Multiple Regression and Beyond

by Keith, Timothy Z.

• ISBN13: 9780205326440
• ISBN10: 0205326447
• Edition: 1st
• Format: Hardcover
• Copyright: 2005-06-24
• Publisher: Pearson

Summary

This book is designed to provide a conceptually-oriented introduction to multiple regression. It is divided into two main parts. The author concentrates on multiple regression analysis in the first part and structural equation modeling in the second part. Provides an introduction to multiple regression analysis, path analysis, confirmatory factor analysis, and structural equation modeling. Students gain a foundation in multiple regression. Multiple regression is used as a jumping-off place for structural equation modeling with measured variables (path analysis). Latent variables are introduced as a way of obviating some of the threats to path models, and SEM with latent variables is presented as the culmination of the previous topics. (not so sure about this one) Conceptual, rather than formula oriented, approach

Table of Contents

Preface	p. xiii
Multiple Regression	p. 1
Introduction and Simple (Bivariate) Regression	p. 1
Simple (Bivariate) Regression	p. 2
Regression in Perspective	p. 13
Other Issues	p. 17
Review of Some Basics	p. 19
Working with Extant Data Sets	p. 20
Summary	p. 22
Exercises	p. 23
Notes	p. 23
Multiple Regression: Introduction	p. 25
A New Example: Regressing Grades on Homework and Parent Education	p. 26
Questions	p. 33
Direct Calculation of [beta] and R[superscript 2]	p. 39
Summary	p. 41
Exercises	p. 41
Notes	p. 42
Multiple Regression: More Detail	p. 43
Why R[superscript 2] '' r[superscript 2]+r[superscript 2]	p. 43
Predicted Scores and Residuals	p. 46
Least Squares	p. 51
Regression Equation = Creating a Composite?	p. 52
Assumptions of Regression and Regression Diagnostics	p. 54
Summary	p. 54
Exercises	p. 55
Note	p. 55
Three and More Independent Variables and Related Issues	p. 56
Three Predictor Variables	p. 56
Rules of Thumb: Magnitude of Effects	p. 61
Four Independent Variables	p. 62
Common Causes and Indirect Effects	p. 66
The Importance of R[superscript 2]?	p. 68
Prediction and Explanation	p. 69
Summary	p. 71
Exercises	p. 72
Notes	p. 73
Three Types of Multiple Regression	p. 74
Simultaneous Multiple Regression	p. 76
Sequential Multiple Regression	p. 78
Stepwise Multiple Regression	p. 92
The Purpose of the Research	p. 99
Combining Methods	p. 101
Summary	p. 102
Exercises	p. 103
Notes	p. 104
Analysis of Categorical Variables	p. 105
Dummy Variables	p. 106
Other Methods of Coding Categorical Variables	p. 114
Unequal Group Sizes	p. 118
Additional Methods and Issues	p. 125
Summary	p. 125
Exercises	p. 127
Notes	p. 127
Categorical and Continuous Variables	p. 129
Sex, Achievement, and Self-Esteem	p. 130
Interactions	p. 132
A Statistically Significant Interaction	p. 137
Specific Types of Interactions Between Categorical and Continuous Variables	p. 141
Caveats and Additional Information	p. 156
Summary	p. 158
Exercises	p. 159
Notes	p. 160
Continuous Variables: Interactions and Curves	p. 161
Interactions Between Continuous Variables	p. 161
Moderation, Mediation, and Common Cause	p. 168
Curvilinear Regression	p. 170
Summary	p. 178
Exercises	p. 178
Note	p. 179
Multiple Regression: Summary, Further Study, and Problems	p. 180
Summary	p. 180
Assumptions and Regression Diagnostics	p. 186
Topics for Additional Study	p. 202
Problems with Mr?	p. 207
Exercises	p. 211
Note	p. 211
Beyond Multiple Regression	p. 212
Path Modeling: Structural Equation Modeling with Measured Variables	p. 212
Introduction to Path Analysis	p. 213
A More Complex Example	p. 223
Summary	p. 233
Exercises	p. 235
Notes	p. 236
Path Analysis: Dangers and Assumptions	p. 238
Assumptions	p. 238
The Danger of Common Causes	p. 240
Intervening (Mediating) Variables	p. 245
Other Possible Dangers	p. 247
Dealing with Danger	p. 249
Review: Steps in a Path Analysis	p. 250
Summary	p. 251
Exercises	p. 252
Notes	p. 253
Analyzing Path Models Using SEM Programs	p. 254
SEM Programs	p. 254
Reanalysis of the Parent Involvement Path Model	p. 256
Advantages of SEM Programs	p. 260
More Complex Models	p. 274
Advice: Mr Versus SEM Programs	p. 283
Summary	p. 284
Exercises	p. 286
Notes	p. 288
Error: The Scourge of Research	p. 289
Effects of Unreliability	p. 290
Effects of Invalidity	p. 295
Latent Variable SEM and Errors of Measurement	p. 298
Summary	p. 303
Exercises	p. 304
Notes	p. 304
Confirmatory Factor Analysis	p. 305
Factor Analysis or the Measurement Model	p. 305
An Example with the Das	p. 306
Testing Competing Models	p. 314
Hierarchical Models	p. 318
Model Fit and Model Modification	p. 321
Additional uses of CFA	p. 325
Summary	p. 328
Exercises	p. 330
Putting It All Together: Introduction to Latent Variable SEM	p. 331
Putting the Pieces Together	p. 332
An Example: Effects of Peer Rejection	p. 333
Competing Models	p. 341
Model Modifications	p. 344
Summary	p. 346
Exercises	p. 348
Latent Variable Models: More Advanced Topics	p. 350
Single Indicators and Correlated Errors	p. 350
Multisample Models	p. 362
Replication and Cross-Validation	p. 374
Dangers, Revisited	p. 377
Summary	p. 381
Exercises	p. 383
Notes	p. 384
Summary: Path Analysis, CFA, and SEM	p. 385
Summary	p. 385
Issues Incompletely or Not Covered	p. 394
Additional Resources	p. 397
Data Files	p. 400
Sample Statistical Programs and Multiple Regression Output	p. 433
Sample Output from SEM Programs	p. 452
Partial and Semipartial Correlation	p. 482
Review of Basic Statistics Concepts	p. 491
References	p. 511
Name Index	p. 517
Subject Index	p. 520
Table of Contents provided by Ingram. All Rights Reserved.