Modern statistics for the social and behavioral sciences : a practical introduction

Note continued: 13.6.4. R Function cmanova --
13.7. Multivariate Regression --
13.7.1. Multivariate Regression Using R --
13.7.2. Robust Multivariate Regression --
13.7.3. R Function mlrreg and mopreg --
13.8. Principal Components --
13.8.1. R Functions prcomp and regpca --
13.8.2. Robust Principal Components --
13.8.3. R Functions outpca, robpca, robpcaS, Ppca, and Ppca.summary --
13.9. Exercises --
ch. 14 Robust Regression and Measures of Association --
14.1. Robust Regression Estimators --
14.1.1. The Theil --
Sen Estimator --
14.1.2. R Functions tsreg, tshdreg, and regplot --
14.1.3. Least Median of Squares --
14.1.4. Least Trimmed Squares and Least Trimmed Absolute Value Estimators --
14.1.5. R Functions lmsreg, ltsreg, and ltareg --
14.1.6. M-estimators --
14.1.7. R Function chreg --
14.1.8. Deepest Regression Line --
14.1.9. R Function mdepreg --
14.1.10. Skipped Estimators --
14.1.11. R Functions opreg and opregMC --
14.1.12. S-estimators and an E-type Estimator --
14.1.13. R Function tstsreg --
14.2. Comments on Choosing a Regression Estimator --
14.3. Inferences Based on Robust Regression Estimators --
14.3.1. Testing Hypotheses About the Slopes --
14.3.2. Inferences About the Typical Value of Y Given X --
14.3.3. R Functions regtest, regtestMC, regci, regciMC, regYci, and regYband --
14.3.4. Comparing Measures of Location via Dummy Coding --
14.4. Dealing with Curvature: Smoothers --
14.4.1. Cleveland\'s Smoother --
14.4.2. R Functions lowess, lplot, lplot.pred, and lplotCI --
14.4.3. Smoothers Based on Robust Measures of Location --
14.4.4. R Functions rplot, rplotCIS, rplotCI, rplotCIv2, rplotCIM, rplot.pred, qhdsm, and qhdsm.pred --
14.4.5. Prediction When X Is Discrete: The R Function rundis --
14.4.6. Seeing Curvature with More Than Two Predictors --
14.4.7. R Function prplot --
14.4.8. Some Alternative Methods --
14.4.9. Detecting Heteroscedasticity Using a Smoother --
14.4.10. R Function rhom --
14.5. Some Robust Correlations and Tests of Independence --
14.5.1. Kendall\'s tau --
14.5.2. Spearman\'s rho --
14.5.3. Winsorized Correlation --
14.5.4. R Function wincor --
14.5.5. OP or Skipped Correlation --
14.5.6. R Function scor --
14.5.7. Inferences about Robust Correlations: Dealing with Heteroscedasticity --
14.5.8. R Functions corb and scorci --
14.6. Measuring the Strength of an Association Based on a Robust Fit --
14.7. Comparing the Slopes of Two Independent Groups --
14.7.1. R Function reg2ci --
14.8. Tests for Linearity --
14.8.1. R Functions lintest, lintestMC, and linchk --
14.9. Identifying the Best Predictors --
14.9.1. Inferences Based on Independent Variables Taken in Isolation --
14.9.2. R Functions regpord, ts2str, and sm2strv7 --
14.9.3. Inferences When Independent Variables Are Taken Together --
14.9.4. R Function reglVcom --
14.10. Interactions and Moderator Analyses --
14.10.1. R Functions olshc4.inter, ols.plot.inter, regci.inter, reg.plot.inter and adtest --
14.10.2. Graphical Methods for Assessing Interactions --
14.10.3. R Functions kercon, runsm2g, regi --
14.11. ANCOVA --
14.11.1. Classic ANCOVA --
14.11.2. Robust ANCOVA Methods Based on a Parametric Regression Model --
14.11.3. R Functions ancJN, ancJNmp, anclin, reg2plot, and reg2g.p2plot --
14.11.4. ANCOVA Based on the Running-interval Smoother --
14.11.5. R Functions ancsm, Qancsm, ancova, ancovaWMW, ancpb, ancov-aUB, ancboot, ancdet, runmean2g, qhdsm2g, and 12plot --
14.11.6. R Functions Dancts, Dancols, Dancova, Dancovapb, DancovaUB, and Dancdet --
14.12. Exercises --
ch. 15 Basic Methods for Analyzing Categorical Data --
15.1. Goodness of Fit --
15.1.1. R Functions chisq.test and pwr.chisq.test --
15.2. A Test of Independence --
15.2.1. R Function chi.test.ind --
15.3. Detecting Differences in the Marginal Probabilities --
15.3.1. R Functions contab and mcnemar.test --
15.4. Measures of Association --
15.4.1. The Proportion of Agreement --
15.4.2. Kappa --
15.4.3. Weighted Kappa --
15.4.4. R Function Ckappa --
15.5. Logistic Regression --
15.5.1. R Functions glm and logreg --
15.5.2. A Confidence Interval for the Odds Ratio --
15.5.3. R Function ODDSR. CI --
15.5.4. Smoothers for Logistic Regression --
15.5.5. R Functions logrsm, rplot.bin, and logSM --
15.6. Exercises --
Appendix A Answers to Selected Exercises --
Appendix B TABLES --
Appendix C BASIC MATRIX ALGEBRA. Note continued: 7.8.9. R Function ks --
7.8.10. Comparing All Quantiles Simultaneously: An Extension of the Kolmogorov-Smirnov Test --
7.8.11. R Function sband --
7.9. Graphical Methods for Comparing Groups --
7.9.1. Error Bars --
7.9.2. R Functions ebarplot and ebarplot.med --
7.9.3. Plotting the Shift Function --
7.9.4. Plotting the Distributions --
7.9.5. R Function sumplot2g --
7.9.6. Other Approaches --
7.10. Comparing Measures of Variation --
7.10.1. R Function comvar2 --
7.10.2. Brown-Forsythe Method --
7.10.3. Comparing Robust Measures of Variation --
7.11. Measuring Effect Size --
7.11.1. R Functions yuenv2 and akp.effect --
7.12. Comparing Correlations and Regression Slopes --
7.12.1. R Functions twopcor, twolsreg, and tworegwb --
7.13. Comparing Two Binomials --
7.13.1. Storer-Kim Method --
7.13.2. Beal\'s Method --
7.13.3. R Functions twobinom, twobici, bi2KMSv2, and power.prop.test --
7.13.4. Comparing Two Discrete Distributions --
7.13.5. R Function disc2com --
7.14. Making Decisions About which Method to Use --
7.15. Exercises --
ch. 8 Comparing Two Dependent Groups --
8.1. The Paired T Test --
8.1.1. When Does the Paired T Test Perform Well? --
8.1.2. R Function t. test --
8.2. Comparing Robust Measures of Location --
8.2.1. R Functions yuend, ydbt, and dmedpb --
8.2.2. Comparing Marginal M-Estimators --
8.2.3. R Function rmmest --
8.2.4. Measuring Effect Size --
8.2.5. R Function D.akp.effect --
8.3. Handling Missing Values --
8.3.1. R Functions rm2miss and rmmismcp --
8.4. A Different Perspective when Using Robust Measures of Location --
8.4.1. R Functions loc2dif and 12drmci --
8.5. The Sign Test --
8.5.1. R Function signt --
8.6. Wilcoxon Signed Rank Test --
8.6.1. R Function wilcox.test --
8.7. Comparing Variances --
8.7.1. R Function comdvar --
8.8. Comparing Robust Measures of Scale --
8.8.1. R Function rmrvar --
8.9. Comparing All Quantiles --
8.9.1. R Functions lband --
8.10. Plots for Dependent Groups --
8.10.1. R Function g2plotdifxy --
8.11. Exercises --
ch. 9 One-Way Anova --
9.1. Analysis of Variance for Independent Groups --
9.1.1. A Conceptual Overview --
9.1.2. ANOVA via Least Squares Regression and Dummy Coding --
9.1.3. R Functions anova, anoval, aov, and fac2list --
9.1.4. Controlling Power and Choosing the Sample Sizes --
9.1.5. R Functions power.anova.test and anova.power --
9.2. Dealing with Unequal Variances --
9.2.1. Welch\'s Test --
9.3. Judging Sample Sizes and Controlling Power when Data are Available --
9.3.1. R Functions bdanoval and bdanova2 --
9.4. Trimmed Means --
9.4.1. R Functions t1way, tlwayv2, t1wayF, and g5plot --
9.4.2. Comparing Groups Based on Medians --
9.4.3. R Function med1way --
9.5. Bootstrap Methods --
9.5.1. A Bootstrap-t Method --
9.5.2. R Functions t1waybt and BFBANOVA --
9.5.3. Two Percentile Bootstrap Methods --
9.5.4. R Functions b1way, pbadepth, and Qanova --
9.5.5. Choosing a Method --
9.6. Random Effects Model --
9.6.1. A Measure of Effect Size --
9.6.2. A Heteroscedastic Method --
9.6.3. A Method Based on Trimmed Means --
9.6.4. R Function rananova --
9.7. Rank-Based Methods --
9.7.1. The Kruskall --
Wallis Test --
9.7.2. R Function kruskal.test --
9.7.3. Method BDM --
9.7.4. R Functions bdm and bdmP --
9.8. Exercises --
ch. 10 Two-Way and Three-Way Designs --
10.1. Basics of a Two-Way Anova Design --
10.1.1. Interactions --
10.1.2. R Functions interaction.plot and interplot --
10.1.3. Interactions When There Are More Than Two Levels --
10.2. Testing Hypotheses About Main Effects and Interactions --
10.2.1. R function anova --
10.2.2. Inferences About Disordinal Interactions --
10.2.3. The Two-Way ANOVA Model --
10.3. Heteroscedastic Methods for Trimmed Means, Includingmeans --
10.3.1. R Function t2way --
10.4. Bootstrap Methods --
10.4.1. R Functions pbad2way and t2waybt --
10.5. Testing Hypotheses Based on Medians --
10.5.1. R Function m2way --
10.6. A Rank-Based Method for a Two-Way Design --
10.6.1. R Function bdm2way --
10.6.2. The Patel --
Hoel Approach to Interactions --
10.7. Three-Way Anova --
10.7.1. R Functions anova and t3way --
10.8. Exercises --
ch. 11 Comparing More than Two Dependent Groups --
11.1. Comparing Means in a One-Way Design --
11.1.1. R Function aov --
11.2. Comparing Trimmed Means When Dealing With a One-Way Design --
11.2.1. R Functions rmanova and rmdat2mat --
11.2.2. A Bootstrap-t Method for Trimmed Means --
11.2.3. R Function rmanovab --
11.3. Percentile Bootstrap Methods for a One-Way Design --
11.3.1. Method Based on Marginal Measures of Location --
11.3.2. R Function bdlway --
11.3.3. Inferences Based on Difference Scores --
11.3.4. R Function rmdzero --
11.4. Rank-Based Methods for a One-Way Design --
11.4.1. Friedman\'s Test --
11.4.2. R Function friedman.test --
11.4.3. Method BPRM --
11.4.4. R Function bprm --
11.5. Comments on Which Method to Use --
11.6. Between-By-Within Designs --
11.6.1. Method for Trimmed Means --
11.6.2. R Function bwtrim and bw2list --
11.6.3. A Bootstrap-t Method --
11.6.4. R Function tsplitbt --
11.6.5. Inferences Based on M-estimators and Other Robust Measures of Location --
11.6.6. R Functions sppba, sppbb, and sppbi --
11.6.7. A Rank-Based Test --
11.6.8. R Function bwrank --
11.7. Within-By-Within Design --
11.7.1. R Function wwtrim --
11.8. Three-Way Designs --
11.8.1. R Functions bbwtrim, bwwtrim, and wwwtrim --
11.8.2. Data Management: R Functions bw2list and bbw2list --
11.9. Exercises --
ch. 12 Multiple Comparisons --
12.1. One-Way Anova and Related Situations, Independent Groups --
12.1.1. Fisher\'s Least Significant Difference Method --
12.1.2. The Tukey --
Kramer Method --
12.1.3. R Function TukeyHSD --
12.1.4. Tukey --
Kramer and the ANOVA F Test --
12.1.5. Step-Down Methods --
12.1.6. Dunnett\'s T3 --
12.1.7. Games --
Howell Method --
12.1.8. Comparing Trimmed Means --
12.1.9. R Functions lincon, stepmcp and twoKlin --
12.1.10. Alternative Methods for Controlling FWE --
12.1.11. Percentile Bootstrap Methods for Comparing Trimmed Means, Medians, and M-estimators --
12.1.12. R Functions medpb, linconpb, pbmcp, and p.adjust --
12.1.13. A Bootstrap-t Method --
12.1.14. R Function linconbt --
12.1.15. Rank-Based Methods --
12.1.16. R Functions cidmul, cidmulv2, and bmpmul --
12.1.17. Comparing the Individual Probabilities of Two Discrete Distributions --
12.1.18. R Functions binband, splotg2, cumrelf, and cumrelfT --
12.1.19. Comparing the Quantifies of Two Independent Groups --
12.1.20. R Functions qcomhd and qcomhdMC --
12.1.21. Multiple Comparisons for Binomial and Categorical Data --
12.1.22. R Functions skmcp and discmcp --
12.2. Two-Way, Between-By-Between Design --
12.2.1. Scheffe\'s Homoscedastic Method --
12.2.2. Heteroscedastic Methods --
12.2.3. Extension of Welch --
Sidak and Kaiser --
Bowden Methods to Trimmed Means --
12.2.4. R Function kbcon --
12.2.5. R Functions con2way and conCON --
12.2.6. Linear Contrasts Based on Medians --
12.2.7. R Functions msmed and mcp2med --
12.2.8. Bootstrap Methods --
12.2.9. R Functions mcp2a, bbmcppb, bbmcp --
12.2.10. The Patel --
Hoel Rank-Based Interaction Method --
12.2.11. R Function rimul --
12.3. Judging Sample Sizes --
12.3.1. Tamhane\'s Procedure --
12.3.2. R Function tamhane --
12.3.3. Hochberg\'s Procedure --
12.3.4. R Function hochberg --
12.4. Methods for Dependent Groups --
12.4.1. Linear Contrasts Based on Trimmed Means --
12.4.2. R Function rmmcp --
12.4.3. Comparing M-estimators --
12.4.4. R Functions rmmcppb, dmedpb, dtrimpb, and boxdif --
12.4.5. Bootstrap-t Method --
12.4.6. R Function bptd --
12.4.7. Comparing the Quantiles of the Marginal Distributions --
12.4.8. R Function Dqcomhd --
12.5. Between-By-Within Designs --
12.5.1. R Functions bwmcp, bwamcp, bwbmcp, bwimcp, spmcpa, spmcpb, spmcpi, and bwmcppb --
12.6. Within-By-Within Designs --
12.6.1. Three-Way Designs --
12.6.2. R Functions con3way, mcp3atm, and rm3mcp --
12.6.3. Bootstrap Methods for Three-Way Designs --
12.6.4. R Functions bbwmcp, bwwmcp, bwwmcppb, bbbmcppb, bbwmcppb, bwwmcppb, and wwwmcppb --
12.7. Exercises --
ch. 13 Some Multivariate Methods --
13.1. Location, Scatter, and Detecting Outliers --
13.1.1. Detecting Outliers Via Robust Measures of Location and Scatter --
13.1.2. R Functions cov.mve and cov.mcd --
13.1.3. More Measures of Location and Covariance --
13.1.4. R Functions rmba, tbs, and ogk --
13.1.5. R Function out --
13.1.6. A Projection-Type Outlier Detection Method --
13.1.7. R Functions outpro, outproMC, outproad, outproadMC, and out3d --
13.1.8. Skipped Estimators of Location --
13.1.9. R Function smean --
13.2. One-Sample Hypothesis Testing --
13.2.1. Comparing Dependent Groups --
13.2.2. R Functions smeancrv2, hotel, and rmdzeroOP --
13.3. Two-Sample Case --
13.3.1. R Functions smcan2, mat2grp, matsplit, and mat2list --
13.3.2. R functions matsplit, mat2grp, and mat2list --
13.4. Manova --
13.4.1. R Function manova --
13.4.2. Robust MANOVA Based on Trimmed Means --
13.4.3. R Functions MULtr.anova and MULAOVp --
13.5. A Multivariate Extension of the Wilcoxon-Mann-Whitney Test --
13.5.1. Explanatory Measure of Effect Size: A Projection-Type Generalization --
13.5.2. R Function mulwmwv2 --
13.6. Rank-Based Multivariate Methods --
13.6.1. The Munzel --
Brunner Method --
13.6.2. R Function mulrank --
13.6.3. The Choi --
Marden Multivariate Rank Test