Optimal Data Analysis: A Guidebook with Software for Windows (Book with CD-ROM)

by Yarnold, Paul R.

• ISBN13: 9781557989819
• ISBN10: 1557989818
• Edition: 1st
• Format: Paperback
• Copyright: 2004-10-01
• Publisher: AMERICAN PSYCHOLOGICAL ASSOCIATION

Summary

Optimal Data Analysis: A Guidebook With Software for Windows offers the only statistical analysis paradigm that maximizes (weighted) predictive accuracy. This unique book fully explains this paradigm and includes simple-to-use software that empowers a universe of associated analyses. For any specific sample and exploratory or confirmatory hypothesis, optimal data analysis (ODA) identifies the statistical model that yields maximum predictive accuracy, assesses the exact Type I error rate, and evaluates potential cross-generalizability.

Table of Contents

Preface	p. xi
Acknowledgments	p. xiii
Introduction to the ODA Paradigm	p. 3
What Is ODA?	p. 3
Why Is ODA Superior to Other Data Analysis Programs?	p. 5
Who Is the Audience for This Book and Software?	p. 6
How Should the Reader Use This Book?	p. 6
Basic Steps and Key Concepts	p. 7
Historical Perspective	p. 10
Thirty Hypothetical Applications	p. 11
Now It Is Time to Start Analyzing Data	p. 27
Using the ODA Software	p. 29
The ODA Commands	p. 29
Running the ODA Software	p. 37
Using ODA With PFE	p. 38
Creating a Data Set for Analysis by ODA	p. 45
Evaluating Classification Performance	p. 57
How to Obtain an ODA Model	p. 61
Selecting Among Multiple Optimal Models	p. 64
Assessing Model Stability	p. 68
Standardizing Transformations	p. 70
Evaluating Statistical Significance	p. 73
Analytic Methodology	p. 74
Fisher's Randomization Methodology	p. 77
Monte Carlo Methodology	p. 78
Specifying the Type I Error Rate	p. 80
A Priori Alpha Splitting	p. 83
Two-Category Class Variables	p. 87
Applications Involving Binary Attributes	p. 87
Applications Involving Polychotomous Attributes	p. 91
Applications Involving Ordinal Attributes	p. 93
Applications Involving Continuous Attributes	p. 101
Multicategory Class Variables	p. 107
Applications Involving Binary Attributes	p. 108
Applications Involving Polychotomous Attributes	p. 108
Applications Involving Ordinal Attributes	p. 112
Applications Involving Continuous Attributes	p. 115
Reliability Analysis	p. 121
Inter-Rater Reliability	p. 122
Parallel Forms Reliability	p. 128
Split-Half Reliability	p. 130
Temporal Reliability	p. 132
Nonlinear Reliability	p. 135
Intraclass Correlation	p. 138
Validity Analysis	p. 141
Hold-Out (Cross-Generalizability) Validity	p. 142
Construct Validity	p. 148
Convergent and Discriminant Validity	p. 149
Optimizing Suboptimal Multivariable Models	p. 155
Optimizing Fisher's Linear Discriminant Analysis	p. 157
Optimizing Logistic Regression Analysis	p. 160
Optimizing Complex Models	p. 165
Multiple Sample Analysis	p. 167
Pooling Samples and Simpson's Paradox	p. 168
The ODA Generalizability Algorithm	p. 170
Evaluating Model Generalizability Across Samples	p. 172
Analyzing Randomized Block Designs	p. 178
Optimizing Multiple Suboptimal Multiattribute Models	p. 181
Sequential Analyses	p. 187
Identifying Structure in Markov Transition Tables	p. 187
Analyzing Turnover Tables	p. 193
Autocorrelation (Time Series) Analysis	p. 198
Repeated Measures (Within-Subjects) Analysis	p. 203
Single-Case (N-of-1) Analysis	p. 207
Iterative Decomposition Analysis	p. 209
Stopping Rules for Iterative Analyses	p. 212
Structural Decomposition With Sequential Data	p. 214
Reliability, Bias, and Random Error	p. 223
Validity, Bias, and Random Error	p. 225
Epilogue: The Future of ODA	p. 229
General-Purpose MultiODA Models	p. 232
Special-Purpose MultiODA Models	p. 233
Nonlinear Classification Tree Analysis	p. 237
Users of ODA	p. 239
Dunn and Sidak Adjusted Per-Comparison p	p. 241
Troubleshooting: Common Problems and Their Possible Solutions	p. 249
References	p. 253
Index	p. 275
About the Authors	p. 287
Table of Contents provided by Rittenhouse. All Rights Reserved.