Statistical principles in experimental design
In: McGraw-Hill series in psychology
"Written primarily for students and research workers in the area of the behavioral sciences, this book is meant to provide a text and comprehensive reference source on statistical principles underlying experimental design. Particular emphasis is given to those designs that are likely to prove useful in research in the behavioral sciences. The book primarily emphasizes the logical basis of principles underlying designs for experiments rather than mathematical derivations associated with relevant sampling distributions. The topics selected for inclusion are those covered in courses taught by the author during the past several years. Students in these courses have widely varying backgrounds in mathematics and come primarily from the fields of psychology, education, economics, sociology, and industrial engineering. It has been the intention of the author to keep the book at a readability level appropriate for students having a mathematical background equivalent to freshman college algebra. From experience with those sections of the book which have been used as text material in dittoed form, there is evidence to indicate that, in large measure, the desired readability level has been attained. Admittedly, however, there are some sections in the book where this readability goal has not been achieved. The first course in design, as taught by the author, has as a prerequisite a basic course in statistical inference. The contents of Chaps. 1 and 2 review the highlights of what is included in the prerequisite material. These chapters are not meant to provide the reader with a first exposure to these topics. They are intended to provide a review of terminology and notation for the concepts which are more fully developed in later chapters. By no means is all the material included in the book covered in a one semester course. In a course of this length, the author has included Chaps. 3, 4, parts of 5, 6, parts of 7, parts of 10, and parts of 11. Chapters 8 through 11 were written to be somewhat independent of each other. Hence one may read, with understanding, in these chapters without undue reference to material in the others. In general, the discussion of principles, interpretations of illustrative examples, and computational procedures are included in successive sections within the same chapter. However, to facilitate the use of the book as a reference source, this procedure is not followed in Chaps. 5 and 6. Basic principles associated with a large class of designs for factorial experiments are discussed in Chap. 5. Detailed illustrative examples of these designs are presented in Chap. 6. For teaching purposes, the author includes relevant material from Chap. 6 with the corresponding material in Chap. 5. Selected topics from Chaps. 7 through 11 have formed the basis for a second course in experimental design. Relatively complete tables for sampling distributions of statistics used in the analysis of experimental designs are included in the Appendix. Ample references to source materials having mathematical proofs for the principles stated in the text are provided"--Preface. (PsycINFO Database Record (c) 2008 APA, all rights reserved).