292 Index Confidence intervals (Cont.) for problem occurrence, 26 for task-time data, 29 Confidence level, 20 Continuous data, 3, 273 Continuous measure, 74 Continuous ratings, converting to discrete, 52­54 Control variability, 120­121 Controversies, 241 Conversion rates, 15, 77, 77 Correlation, 269 Criterion, 42, 47 Critical value, 283 Cronbach's alpha, see Coefficient alpha CSUQ, see Computer System Usability Questionnaire CTT, see Classical test theory Cumulative problem discovery goal, 157 CUSI, see Computer Usability Satisfaction Inventory CxPi, see Forrester Customer Experience Index F Factor analysis, 190­191, 213 confirmatory, 225­226 of data, 209, 228, 231 exploratory, 225 Fechner's Law, 217 Fidelity questionnaire, 211 Fisher exact test, 78 Fisher­Irwin test, 128 Formative conception, 105 Formative test, 10 Formative user research binomial probability model, 148­149 Magic Number 5 for, 160, 162­163 planning, 146­148, 147, 174­178 sample sizes estimation for, 143­148, 146, 179 Forrester Customer Experience Index (CxPi), 230­231 Functional magnetic resonance imaging (fMRI), 105 G D Data collection, 12 types of, 273 Data-quality errors, 74 Decision map continuous data, analyses of, 3 for sample sizes, comparing data, 4 Deflation adjustment, 153, 169­172 Demographic questionnaire, 188 Discordant pairs, 86­87, 91 Discrete data, 273 Discrete-binary data, decision map for, 3 Double-deflation, 170­171 General linear modeling, 270 Geometric mean, 31­33, 111­112, 275 Good­Turing adjustment, 150­151, 153 H Hedonic quality (HQ), 228 Heterogeneity of probability, 174 HFRG, see Human Factors Research Group HQ, see Hedonic quality Human Factors Research Group (HFRG), 190 Human­computer interaction (HCI), 255 Hypothesis testing, 118, 118, 287­288 Hypothetical usability study, 150, 152 I IBM questionnaires, positive tone in, 206 Independent proportions, 128­131 Interquartile range, 154 Interval, 242 Intranet Satisfaction Questionnaire (ISQ), 225 IRT, see Item response theory ISQ, see Intranet Satisfaction Questionnaire Item response theory (IRT), 212 E Economy, 186 Eight is not enough, 164­167 Empirical Rule, 276, 278, 283 Equivalent confidence, 126 ER, see Expectation ratings Errata sheet, 42 Errors, 14 margin of, 6, 19, 283 in statistics, 288­289 Estimating values, 108­114 Events, composite estimates of p for, 149 Exact probability, 46 Expanded binomial models, 173 Expectation ratings (ER), 215­216 Eye-tracking data analysis, 270 L Laplace method, 25 Large samples for binomial sample size estimation, 121­123 test, 49­50 Likelihood of discovery, for sample sizes, 147 Linear regression, 269 Logic of hypothesis testing, 287­288