concepts to practice. … The examples are taken from a range of areas, including pharmaceutical science and industrial manufacturing. … A companion website includes SAS code and data sets.
The book covers the basics that you would expect … . A wide variety of other topics are also covered, including split-plot designs, mixture experiments and robust-parameter design. Of particular interest to medical statisticians may be the chapter on crossover and repeated measurement studies …
The inclusion of so many exercises makes this book ideal for teaching. …
I think this book achieves its objectives. It is a comprehensive text on an important subject and it is sure to make designing and analysing experiments in SAS more straightforward. The inclusion of advanced topics and modern methods is a particular benefit in this regard.
—David Woods, Statistics in Medicine, 2011
The exposition throughout is first rate. The presentation and organization, the coverage of the topics, and the discussions of the examples are all excellent. If you are an SAS user needing help with experimental design, you will certainly profit from this text.
—International Statistical Review (2011), 79, 1
… The book’s strongest point is its wealth of practical examples from a wide range of fields, such as agriculture, industrial production, psychology, pharmacology etc. … the examples are very helpful for grasping the ideas behind applied experimentation. … a very useful addition to the library of anyone with an already strong understanding of linear models, some familiarity with SAS, and interest or experience in applied experimentation. … [also] useful for statistically skilled readers who want to use software other than SAS for design and analysis of experiments. Applied experimenters without a strong statistical background or at least interest will benefit from individual examples…
—Journal of Statistical Software, December 2010, Volume 37
