Applied Statistics: From Bivariate through Multivariate Techniques.
by Rebecca Warner
Sage Publications, Inc.
Contact: rmw@unh.edu
Most recent update for this page: March 4, 2009
The Sage study site for the book is at: http://www.sagepub.com/warnerstudy
Instructor ancillary materials including PowerPoint, answers to comprehension questions at the end of each chapter, data sets used in the text, additional data sets for student projects, and SAS examples can be requested from Sage as an Instructor Resource CD.
Known errors in the text are updated as I become aware of them . If you notice errors in the text please send me an email so that these can be corrected in future editions.
To read the publisher description including Table of Contents or to request a complimentary instructor copy, go to the Sage website.
This book helps students to answer fundamental questions in statistics, such as the following:
Why is there variance?
What does it mean to say "p < .05"?
How do researcher decisions about treatment dosage levels and control over extraneous variables tend to influence the magnitudes of test statistics such as t ratios?
What do we mean when we refer to a partition of variance, or a proportion of explained or predicted variance?
What does it mean to say that we have "statistically controlled for" one variable while assessing the nature and strength of association between other variables?
How do we evaluate whether scores on variables are "contingent"?
What features of experimental design help us to rule out rival explanations and to interpret results as possible evidence of a causal association between variables?
How do multivariate analyses such as MANOVA and Discriminant Analysis make it possible for us to examine sets of intercorrelated predictor or outcome variables?
The book begins with a review of bivariate analyses (t test, Pearson correlation) and provides a clear introduction to widely used topics in multivariate statistics, including multiple regression, discriminant analysis, MANOVA, factor analysis, and binary logistic regression.
The approach is applied and does not require formal mathematics; equations are accompanied by verbal explanations. Students are asked to think about the meaning of equations, and to consider how quantitative outcomes are related to research design decisions. For example, how do decisions about treatment dosage levels and sample size tend to influence the differences between sample means, and the magnitudes of t and F ratios?
Each chapter presents a complete empirical research example to illustrate the application of a specific method, such as multiple regression. Although SPSS examples are used throughout the book, the conceptual material will be helpful for users of different programs. A companion study site includes all datasets used in the book and additional datasets for student projects in both SPSS and Excel file formats; SAS screen shots, script and output for all of the examples in each chapter; and an extensive Microsoft® PowerPoint® presentation for each chapter. Complete answers to the comprehension questions at the end of each chapter will be available only to instructors so that these end of chapter questions can be used in graded assignments.
Features:
The text begins with a clear review and a fresh perspective on important concepts including effect size, variance partitioning, the logic of null hypothesis statistical significance tests, and statistical control. Depending on student background and the level of the course, instructors can begin with chapters that review basic material, or start with more advanced topics and refer to earlier chapters as supplemental review material. The material is suitable for either an advanced undergraduate or graduate level course.
Chapters 10 through 13 examine three variable research situations in detail and teach students how to think about statistical control: How does the nature and strength of the association between an X1 predictor variable and a Y outcome variable change when we statistically control for another (X2) variable? This understanding of statistical control is essential for comprehension of multivariate analyses.
The book includes a chapter on reliability, validity, and multiple item scales, and draws extensively on path models to illustrate theories about possible causal and non-causal associations among variables, beginning with simple three variable research situations.
Graphics are used to explain concepts such as variance partitioning, statistical control, and factor rotation.
About the Author:
Rebecca M. Warner received a B.A. from Carnegie-Mellon University in Social Relations in 1973 and a Ph.D. in Social Psychology from Harvard in 1978. She has taught statistics for more than 25 years; her courses have included Introductory and Intermediate Statistics as well as advanced topics seminars in Multivariate Statistics, Structural Equation Modeling, and Time Series Analysis. She is currently a Full Professor in the Department of Psychology at the University of New Hampshire. She received a UNH Liberal Arts Excellence in Teaching Award in 1992. She is a Fellow in the Association for Psychological Science and a member of the American Psychological Association, Division 5 of APA (Evaluation, Measurement, and Statistics), the Association for Psychological Science, the International Association for Relationships Research, the Society of Experimental Social Psychology, and the Society for Personality and Social Psychology. She has consulted on statistics and data management for the World Health Organization in Geneva and served as a visiting faculty member at Shandong Medical University in China. Her previous book, The spectral analysis of time-series data, was published by Guilford in 1998. She has published articles on statistics, health psychology, and social psychology in numerous journals including the Journal of Personality and Social Psychology; she has served as a reviewer for many journals including Psychological Bulletin, Psychological Measurement, and Psychometrika. Dr. Warner has taught in the Semester at Sea program and worked with Project Orbis to set up a data management system for an airplane-based surgical education program.
She has also written a novel, A. D. 62: Pompeii under the pen name Rebecca East.
Current Data Collection and Data Analysis:
(1) We are obtaining observer ratings of interviewer behaviors for video recordings of interviews the medical students have conducted with simulated patients. Analyses will examine whether quality of interview skills can be predicted from scores on Emotional Intelligence, Physician Empathy, MCAT Verbal scores, and other background information.
(2) We have obtained self-evaluations of ability to provide social support from undergraduates, along with their scores on Jack Mayer's MSCEIT (an ability based test of Emotional Intelligence). In addition, for half of our sample, we also have ratings of quality of social support made by a "significant other" person (a parent, friend, or roommate). We included several different self report measures of quantity, quality, and type of social support. Initial factor analyses will help us to evaluate whether self reported social support can be described as multidimensional (and if so, to identify some possible dimensions). Later analyses will examine whether self and other evaluations of quality of social support are predictable from Emotional Intelligence and from student gender.
(3) We are also examining associations among Emotional Intelligence, Machiavellianism, alexithymia, and self reported success in deception and manipulation. We expect that persons high on both Emotional Intelligence and Machiavellianism may be more skillful at deception and manipulation than other groups.
Recent Conference Presentation Posters:
Rebecca Warner's Other Books:
Warner, R. M. (1998). Spectral analysis of time-series data. New York: Guilford.
(review of Spectral Analysis in Technometrics)
East, R. (2003). A. D. 62: Pompeii, a novel. (link to Amazon listing)
Selected Journal Articles:
The following are downloadable PDF files. Other journal articles are available upon request.
Michaud, S. & Warner, R. (1997). Gender differences in self-reported response to troubles talk. Sex Roles, 37, 527-540.(and the Communication Styles Survey described in this paper).
Warner, R. M. (1979). Periodic rhythms in conversational speech. Language and Speech, 22, 381-396.
Other Links:
Most Influential Teachers:
Colleagues, Collaborators, and Friends:
Selected Statistics Web Sites:
Dave Kenny's Discussion of Mediation