824 Subject Index about population proportion, 390–404, 428–429 P-value in, 391–393, 406–408 requirements, 405, 411, 417 sign test for, 410 simulations for, 191 about standard deviation, 416–425, 430 technology for, 406–408 test statistic, 405, 411, 417 about variance, 416–425, 430 Wilcoxon signed-ranks test for, 410 Class boundaries, 45, 46 midpoints, 45 size, 89 width, 45, 46 Classical approach to probability, 147 Clinical trials, 16 alternatives to, 595 observational studies compared with, 27 short cuts, 565 Clopper-Pearson method, 323 Cluster sampling, 28, 29 Coefficient confidence, 313 linear correlation, 75, 510–514 Pearson correlation, 740 Pearson product moment correlation, 510 regression, 557 Spearman’s rank correlation, 677, 743 of variation, 114 Coefficient of determination, 548 adjusted, 554–555 multiple, 554–555 Coefficient of variation (CV) defining, 114 round-off rule for, 114 Cohort study, 30–31 Coincidences, 176 Coin toss, 203–204 Coin weights, 758 Cola weights, 758 Column factor, 629 Combinations defining, 182 mnemonics for, 182 permutations and, 185 rule, 184–185 Commercial jet safety, 596 Commute times, 756 Complementary event addition rule and, 160–161 defining, 151 probability and, 151 rule of, 160 Complements of “at least one”, 171–172 multiplication rule of, 171 probability and, 171–172 Completely randomized design, 31, 619 Composite sampling, 180 Conclusions, 6–7 misleading, 7 Conditional probability, 172–175 defining, 172 formal approach for, 173 intuitive approach for, 172–173 notation, 172–173 Confidence coefficient, 313 Confidence interval, 312 analyzing polls, 318–319 better-performing, 321–323 bootstrap sample for, 335, 350, 356, 357 for claims testing, 395, 396, 405, 409, 420 Clopper-Pearson method, 323 for comparisons, 349–350 constructing, 317, 348–349 critical values, 314–315 data comparisons with, 336–337 defining, 313, 544 as equivalent method, 471 for hypothesis tests, 319, 349–350, 376, 383–384 in independent samples, 457, 459–460 interpreting, 315–316, 331 manual calculation, 318 margin of error and, 336–337, 461 for matched pairs, 471 for mean, 619 notation, 316 one-sided, 329 plus four method, 322 point estimate and, 319, 336–337 for population mean, 329–330 for population proportion, 316–317 for population standard deviation, 348 population variance, 348 procedure for, 332–333 rationale for, 350–351, 447–448 requirement check, 316, 317 round-off rule, 316 software/calculator results, 351–352, 448–449 technology and, 317 for two dependent samples, 471 in two proportions, 443, 446–448 Wald, 321–322 Wilson score, 322 Confidence level defining, 313 process success rate, 315–316 Confounding, 31 Confusion of the inverse, 174–175 Constitution, 47 Context, 3–5 Contingency tables alternative hypothesis, 591 basic concepts, 590–595 critical thinking, 600–603 critical values in, 591 defining, 590 degrees of freedom in, 591 expected frequencies in, 592 expected value in, 592–593 notation, 591 null hypothesis, 591 observed frequencies in, 592 P-values in, 591 requirements, 591 software/calculator results, 599 test statistic in, 591 2 * 2, 594 Continuity correction, 303 Continuous data, 16–17 Continuous random variable, 206 Control charts, 704–710 for attributes, 713–716 constants, 707 defining, 704 graphs, 706, 709 interpreting, 705–706 for mean, 708–709 notation, 706, 708 for p, 713–716 requirements, 706, 708 software/calculator results, 710 for variation, 706–708 Convenience sampling, 28, 29 Convictions, 172 Correlation, 73–79 basic concepts of, 508–518 causation and, 7, 73, 517–518 common errors involving, 518 defining, 73, 508 determining, 75–77 formal hypothesis test and, 518–520 linear, 73, 77–78, 508–517, 678 negative, 509 nonlinear relationship in, 518 positive, 509 randomization test for, 520–521 rank, 677–686 regression and, 533 software/calculator results for, 521–522 spurious, 516–517 Count five for standard deviation, 484 for variance, 484 Course evaluations, 755 Coverage probability, 321–322 COVID-19 virus pandemic, 566 Critical region, 379 Critical thinking contingency tables, 600–603 goodness-of-fit and, 586–590 Kruskal-Wallis test, 675–676 one-way ANOVA, 623–625 sign test, 655–657 standard deviation for, 224–226 in two-way ANOVA, 634–636 Wilcoxon rank-sum test, 668–670 Critical values, 734 in Bonferroni multiple comparison test, 620 Claims testing (continued)
RkJQdWJsaXNoZXIy NjM5ODQ=