758 Index
analytic view, 112
approximate regression coefficients, 552–553
a priori comparisons, 365, 369–384
Bonferroni t(Dunn’s test), 377
choice of coefficients, 373
Holm and Larzelere and Mulaik tests, 380
Larzelere and Mulaik test, 381
linear contrasts, 371
multiple ttests, 369
multistage Bonferroni procedures, 379
orthogonal coefficients, 375
orthogonal contrasts, 375
sum of squares for contrasts, 372
test of significance, 373
trimmed means, 383
a priori power, 239
arcsine transformation, 341
array, 264
assessing whether data are normally distributed, 76–79
assumption of independence, 152–153
assumptions, 320, 644
assumptions of chi-square, 152
assumptions underlying regression and
correlation, 264–266
asymmetric relationships, 631
average deviation, 40
axes in Q-Q plots, 77–79
backward elimination, 548
balanced designs, 332
bar charts, 67
basic laws of probability, 114
Bayes, Thomas, 123
Bayesian statistics, 127
Bayes’ theorem, 123–127
Behrens-Fisher problem, 214
Benjamini and Hochberg’s linear step up
(LSU) procedure, 397
Benjamini-Hochberg test, 396
Bernoulli trial, 127
beta, 97
bimodal distribution, 27
binomial distribution, 127–131
mean and variance of, 130–131
plotting, 128–130
to test hypotheses, 131–133
biserial and tetrachoric correlation, 303
biserial-correlation coefficient, 303
bivariate normal models, 246
Bonferroni inequality, 377
Bonferroni t (Dunn’s test), 377
bootstrapping as a general approach, 661–663
bootstrapping confidence limits on a correlation
coefficient, 670–673
bootstrapping with one sample, 663–665
boring is attractive, 403
box-and-whisker plot, 48
boxplots, 48–51calculating phi, 298
calculating rpb, 295
calculating rs, 304
calculating tau, 305–306
calculation for nested designs, 435–436
calculation of chi-square, 146–147
calculation of simple effects, 425
calculations in the analysis of variance, 324–330
degrees of freedom, 327
Fstatistic, 328
mean squares, 327
sources of variation, 327
SSerror, 326
SStotal, 326
SStreat, 326
sum of squares, 324
the summary table, 327
case-control design, 160
casewise deletion, 549
categorical data, 4
categorical data and chi-square, 139–167
4 3 2 design, 148–151
chi-square distribution, 140–141
chi-square for ordinal data, 151–152
chi-square goodness-of-fit test, one-way
classification, 141–145
dependent or repeated measurements, 153–155
effect sizes, 159–165
likelihood ratio tests, 156–157
Mantel-Haenszel statistic, 157–159
measure of agreement, 165–166
one- and two-tailed tests, 155–156
summary of the assumptions of chi-square, 152–153
two way classification variables, contingency table
analysis, 145–148
writing up the results, 167
cell, 145, 416
censored data, 564
center, 558
centering, 54
central limit theorem, 180
chi-square, 140. See alsocategorical data and chi-square
chi-square assumptions
assumption of independence, 152–153
inclusion of nonoccurrences, 153
chi-square distribution, 140–141
chi-square for ordinal data, 151–152
chi-square goodness-of-fit test, one-way
classification, 141–145
more than two categories example, 144
tabled chi-square distribution, 143–144