.

0 (hodnocen0 x )

BK

4th ed.

Oxford : Oxford University Press, 2008

xvii, 431 s. : il. ; 24 cm

ISBN 978-0-19-923188-1 (brož.)

Obsahuje bibliografické odkazy a rejstřík

000190987

Contents // 1. Introduction 1 // 2. Basic concepts 5 Searching the literature 6 Critical review 6 Empirical forms of validity 9 // The two traditions of assessment 11 The reduction of measurement error 14 Summary 15 // 3. Devising the items 17 The source of items 18 Content validity 23 // Generic versus specific scales and the ‘fidelity versus bandwidth’ issue 27 Translation 29 // 4. Scaling responses 37 Introduction 37 Some basic concepts 37 Categorical judgements 38 Continuous judgements 40 To rate or to rank 66 Multidimensional scaling 68 // 5. Selecting the items 77 Interpretability 77 Face validity 82 // Frequency of endorsement and discrimination 83 Homogeneity of the items 85 Multifactor inventories 96 When homogeneity does not matter 97 Putting it all together 98 // 6. Biases in responding 103 The differing perspectives 103 // Answering questions: the cognitive requirements 104 Optimizing and satisficing 108 // xiv CONTENTS // Social desirability and faking good 110 Deviation and faking bad 115 Yea-saying or acquiescence 118 End-aversion, positive skew, and halo 119 Framing 122 // Biases related to the measurement of change 123 Estimates of the prior state — implicit theory of change 125 Reconciling the two positions 125 Proxy reporting 126 Testing the items 127 // 7. From items to scales 135 Weighting the items 135 Missing items 139 Multiplicative composite scores 140 Transforming the final score 143 Percentiles 144 // Standard and standardized scores 146 Normalized

scores 149 Age and sex norms 149 Establishing cut points 151 // Methods based on characteristics of the distribution 152 // Methods based on judgement 154 // Absolute methods 154 // Receiver operating characteristic curves 156 // Summary 163 // 8. Reliability 167 Basic concepts 167 Philosophical implications 170 Terminology 173 // Defining reliability 174 // Other considerations in calculating the reliability of a test 177 // The observer nested within subject 179 // Multiple observations 180 // Other types of reliability 182 // Different forms of the reliability coefficient 183 // Kappa coefficient versus the ICC 188 // The method of Bland and Altman 190 // Issues of interpretation 190 // Improving reliability 196 // CONTENTS XV // Standard error of the reliability coefficient and sample size 198 // Reliability generalization 202 // The average value of r and a 203 // The variance of the reliability estimates 204 // Combining estimates 205 // Factors affecting the reliability 206 // Summary 207 // 9. Generalizability theory 211 // Generalizability theory fundamentals 213 // An Example 214 // The First Step—the ANOVA 215 // Step 2—From ANOVA to G coefficients 218 // Relative vs. Absolute Error 219 // Equivalent for the nested design 222 // Generalizability of an average 222 // Step 3 - from G study to D study 223 // ANOVA for statisticians and ANOVA for psychometricians 224 // Confidence intervals for G coefficients 225 // The general rules to compute G coefficients 226

Getting the computer to do it for you 227 // Some Common Examples 228 // Uses and abuses of G theory 244 // Summary 245 // 10. Validity 247 // Why assess validity? 247 Reliability and validity 248 A history of the ‘types’ of validity 249 Content validation 252 Criterion validation 254 Construct validation 257 Construct validational studies 258 Extreme groups 261 // Convergent and discriminant validation 262 Consequential validation 263 The multitrait-multimethod matrix 264 Summary 265 // Responsiveness and sensitivity to change 266 // xvi CONTENTS // Validity and ‘types of indices’ 267 Biases in validity assessment 26S Unreliability of the criterion 27\\ // Changes in the sample 273 Validity generalization 274 Summary 274 // 11. Measuring change 277 Introduction 277 // The goal of measurement of change 277 Why not measure change directly? 27S // Measures of association—reliability and sensitivity to change 2S0 Difficulties with change scores in experimental designs 285 Change scores and quasi-experimental designs 286 Measuring change using multiple observations: growth curves 288 How much change is enough? 293 Summary 295 // 12. Item response theory 299 Problems with classical test theory 299 // The introduction of item response theory 301 // Item characteristic curves 302 // The one-parameter model 304 // The two- and three-parameter models 306 // Polytomous models 309 // Item information 312 // Item fit 313 // Person fit 315 // Differential item functioning 315

Unidimensionality and local independence 316 // The standard error of measurement 320 // Equating tests 321 // Sample size 322 // Mokken scaling 323 // Advantages 324 // Disadvantages 326 // Computer programs 327 // 13. Methods of administration 331 Face-to-face interviews 331 Advantages 331 // CONTENTS xvi i // Disadvantages 332 Telephone questionnaires 334 Random digit dialling 336 Advantages 337 Disadvantages 338 Mailed questionnaires 340 The necessity of persistence 346 Computer-assisted administration 348 Using e-mail and the Web 351 Personal data assistants 354 Reporting response rates 356 // 14. Ethical considerations 365 // 15. Reporting test results 373 // Standards for Educational and Psychological Testing 374 The STARD initiative 376 Summary 379 // Appendices A Further reading 381 ? Where to find tests 387 // C A (very) brief introduction to factor analysis 409 Author Index 415 Subject Index 423