Design and Analysis of Experiments (Springer Texts in Statistics) - Softcover

Dean, Angela M.; Voss, Daniel

Softcover

ISBN 10: 1475772920 ISBN 13: 9781475772920

Publisher: Springer, 2013

Synopsis

Experimental design is essential for statistical work in many fields including medicine, agriculture, and engineering. This book will serve as a modern and comprehensive reference. The book includes sample SAS programs.

About this title

Review

This is a readable book presenting the basic concepts, principles, and techniques of design and analysis of experiments. Written with a view to making it accessible to a wide audience, the authors make concerted efforts to avoid using calculus and linear algebra and, wherever needed, a low mathematical level is used for presentation. Rather than performing exploratory data analysis, the concentration here is on the use of prespecified models and preplanned analyses. Model assumptions are clearly stated, and checked through the use of residual plots rather than formal tests. All analyses are presented by using standard linear models under the assumption of normality. It is the experimentwise control of the error rate and confidence levels on which the presentation is focused as opposed to individual error rates and confidence levels. The popular "Taguchi techniques", used extensively in an industrial set-up, are included and appear throughout several chapters.

The book contains enough material for an instructor to offer a course ranging from one semester to one year. An attractive feature of this book is the inclusion of numerous real experiments which were either run by students or extracted from published articles---thus bringing home to students the practical utility of statistical designs. The authors have done a commendable job in presenting, explaining, and elucidating the fundamental concepts of design and analysis of experiments through illustrative examples. A carefully selected set of exercises is provided at the end of each chapter for students to test their understanding of the material.

--Mathematical Reviews

From the Back Cover

This textbook takes a strategic approach to the broad-reaching subject of experimental design by identifying the objectives behind an experiment and teaching practical considerations that govern design and implementation, concepts that serve as the basis for the analytical techniques covered. Rather than a collection of miscellaneous approaches, chapters build on the planning, running, and analyzing of simple experiments in an approach that results from decades of teaching the subject. In most experiments, the procedures can be reproduced by readers, thus giving them a broad exposure to experiments that are simple enough to be followed through their entire course. Outlines of student and published experiments appear throughout the text and as exercises at the end of the chapters. The authors develop the theory of estimable functions and analysis of variance with detail, but at a mathematical level that is simultaneously approachable. Throughout the book, statistical aspects of analysis complement practical aspects of design. 
This new, second edition includes

• an additional chapter on computer experiments
• additional "Using R” sections at the end of each chapter to illustrate R code and output 
• updated output for all SAS programs and use of SAS Proc Mixed
• new material on screening experiments and analysis of mixed models

Angela Dean, PhD, is Professor Emeritus of Statistics and a member of the Emeritus Academy at The Ohio State University, Columbus, Ohio. She is a fellow of the American Statistical Association and the Institute of Mathematical Statistics.  Her research interests include design of screening and computer experiments.
Daniel Voss, PhD, is Professor Emeritus of Mathematics and Statistics and former Interim Dean of the College of Science and Mathematics at Wright State University, Dayton, Ohio.  His research interests include the analysis of saturated fractional factorial experiments, and the equivalence of hypothesis testing and confidence interval estimation.
Danel Draguljic, PhD, is Assistant Professor of Mathematics at Franklin &  Marshall College, Lancaster, Pennsylvania. His research interests include design of screening experiments, design of computer experiments, and statistics education.

Edition details

• Publisher: Springer
• Publication date: 2013
• ISBN 10: 1475772920
• ISBN 13: 9781475772920
• Binding: Paperback
• Language: English
• Number of pages: 762