830 Subject Index P-values 0.05 vs. 0.005, 393 in Bonferroni multiple comparison test, 620 for claims testing, 391–393 in claims testing, 406–408 in contingency tables, 591 controversy over, 381 defining, 77, 380 for goodness-of-fit, 578 in hypothesis testing, 379–381 in independent samples, 456, 457–458 interpreting, 77 Kruskal-Wallis test, 672 for linear correlation, 77–78, 678 for linear correlation coefficient, 511 in multiple regression equation, 555 in one-way ANOVA, 613–614 randomization to estimate, 491–492 rank correlation, 678 in Sign test, 648 in Standard deviation, 481 from technology, 594 without technology, 408 technology and, 515 test statistics and, 615–616 in two proportions, 443–445 in two-tailed tests, 420 in variance, 481 Wilcoxon rank-sum test, 665 Wilcoxon signed-ranks test, 658 Q Quadratic mean, 104 Qualitative data, 15 Quality control, 715 Quantiles, 123 Quantitative data, 15–16 Quartiles, 121–130 defining, 126 Questions loaded, 8 order of, 8 Queueing theory, 86–87 Queues, 755 R Randomization, 27, 28 bootstrap resampling vs., 490 in claims testing, 395–396, 410 correlation analysis using, 520–521 defining, 426, 490 in holistic statistics, 727–728 in hypothesis testing, 384 for matched pairs, 495 P-value estimation using, 491–492 software/calculator results, 496–497 for two-samples inferences, 447, 460, 490–492, 493, 496 usefulness of, 491 Randomized block design, 31–32 Random sample, 28 Random sampling error, 32–33 Random variable continuous, 206 defining, 205 discrete, 206 numerical, 206 Random variation, 704 Range, 106 Range chart. See R chart Range rule of thumb, 210, 224 for sample size, 338 for significant values, 109, 210 for standard deviation, 108–110 Range tests, 619 Rank correlation, 677–686 advantages of, 678 critical values, 678 defining, 677 disadvantage of, 678 efficiency, 678 nonlinear patterns, 681–682 notation, 677 procedure, 679 P-value, 678 requirements, 678 software/calculator results, 682–683 test statistic, 678 Ranks defining, 645–646 handling ties among, 645–646 in nonparametric tests, 645–646 Rare event rule, 152–153 hypothesis testing and, 288 for inferential statistics, 211 Ratio test, 19 R chart, 706 graphs, 706 notation, 706 requirements, 706 Real estate agents, 455 Redundancy, 164 multiplication rule and, 165–166 Region, 249 Regression, 73–79 advanced concepts, 534–538 bad models, 533, 534 basic concepts of, 529–534 correlation and, 533 defining, 78 equation, 79, 529 fitted line plot, 521 good models, 533, 534 influential points in, 535 least-squares property, 536–537 logistic, 557–559 nonlinear, 564–570 outliers in, 535 predictions from, 533–534 residuals in, 536–537 scope, 533 software/calculator results, 539 Regression coefficients, 557 Regression equation interpreting, 534–535 slope in, 530 technology and, 531 y-intercept in, 530 Regression line, 529 equation of, 530 formulas for, 530 graphing, 532 notation, 530 requirements, 530 Rejection region, 379 Relative frequency approximation, 146–147 Relative frequency distribution, 48–49 Relative frequency histogram, 55–56 Relative frequency polygon, 66 Replacements sampling with, 163, 280 sampling without, 163, 280 Replication, 28, 729 in experiment design, 28 Resampling, 425–433 bootstrap. See Bootstrap resampling claim testing using, 428–430, 492–494 with matched pairs, 494–495 randomization. See Randomization with replacement, 426 software/calculator results, 431 usefulness of, 425–426 Residual plot, 537–538 Residuals, 536–537 Resistant statistics, 89, 103 Response rates declining, 23 low, 8 Response variable, 529 Retrospective study, 30–31 Rice, arsenic in, 751 Right-tailed test, 379 Rigorously controlled design, 32 one-way ANOVA in, 619 Robocalls, 23, 339 Root mean square (RMS), 104 Rosenthal effect, 22 Rounding errors, 93 linear correlation coefficient, 511 measures of center, 93 in probabilities, 148 sample size, 338 slope, 530 two proportions, 443 y-intercept, 530 Round-off rules for coefficient of variation, 114 confidence interval, 316
RkJQdWJsaXNoZXIy NjM5ODQ=