Applied multivariate statistics for the social sciences / James Stevens.

Includes bibliographical references and index.

1.2 Type I Error, Type II Error, and Power 3 -- 1.3 Multiple Statistical Tests and the Probability of Spurious Results 6 -- 1.4 Statistical Significance Versus Practical Significance 9 -- 1.5 Outliers 12 -- 1.6 Research Examples for Some Analyses Considered in This Text 17 -- 1.7 SAS and SPSS Statistical Packages 24 -- 1.8 SPSS for Windows -- Releases 9.0 and 10.0 34 -- 1.9 Data Files 36 -- 1.10 Data Editing 40 -- 1.11 SPSS Output Navigator 45 -- 1.12 Some Issues Unique to Multivariate Analysis 48 -- 1.13 Data Collection and Integrity 49 -- Appendix 1 Defining a Measure of Statistical Distance 50 -- Appendix 2 Milk Data 52 -- Chapter 2 Matrix Algebra -- 2.2 Addition, Subtraction, and Multiplication of a Matrix by a Scalar 59 -- 2.3 Obtaining the Matrix of Variances and Covariances 62 -- 2.4 Determinant of a Matrix 64 -- 2.5 Inverse of a Matrix 70 -- 2.6 Eigenvalues 73 -- 2.7 SPSS Matrix Procedure 75 -- 2.8 SAS IML Procedure 76 -- Chapter 3 Multiple Regression -- 3.2 Simple Regression 82 -- 3.3 Multiple Regression for Two Predictors -- Matrix Formulation 86 -- 3.4 Mathematical Maximization Nature of Least Squares Regression 88 -- 3.5 Breakdown of Sum of Squares in Regression and F Test for Multiple Correlation 89 -- 3.6 Relationship of Simple Correlations to Multiple Correlation 91 -- 3.7 Multicollinearity 91 -- 3.8 Model Selection 93 -- 3.9 Two Computer Examples 98 -- 3.10 Checking Assumptions for the Regression Model 110 -- 3.11 Model Validation 113 -- 3.12 Importance of the Order of the Predictors in Regression Analysis 119 -- 3.13 Other Important Issues 121 -- 3.14 Outliers and Influential Data Points 125 -- 3.15 Further Discussion of the Two Computer Examples 138 -- 3.16 Sample Size Determination for a Reliable Prediction Equation 143 -- 3.17 Logistic Regression 146 -- 3.18 Other Types of Regression Analysis 155 -- 3.19 Multivariate Regression 155 -- 3.20 Summary of Important Points 159 -- Chapter 4 Two-Group Multivariate Analysis Of Variance -- 4.2 Four Statistical Reasons for Preferring a Multivariate Analysis 174 -- 4.3 Multivariate Test Statistic as a Generalization of Univariate t 175 -- 4.4 Numerical Calculations for a Two-Group Problem 177 -- 4.5 Three Post Hoc Procedures 181 -- 4.6 SAS and SPSS Control Lines for Sample Problem and Selected Printout 183 -- 4.7 Multivariate Significance but No Univariate Significance 184 -- 4.8 Multivariate Regression Analysis for the Sample Problem 188 -- 4.9 Power Analysis 192 -- 4.10 Ways of Improving Power 195 -- 4.11 Power Estimation on SPSS MANOVA 197 -- 4.12 Multivariate Estimation of Power 197 -- Chapter 5 K-Group Manova: A Priori And Post Hoc Procedures -- 5.2 Multivariate Regression Analysis for a Sample Problem 209 -- 5.3 Traditional Multivariate Analysis of Variance 210 -- 5.4 Multivariate Analysis of Variance for Sample Data 212 -- 5.5 Post Hoc Procedures 217 -- 5.6 Tukey Procedure 222 -- 5.7 Planned Comparisons 225 -- 5.8 Test Statistics for Planned Comparisons 228 -- 5.9 Multivariate Planned Comparisons on SPSS MANOVA 231 -- 5.10 Correlated Contrasts 235 -- 5.11 Studies Using Multivariate Planned Comparisons 241 -- 5.12 Stepdown Analysis 243 -- 5.13 Other Multivariate Test Statistics 243 -- 5.14 How Many Dependent Variables for a MANOVA? 245 -- 5.15 Power Analysis -- A Priori Determination of Sample Size 245 -- Appendix Novince (1977) Data for Multivariate Analysis of Variance Presented in Tables 5.3 and 5.4 249 -- Chapter 6 Assumptions In Manova -- 6.2 ANOVA and MANOVA Assumptions 257 -- 6.3 Independence Assumption 258 -- 6.4 What Should Be Done With Correlated Observations? 260 -- 6.5 Normality Assumption 261 -- 6.6 Multivariate Normality 262 -- 6.7 Assessing Univariate Normality 263 -- 6.8 Homogeneity of Variance Assumption 268 -- 6.9d Homogeneity of the Covariance Matrices 269 -- 6.10 General Procedure for Assessing Violations in MANOVA 276 -- Appendix Multivariate Test Statistics for Unequal Covariance Matrices 279 -- Chapter 7 Discriminant Analysis -- 7.2 Descriptive Discriminant Analysis 286 -- 7.3 Significance Tests 287 -- 7.4 Interpreting the Discriminant Functions 288 -- 7.5 Graphing the Groups in the Discriminant Plane 289 -- 7.6 Rotation of the Discriminant Functions 296 -- 7.7 Stepwise Discriminant Analysis 296 -- 7.8 Two Other Studies That Used Discriminant Analysis 297 -- 7.9 Classification Problem 301 -- 7.10 Linear vs. Quadratic Classification Rule 316 -- 7.11 Characteristics of a Good Classification Procedure 316 -- Chapter 8 Factorial Analysis Of Variance -- 8.2 Advantages of a Two-Way Design 322 -- 8.3 Univariate Factorial Analysis 324 -- 8.4 Factorial Multivariate Analysis of Variance 331 -- 8.5 Weighting of the Cell Means 332 -- 8.6 Three-Way MANOVA 335 -- Chapter 9 Analysis Of Covariance -- 9.2 Purposes of Covariance 340 -- 9.3 Adjustment of Posttest Means and Reduction of Error Variance 342 -- 9.4 Choice of Covariates 345 -- 9.5 Assumptions in Analysis of Covariance 347 -- 9.6 Use of ANCOVA With Intact Groups 350 -- 9.7 Alternative Analyses for Pretest-Posttest Designs 351 -- 9.8 Error Reduction and Adjustment of Posttest Means for Several Covariates 353 -- 9.9 MANCOVA -- Several Dependent Variables and Several Covariates 354 -- 9.10 Testing the Assumption of Homogeneous Regression Hyperplanes on SPSS 355 -- 9.11 Two Computer Examples 356 -- 9.12 Bryant-Paulson Simultaneous Test Procedure 361 -- Chapter 10 Stepdown Analysis -- 10.2 Four Appropriate Situations for Stepdown Analysis 375 -- 10.3 Controlling on Overall Type I Error 376 -- 10.4 Stepdown F's for Two Groups 377 -- 10.5 Comparison of Interpretation of Stepdown F's vs.

Univariate F's 379 -- 10.6 Stepdown F's for k Groups -- Effect of Within and Between Correlations 381 -- Chapter 11 Confirmatory And Exploratory Factor Analysis -- 11.2 Nature of Principal Components 386 -- 11.3 Three Uses for Components as a Variable Reducing Scheme 388 -- 11.4 Criteria for Deciding on How Many Components to Retain 389 -- 11.5 Increasing Interpretability of Factors by Rotation 391 -- 11.6 What Loadings Should Be Used for Interpretation? 393 -- 11.7 Sample Size and Reliable Factors 395 -- 11.8 Four Computer Examples 395 -- 11.9 Communality Issue 409 -- 11.11 Exploratory and Confirmatory Factor Analysis 411 -- 11.12 PRELIS 415 -- 11.13 A LISREL Example Comparing Two A Priori Models 419 -- 11.14 Identification 427 -- 11.15 Estimation 429 -- 11.16 Assessment of Model Fit 430 -- 11.17 Model Modification 435 -- 11.18 LISREL 8 Example 437 -- 11.19 EQS Example 445 -- 11.20 Some Caveats Regarding Structural Equation Modeling 449 -- Chapter 12 Canonical Correlation -- 12.2 Nature of Canonical Correlation 472 -- 12.3 Significance Tests 473 -- 12.4 Interpreting the Canonical Variates 475 -- 12.5 Computer Example Using SAS CANCORR 476 -- 12.6 A Study That Used Canonical Correlation: Relationship Between Student Needs and Teacher Ratings 479 -- 12.7 Using SAS for Canonical Correlation on Two Sets of Factor Scores 481 -- 12.8 Redundancy Index of Stewart and Love 483 -- 12.9 Rotation of Canonical Variates 485 -- 12.10 Obtaining More Reliable Canonical Variates 485 -- Chapter 13 Repeated Measures Analysis -- 13.2 Single-Group Repeated Measures 496 -- 13.3 Multivariate Test Statistic for Repeated Measures 497 -- 13.4 Assumptions in Repeated Measures Analysis 500 -- 13.5 Computer Analysis of the Drug Data 502 -- 13.6 Post Hoc Procedures in Repeated Measures Analysis 506 -- 13.7 Should We Use the Univariate or Multivariate Approach? 509 -- 13.8 Sample Size for Power = .80 in Single-Sample Case 510 -- 13.9 Multivariate Matched Pairs Analysis 512 -- 13.10 One Between and One Within Factor -- A Trend Analysis 512 -- 13.11 Post Hoc Procedures for the One Between and One Within Design 519 -- 13.12 One Between and Two Within Factors 521 -- 13.13 Two Between and One Within Factors 526 -- 13.14 Two Between and Two Within Factors 532 -- 13.15 Totally Within Designs 532 -- 13.16 Planned Comparisons in Repeated Measures Designs 534 -- 13.17 Profile Analysis 536 -- 13.18 Doubly Multivariate Repeated Measures Designs 538 -- Chapter 14 Categorical Data Analysis: The Log Linear Model -- 14.2 Sampling Distributions: Binomial and Multinomial 561 -- 14.3 Two Way Chi Square -- Log Linear Formulation 564 -- 14.4 Three-Way Tables 567 -- 14.5 Model Selection 576 -- 14.6 Collapsibility 578 -- 14.7 Odds (Cross-Product) Ratio 582 -- 14.8 Normed Fit Index and Residual Analysis 583 -- 14.9 Residual Analysis 584 -- 14.10 Cross-Validation 585 -- 14.11 Higher Dimensional Tables -- Model Selection -- 14.12 Contrasts for the Log Linear Model 586 -- 14.13 Log Linear Analysis for Ordinal Data 590 -- 14.14 Sampling and Structural (Fixed) Zeros 595 -- Appendix A Statistical Tables 614 -- Appendix B Data Sets 634 -- Appendix C Obtaining Nonorthogonal Contrasts in Repeated Measures Designs 653.

CD-ROM contains: Data sets.