Making Sense of Multivariate Data Analysis: An Intuitive Approach

ISBN-13: 9781412904018 / Angielski / Miękka / 2004 / 256 str.

John Spicer

Opis

Making Sense of Multivariate Data Analysis is a short introduction to multivariate data analysis (MDA) for students and practitioners in the behavioral and social sciences. It provides a conceptual overview of the foundations of MDA and of a range of specific techniques including multiple regression, logistic regression, discriminant analysis, multivariate analysis of variance, factor analysis, and log-linear analysis. As a conceptual introduction, the book assumes no prior statistical knowledge, and contains very few symbols or equations. Its primary objective is to expose the conceptual unity of MDA techniques both in their foundations and in the common analytic strategies that lie at the heart of all of the techniques. Although introductory, the book encourages the reader to reflect critically on the general strengths and limitations of MDA techniques. Each chapter includes references for further reading accessible to the beginner.

Recenzje

"This book serves as a resource for readers who want to have an overall view of what encompasses multivariate analyses. The author has discussed some important issues rather philosophically (e.g., theory vs. data analysis). These points are valuable even for readers who have extensive training with multivariate analyses." Jenn-Yun Tein

Spis treści

Preface
Part I. The Core Ideas
1. What Makes a Difference?
1. 1 Analyzing Data in the Form of Scores
1.2 Analyzing Data in the Form of Categories
1.3 Further Reading
2. Deciding Whether Differences Are Trustworthy
2.1 Sampling Issues
2.2 Measurement Issues
2.3 The Role of Chance
2.4 Statistical Assumptions
2.5 Further Reading
3. Accounting for Differences in a Complex World
3.1 Limitations of Bivariate Analysis
3.2 The Multivariate Strategy
3.3 Common Misinterpretations of Multivariate Analyses
3.4 Further Reading
Part II. The Techniques
4. Multiple Regression
4.1 The Composite Variable in Multiple Regression
4.2 Standard Multiple Regression in Action
4.3 Trustworthiness in Regression Analysis
4.4 Accommodating Other Types of Independent Variables
4.5 Sequential Regression Analysis
4.6 Further Reading
5. Logistic Regression and Discriminant Analysis
5.1 Logistic Regression
5.2 Discriminant Analysis
5.3 Further Reading
6. Multivariate Analysis of Variance
6.1 One-Way Analysis of Variance
6.2 Factorial Analysis of Variance
6.3 Multivariate Analysis of Variance
6.4 Within-Subjects ANOVA and MANOVA
6.5 Issues of Trustworthiness in MANOVA
6.6 Analysis of Covariance
6.7 Further Reading
7. Factor Analysis
7.1 The Composite Variable in Factor Analysis
7.2 Factor Analysis in Action
7.3 Issues of Trustworthiness in Factor Analysis
7.4 Confirmatory Factor Analysis
7.5 Further Reading
8. Log-Linear Analysis
8.1 Hierarchical Log-Linear Analysis
8.2 Trustworthiness in Log-Linear Analysis
8.3 Log-Linear Analysis With a Dependent Variable: Logit Analysis
8.4 Further Reading
Bibliography
Index
About the Author

O autorze

John Spicer was an Associate Professor and Head of Psychology at Massey University, New Zealand until the end of 2002, when he took early retirement to devote all of his time to writing books. Earlier, he was a Research Fellow for several years at the University of Auckland, New Zealand, and held Visiting Fellowships at the Universities of Michigan and London. His primary research interests have been in health psychology, and he has published articles mainly on cardiovascular disease and theoretical issues in a variety of international journals. He was coeditor of Social Dimensions of Health and Disease: New Zealand Perspectives (1994). Most of his undergraduate and graduate teaching has focused on research methods, particularly multivariate data analysis. In 2002 he coauthored a chapter on sociological and psychological methods in the fourth edition of the Oxford Textbook of Public Health.