THE

© PSYCHOLOGICAL
: CORPORATION*
A Harcourt Assessment Company
Copy +86
WECHSLER ADULT INTELLIGENCE SCALE - THIRD EDITION

Sue,

WECHSLER MEMORY SCALE — THIRD EDITION

TECHNICAL
MANUAL

THE
PSYCHOLOGICAL
*/ CORPORATION"
ee
tn
A Harcourt Assessment Company
Copyright © 2002 by The Psychological Corporation
Normative data copyright © 1997 by The Psychological Corporation
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval
system, without permission in writing from the publisher.
The Psychological Corporation and the PSI logo are registered trademarks of The Psychological Corporation.
Wechsler Adult Intelligence Scale and WAIS are registered trademarks of The Psychological Corporation.
Wechsler Memory Scale and WMS are registered trademarks of The Psychological Corporation.
Portions of this work were previously published.
Printed in the United States of America

2.354556
7 899 10 1112 AB © Dee
 SPECIAL ACKNOWLEDGMENT

After the decision was made to revise the WAIS-R and the WMS-R instru-
ments, The Psychological Corporation asked Dr. Nelson Butters to form an
advisory panel of individuals recognized as scientists—practitioners. The pur-
pose of the panel was to provide consultation regarding content revision and
research design as well as direction related to the integration of these two
instruments. The core members of this panel were Drs. Robert Bornstein,
Gordon Chelune, Robert Heaton, and Robert Ivnik. The advisory panel
reviewed our progress on the development of the tests at key points and
formally met approximately twice yearly to provide consultation regarding
possible revisions. Unfortunately, Dr. Butters became ill early on in the
project and his untimely death prevented his seeing these tests published.
Nevertheless, his contributions can be clearly seen in the final versions of
the WAIS-III and the WMS-III. Although it was not feasible to incorporate
all of the panel's suggestions into the final scales, their contributions have
greatly enhanced the WAIS-III and the WMS-III. We are exceedingly indebt-
ed to our advisory panel and want to formally acknowledge their invaluable
contributions.
David Tulsky and Jianjun Zhu, WAIS-III Project Directors
Mark E Ledbetter, WMS-III Project Director
 Digitized by the Internet Archive
in 2023 with funding from
Kahle/Austin Foundation

[Link]
 CONTENTS
Special Acknowledgment

Chapter 1. Introduction
Concept of Intelligence
Concept of Memory and Learning
Concept of Working Memory
Antecedents of the WAIS-III and the WMS-III
Development of the Scales
WAIS-HUI
WMS-III
Normative/Psychometric Issues
Content/Scoring Configuration Issues

Chapter 2. Standardization and Norms Development

Standardization
Description of the Samples
Locating and Testing the Samples
Representativeness of the Samples
WMS-III Weighted Sampling
Quality-Control Procedures
Examiner Qualification Procedures
Scoring Studies
Quality Assurance of Scoring and Data Entry
Other Quality-Assurance Procedures
Norms Development
Derivation of Subtest Scaled Scores
Conversion of Scaled Scores to IQ and Index Scores
Derivation of IQ and Index Scores
Age-Adjusted and Reference-Group Norms

Chapter 3. Reliability and Score Differences

Reliability
Reliability Coefficients
Standard Errors of Measurement and Confidence Intervals
Test—Retest Stability
Interscorer Agreement
Score Differences
Differences Between IQ Scores and Index Scores
Statistical Significance of IQ and Index Score Differences
Frequency of IQ and Index Score Differences
Differences Between a Single Subtest Score and an Average of
Subtest Scores
Significance of Differences Between a Single Subtest Score
and an Average of Subtest Scores
 Frequency of Differences Between a Single Subtest Score 70
and an Average of Subtest Scores
Differences Between Subtest Scores 71
Statistical Significance of Differences Between Subtest ek
Scaled Scores
Intersubtest Scatter he
Discrepancies Between Digit Span Forward and Backward 72
Summary ts

Chapter 4. Evidence Base for Validity of the WAIS-IIT and WMS-II 75

Evidence Based on Internal Structure 75
WAIS-III Intercorrelations 76
WMS-III Intercorrelations 80
WAIS-III Factor Analytic Studies 83
Exploratory Factor Analysis 84
Confirmatory Factor Analysis 90
Composition of the Index Scores 93
Development of a Short Version of the WAIS—III 94
Development of a General Ability Index 95
Wechsler Abbreviated Scale of Intelligence 95
WMS-III Factor Analytic Studies 96
Correlation Between the WAIS-III and the WMS-III 102
Joint WAIS—HI/WMS-III Factor Analysis 105
Convergent Evidence for the WAIS-III as a
Measure of Intellectual Functioning 106
Correlations With the WAIS-R 106
Correlations With the WISC-III 109
Correlations With the Standard Progressive Matrices itt
Correlations With the Standford-Binet Intelligence Scale—
Fourth Edition 1i2
Summary we
Convergent Evidence for the WMS-III as a Measure of
Memory Functioning 114
Correlations with the WMS-R 114
Correlations with the Children’s Memory Scale 118
Evidence Based on Test-Criterion Relationships 118
Assessment of Psychoeducational and Developmental Disorders 118
Mental Retardation 118
Attention-Deficit/Hyperactivity Disorder 122
Learning Disabilities 126
Hearing Deficiencies 129
Summary 130
 Assessment of Neurological Disorders and Disorders
Associated With Dementia 131
Alzheimer’s Disease 131
Huntington's Disease 136
Parkinson's Disease 138
Traumatic Brain Injury 141
Multiple Sclerosis 144
Temporal Lobe Epilepsy 145
Assessment of Alcohol-Related Disorders 149
Chronic Alcohol Abuse 150
Korsakoff's Syndrome 152
Assessment of Neuropsychiatric Disorders—Schizophrenia 154
Evidence Based on Relationships With Other Variables 157
WAIS-III Correlations With Selected External Measures 157
Measures of Academic Achievement 157
Measures of Cognitive Ability 160
Measures of Attention and Concentration 163
Measures of Memory 165
Measures of Language 167
Measures of Fine Motor Speed and Fine Motor Dexterity 169
Measures of Spatial Processing 169
Measures of Executive Functioning 172
WMS-II Correlations With Selected External Measures 173
Measures of Cognitive Ability 176
Measures of Attention and Concentration 178
Measures of Memory 180
Measures of Language 182
Measures of Spatial Processing 183
Measure of Executive Functioning 184
Measures of Fine Motor Speed and Fine Motor Dexterity 185
Demographic Studies 187
Prediction of Premorbid Intellectual and Memory Functioning 189
Malingering Studies ife
Sequential Assessment 192
Other Issues 192

Chapter 5. Interpretive Considerations 195

Scores and Descriptive Classifications 195
Use of Confidence Intervals Around Scores 196
Normative Reference Groups 196
Level of Performance 198
WAIS-III 199
Description of IQ Scores 199

vii
 Description of Index Scores 200
IQ Scores Versus Index Scores 201
Optional Procedures of the WAIS-III 201
Discrepancies Between Digit Span Forward and Backward 201
Digit Symbol Optional Procedures 202
WMS-III 203
Primary Indexes 204
Auditory Immediate and Auditory Delayed Indexes 204
Visual Immediate and Visual Delayed Indexes 205
Immediate Memory Index 206
Auditory Recognition Delayed Index 206
General Memory Index 207
Working Memory Index 208
Auditory Process Composites 208
Single-Trial Learning Composite 209
Learning Slope Composite 209
Retention Composite 210
Retrieval Composite 211
Subtests vi
Information and Orientation 211
Logical Memory fi
Verbal Paired Associates ra ee
Word Lists 214
Faces 21>
Family Pictures 245
Visual Reproduction 2415
Letter-Number Sequencing 216
Spatial Span 216
Digit Span 218
Mental Control 218
Patterns and Profiles of Performance 218
WAIS-III Discrepancy Analyses 220
Subtest Score Patterns and Discrepancies nad |
IQ Score and Index Score Discrepancies onl
Directional Discrepancy Scores Deo
Other Score Discrepancies 224
WMS-III Discrepancy Analyses 224
Primary Index Score Comparisons 224
Immediate Versus Delayed Indexes 224
Auditory Versus Visual Indexes 224
Working Memory Versus Immediate Memory Indexes (Gas
Auditory Delayed Index Versus Auditory Recognition 229
Delayed Index
 Auditory Process Composite Score Comparisons oo
Single-Trial Learning Composite Versus Learning Slope ped
Composite
Retention Composite Versus Auditory Delayed Index 226
Differences Between the WAIS-III and the WMS-III 226
Simple-Difference Method 228
Predicted-Difference Method pias)
Differences Between the WAIS-III and the WIAT-II 230
Simple-Difference Method Zod
Predicted-Difference Method 21

Appendix A. Intercorrelation Tables for the WAIS—III and the WMS-III 23a

Appendix B. Discrepancy Score Tables Based on

Predicted-Difference Method ake

Appendix C. Discrepancy Score Tables Based on

Simple-Difference Method 339

Appendix D. Frequencies of WAIS-III IQ Score Discrepancies

Based on Ability Level 363

Appendix E. WAIS-III Digit Span Backward Scaled-Score

Equivalents of Raw Scores 381

Appendix E Inclusion and Exclusion Criteria for

Participation in Special Group Studies 383

Appendix G. Reviewers, Examiners, and Participating

Clinics and Organizations a9li

References 401
Tabies
Table 2.1. Exclusionary Criteria for the Standardization Samples 21

Table 2.2. Percentages of the U.S. Population by

Age, Race/Ethnicity, and Education 24

Table 2.3. Demographic Characteristics of the WAIS—III

Standardization Sample: Percentages by
Age, Race/Ethnicity, and Education p25.

Table 2.4. Demographic Characteristics of the WMS-III

Standardization Sample: Percentages by
Age, Race/Ethnicity, and Education 26

Table 2.5. Percentages of the U.S. Population

by Age, Sex, and Education 27
Table 2.6. Demographic Characteristics of the WAIS—III
Standardization Sample: Percentages by
Age, Sex, and Education 28
Table 2.7. Demographic Characteristics of the WMS-III
Standardization Sample: Percentages by
Age, Sex, and Education 29
Table 2.8. Percentages of the U.S. Population by
Age, Sex, and Race/Ethnicity 30
Table 2.9. Demographic Characteristics of the WAIS-III
Standardization Sample: Percentages by
Age, Sex, and Race/Ethnicity ae
Table 2.10. Demographic Characteristics of the WMS-III
Standardization Sample: Percentages by
Age, Sex, and Race/Ethnicity 32
Table 2.11. Percentages of the U.S. Population by
Age, Race/Ethnicity, and Geographic Region 5
Table 2.12. Demographic Characteristics of the WAIS-III
Standardization Sample: Percentages by
Age, Race/Ethnicity, and Geographic Region 34
Table 2.13. Demographic Characteristics of the WMS-III
Standardization Sample: Percentages by
Age, Race/Ethnicity, and Geographic Region 35
Table 2.14. WAIS-III Subtest Composition of IQ and Index Scores 43
Table 2.15. WMS-III Subtest Composition of Index and
Composite Scores 44
Table 3.1. Reliability Coefficients of the WAIS-III Subtests, IQ
Scales, and Indexes by Age Group 50
Table 3.2. Reliability Coefficients of the WMS-III Primary
Subtest Scores and Primary Indexes by Age Group
Table 3.3. Generalizability Coefficients and Standard Errors of
Measurement of Selected WMS-III Supplemental
Subtest Scores
Table 3.4. Standard Errors of Measurement of the WAIS-III Subtest,
IQ, and Index Scores by Age Group
Table 3.5. Standard Errors of Measurement of the WMS-III Primary
Subtest Scores and Primary Indexes by Age Group
Table 3.6. Stability Coefficients of the WAIS-III Subtests, IQ Scales,
and Indexes: Age Group 16-29
Table 3.7. Stability Coefficients of the WAIS-III Subtests, IQ Scales,
and Indexes: Age Group 30-54
Table 3.8. Stability Coefficients of the WAIS-III Subtests, IQ Scales,
and Indexes: Age Group 55-74
Table 3.9. Stability Coefficients of the WAIS-III Subtests, IQ Scales,
and Indexes: Age Group 75-89
Table 3.10. Stability Coefficients of the WMS-III Primary Subtest
Scores and Primary Indexes: Age Group 16-54
Table 3.11. Stability Coefficients of the WMS-III Primary Subtest
Scores and Primary Indexes: Age Group 55-89
Table 3.12. Stability Coefficients of the WMS-III Supplemental
Subtest Scores
Table 4.1. Intercorrelations of WAIS—III Subtest Scaled Scores and
Sums of Scaled Scores for IQ Scales and Indexes: All Ages
Table 4.2. Intercorrelations of WMS-III Primary Subtest Scaled
Scores: All Ages
Table 4.3. Intercorrelations of WMS-III Indexes and Composites:
All Ages
Table 4.4. Predicted Factor Structure of the WAIS-III

Table 4.5. WISC-III Subtest Set: Exploratory Factor Pattern

Loadings for Four-Factor Solutions
Table 4.6. WAIS-III Subtest Set: Exploratory Factor Pattern
Loadings for Four-Factor Solutions 86

Table 4.7. WAIS-III Factor Pattern Loadings for Verbal

Comprehension Factor by Five Age Bands 87

Table 4.8. WAIS-III Factor Pattern Loadings for Perceptual

Organization Factor by Five Age Bands 88

Table 4.9. WAIS-III Factor Pattern Loadings for Working

Memory Factor by Five Age Bands 88

Table 4.10. WAIS-III Factor Pattern Loadings for Processing

Speed Factor by Five Age Bands 89

Table 4.11. WAIS-III Goodness-of-Fit Statistics for Confirmatory

Factor Analysis 92

Table 4.12. WMS-III Calibration (Standardization) Sample

Goodness-of-Fit Statistics for Confirmatory
Factor Analyses 99
Table 4.13. WMS-III Cross-Validation Sample Goodness-of-Fit
Statistics for Confirmatory Factor Analyses 100
Table 4.14. Standardized Parameter Estimates for Model 4 101
Table 4.15. Interfactor Correlations for the Calibration and
Cross-Validation Samples for Model 4 101
Table 4.16. WMS-III Invariance Statistics for Model 4 for the
Calibration and Cross-Validation Samples 102
Table 4.17. Correlations Between the WAIS-III and the WMS-III 104
Table 4.18. Correlations Between the WAIS-R and the WAIS-III 107
Table 4.19. Expected WAIS-III IQ Scores for Selected WAIS-R IQ Scores 108
Table 4.20. Correlations Between the WISC-III and the WAIS-III 109
Table 4.21. Expected WAIS-III IQ Scores for Selected WISC-III IQ Scores 110
Table 4.22. Correlations Between the SPM and the WAIS-III 111
Table 4.23. Correlations Between the SB-IV and the WAIS-III LTS
Table 4.24. Correlations Between the WMS-R and the WMS-III WARS)
Table 4.25. Correlations Between the CMS and the WMS-III Lig
Table 4.26. Demographic Data of Samples With Psychoeducational
and Developmental Disorders for WAIS-III Studies 119
Table 4.27. WAIS-III Performance of Samples With
Psychoeducational and Developmental Disorders Lio
Table 4.28. Demographic Data of Samples With Psychoeducational
and Developmental Disorders for WMS-III Studies 120
Table 4.29. WMS-III Performance of Samples With
Psychoeducational and Developmental Disorders 120
Table 4.30. WAIS-III Performance of Individuals With
Deafness/ Hearing Deficiencies 130
Table 4.31. Demographic Data of the Samples With Neurological,
Alcohol-Related, and Neuropsychiatric Disorders 132
Table 4.32. WAIS-III and WMS-III Performance of Samples With
Neurological Disorders 133
Table 4.33. WAIS-III and WMS-III Performance of Samples With
Temporal Lobe Epilepsy 148
Table 4.34. WAIS-III and WMS-III Performance of Samples
With Alcohol-Related Disorders 149
Table 4.35. WAIS-III and WMS-III Performance of Samples With
Schizophrenia 154

Table 4.36. Demographic Data of the Samples for the

WAIS-III/External Measures Comparison Studies 158

Table 4.37. Intercorrelations of WAIS-III IQ Scores and the WIAT-II

Subtest and Composite Standard Scores 161

Table 4.38. Correlations Between the WAIS-—III and Other Measures

of Cognitive Ability 162

Table 4.39. Correlations Between the WAIS-III and Measures

of Attention and Concentration 164

Table 4.40. Correlations Between the WAIS-III and Measures

of Memory 166

Table 4.41. Correlations Between the WAIS-III and Measures

of Language 168

Table 4.42. Correlations Between the WAIS-III and Measures

of Fine Motor Speed and Fine Motor Dexterity 170

Table 4.43. Correlations Between the WAIS-III and Measures

of Spatial Processing 171
 Table 4.44. Correlations Between the WAIS-—III and Measures
of Executive Functioning 1f2

Table 4.45. Demographic Data of the Samples for the

WMS-III/External Measures Comparison Studies 174

Table 4.46. Correlations Between the WMS-III and Measures

of Cognitive Ability 1¥%

Table 4.47. Correlations Between the WMS-III and Measures

of Attention and Concentration 179

Table 4.48. Correlations Between the WMS-III and Other Measures

of Memory 181

Table 4.49. Correlations Between the WMS-III and Measures

of Language 182

Table 4.50. Correlations Between the WMS-III and Measures

of Spatial Processing 183

Table 4.51. Correlations Between the WMS-III and Measures

of Executive Functioning 185
Table 4.52. Correlations Between the WMS-III and Measures
of Fine Motor Speed and Fine Motor Dexterity 186
Table 5.1. Relation of IQ and Index Scores to Standard Deviations
From the Mean and Percentile Rank Equivalents 197
Table 5.2. Qualitative Descriptions of IQ and Index Scores 198
Table 5.3. Frequencies (Cumulative Percentages) of Information
and Orientation Total Scores by Age 212
Table 5.4. Frequencies (Cumulative Percentages) of Visual
Reproduction Discrimination Total Scores by Age 217
Table A.1. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 16-17 Years 234
Table A.2. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 18-19 Years 235
Table A.3. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 20-24 Years 236

xiv
Table A.4. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 25-29 Years oT
Table A.5. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 30-34 Years 238
Table A.6. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 35-44 Years 239
Table A.7. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 45-54 Years 240
Table A.8. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 55-64 Years 241
Table A.9. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 65-69 Years 242
Table A.10. Intercorrelations of WAIS-III Subtest Scaled Scores
and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 70-74 Years 243

Table A.11. Intercorrelations of WAIS-—III Subtest Scaled Scores

and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 75-79 Years 244

Table A.12. Intercorrelations of WAIS-III Subtest Scaled Scores

and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 80-84 Years 245

Table A.13. Intercorrelations of WAIS-III Subtest Scaled Scores

and Sums of Scaled Scores for IQ Scales and Indexes:
Ages 85-89 Years 246

Table A.14. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 16-17 Years 247

Table A.15. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 18-19 Years 248

Table A.16. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 20-24 Years 249

xv
Table A.17. Intercorrelations of WMS-III Primary Subtest Scaled
Scores: Ages 25-29 Years 250

Table A.18. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 30-34 Years 251

Table A.19. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 35-44 Years 252

Table A.20. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 45-54 Years 20a

Table A.21. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 55-64 Years 254

Table A.22. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 65-69 Years 255

Table A.23. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 70—74 Years 256

Table A.24. Intercorrelations of WMS-III Primary Subtest Scaled

Scores: Ages 75-79 Years 257
Table A.25. Intercorrelations of WMS-III Primary Subtest Scaled
Scores: Ages 80-84 Years 258
Table A.26. Intercorrelations of WMS-III Primary Subtest Scaled
Scores: Ages 85-89 Years 259
Table A.27. Intercorrelations of WMS-III Indexes and Composites:
Ages 16-17 Years 260
Table A.28. Intercorrelations of WMS-III Indexes and Composites:
Ages 18-19 Years 261
Table A.29. Intercorrelations of WMS-III Indexes and Composites:
Ages 20-24 Years 262
Table A.30. Intercorrelations of WMS-III Indexes and Composites:
Ages 25-29 Years 263
Table A.31. Intercorrelations of WMS-III Indexes and Composites:
Ages 30-34 Years 264
Table A.32. Intercorrelations of WMS-III Indexes and Composites:
Ages 35-44 Years 265
Table A.33. Intercorrelations of WMS-III Indexes and Composites:
Ages 45-54 Years 266
Table A.34. Intercorrelations of WMS-III Indexes and Composites:
Ages 55-64 Years 267
Table A.35. Intercorrelations of WMS-III Indexes and Composites:
Ages 65-69 Years 268
Table A.36. Intercorrelations ofWMS-III Indexes and Composites:
Ages 70-74 Years 269
Table A.37. Intercorrelations of WMS-III Indexes and Composites:
Ages 75-79 Years 270
Table A.38. Intercorrelations of WMS-III Indexes and Composites:
Ages 80-84 Years zt
Table A.39. Intercorrelations of WMS-III Indexes and Composites:
Ages 85-89 Years Zia
Table A.40. WMS-III Intercorrelations of Primary and
Supplemental Subtest Scaled Scores: Ages 16-29 years 273
Table A.41. WMS-III Intercorrelations of Primary and
Supplemental Subtest Scaled Scores: Ages 30-64 years Aas
Table A.42. WMS-III Intercorrelations of Primary and
Supplemental Subtest Scaled Scores: Ages 65-89 years 2nt
Table B.1. WMS-III Index Scores Predicted From WAIS-III FSIQ
Scores: All Ages 280
Table B.2. WMS-III Index Scores Predicted From WAIS-III VIQ
Scores: All Ages 282

Table B.3. WMS-III Index Scores Predicted From WAIS-III PIQ

Scores: All Ages 284

Table B.4. Differences Between Predicted and Obtained WMS-III

Index Scores Required for Statistical Significance by Age
(Predicted-Difference Method—WAIS-III FSIQ) 286

Table B.5. Differences Between Predicted and Obtained WMS-III

Index Scores Required for Statistical Significance by Age
(Predicted-Difference Method—WAIS-III VIQ) 287

Table B.6. Differences Between Predicted and Obtained WMS-III

Index Scores Required for Statistical Significance by Age
(Predicted-Difference Method—WAIS-III PIQ) 288

Table B.7. Frequencies (Cumulative Percentages) of Differences

Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III FSIQ)
Table B.8. Frequencies (Cumulative Percentages) of Differences
Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III VIQ) 289

Table B.9. Frequencies (Cumulative Percentages) of Differences

Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III PIQ) 289

Table B.10. WIAT-II Subtest and Composite Standard Scores

Predicted From WAIS-III FSIQ Scores 290

Table B.11. WIAT-II Subtest and Composite Standard Scores

Predicted FromWaAIS-III VIQ Scores 296

Table B.12. WIAT-II Subtest and Composite Standard Scores

Predicted From WAIS-III PIQ Scores 302

Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted

From WAIS-III VCI Scores 308

Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted

From WAIS-III POI Scores 314

Table B.15. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Required for Statistical
Significance With Prediction Based on WAIS-III Scores 320
Table B.16. Differences Between Predicted and Actual WIAT-II Subtest
and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III FSIQ Scores 323
Table B.17. Differences Between Predicted and Actual WIAT-II Subtest
and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VIQ Scores 326
Table B.18. Differences Between Predicted and Actual WIAT-II Subtest
and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III PIQ Scores PAs)
Table B.19. Differences Between Predicted and Actual WIAT-II Subtest
and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VCI Scores So2
Table B.20. Differences Between Predicted and Actual WIAT-II Subtest
and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III POI Scores 335
Table C.1. Differences Between WAIS-III FSIQ Scores and WMS-III
Index Scores Required for Statistical Significance by Age
(Simple-Difference Method) 340
Table C.2. Differences Between WAIS-III VIQ Scores and WMS-III
Index Scores Required for Statistical Significance by Age
(Simple-Difference Method) 341
Table C.3. Differences Between WAIS-III PIQ Scores and WMS-III
Index Scores Required for Statistical Significance by Age
(Simple-Difference Method) 342
Table C.4. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III FSIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method) 343
Table C.5. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III VIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method) 343
Table C.6. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III PIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method) 343
Table C.7. Differences Between WAIS-III Scores and Actual WIAT-II
Subtest and Composite Standard Scores Required for
Statistical Significance (Simple-Difference Method) 344
Table C.8. Differences Between WAIS-III FSIQ Scores and Actual
WIAT-II Subtest and Composite Standard Scores
Obtained by Various Percentages of the
WIAT-II/WAIS-III Linking Sample 347
Table C.9. Differences Between WAIS-III VIQ Scores and Actual
WIAT-II Subtest and Composite Standard Scores
Obtained by Various Percentages of the
WIAT-II/WAIS-III Linking Sample 350
Table C.10. Differences Between WAIS-III PIQ Scores and Actual
WIAT-II Subtest and Composite Standard Scores
Obtained by Various Percentages of the
WIAT-II/WAIS-III Linking Sample 353

Table C.11. Differences Between WAIS-III VCI Scores and Actual

WIAT-II Subtest and Composite Standard Scores
Obtained by Various Percentages of the
WIAT-II/WAIS-III Linking Sample 356

xix
 Table C.12. Differences Between WAIS-III POI Scores and Actual
WIAT-II Subtest and Composite Standard Scores
Obtained by Various Percentages of the
WIAT-II/WAIS-III Linking Sample 350

Table D.1. Frequencies (Cumulative Percentages) of Differences

Between WAIS-III IQ and Index Scores by Ability Level:
FSIQ <79 364

Table D.2. Frequencies (Cumulative Percentages) of Differences

Between WAIS-III IQ and Index Scores by Ability Level:
FSIQ 80-89 366
Table D.3. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III IQ and Index Scores by Ability Level:
FSIQ 90-109 368
Table D.4. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III IQ and Index Scores by Ability Level:
FSIQ 110-119 370
Table D.5. Frequencies (Cumulative Percentages) of Differences
Between WAIS-III IQ and Index Scores by Ability Level:
FSIQ >120 3i2
Table D.6. Frequencies of Directional Discrepancies
Between WAIS-III IQ and Index Scores 374
Table D.7. Directional Cumulative Frequencies for WMS-III
Index Score Differences 3/9
Table E.1. WAIS-II Digit Span Backward Scaled-Score
Equivalents of Raw Scores 382

Figures
Figure 2.1: Standardization Sampling Sites 20
Figure 2.2. Race/Ethnicity Characteristics of the U.S. Population
and WAIS—III and WMS-III Standardization Samples 23
Figure 2.3. Geographic Region Characteristics of the U.S. Population
and WAIS-III and WMS-III Standardization Samples 228)

XX
 CHAPTER 1

Introduction

In clinical practice, measures of intellectual functioning and memory are

often administered concurrently so that a broad spectrum of cognitive abili-
ties can be examined. In view of this purpose, the Wechsler Adult Intelligence
Scale—Third Edition (WAIS-III; Wechsler, 1997a) and the Wechsler Memory
Scale—Third Edition (WMS-III; Wechsler, 1997b) were codeveloped and
share similar research methodologies, normative samples, and similar
clinical validation procedures. As a result, these two instruments provide
a means of assessing a broad range of cognitive abilities and now allow
for more meaningful comparisons between intellectual ability and
memory functioning.
This update of the WA/S—IIJ—WMS-III Technical Manual contains the psy-
chometric, technical, and basic interpretive information for the WAIS-III
and WMS-III. This manual includes the following updated material:
e Predicted achievement scores and discrepancy tables for the Wechsler
Individual Achievement Test-Second Edition (WIAT-II; The Psychological
Corporation, 2002a, 2002b, 2002c)
¢ Norms for Digit Span Backward
¢ Bidirectional cumulative frequency distributions for WAIS—III
and WMS-III Index scores
¢ New factor analytic studies of the WMS-III
e Review of recent research studies and clinical methods related
to the WAIS-III/WMS-III and
¢ demographically adjusted norms
e the Wechsler Test ofAdult Reading (The Psychological
Corporation, 2001)
e short forms of the WAIS-III
e the Wechsler Abbreviated Scale of Intelligence (WASI; The
Psychological Corporation, 1999)
¢ serial/sequential assessment
¢ malingering studies
¢ clinical studies
¢ factor analytic studies of the combined WAIS—III and WMS—III
Introduction

Chapter 1 of this Manual reviews the basic theories underlying intelligence

and memory. Chapter 2 describes the sampling and data-handling proce-
dures, and Chapter 3 presents data relevant to the reliability of the two
instruments. Chapter 4 presents evidence of the validity and clinical utility
of the WAIS-III and WMS-III, including results from studies conducted since
their original publication. Chapter 5 presents guidelines for interpreting the
various scores obtained from the two scales. The appendixes provide addi-
tional information, such as intercorrelation tables and normative data for
intellectual ability-memory discrepancies and intellectual
ability-academic achievement discrepancies.

Concept of Intelligence
The concept of intelligence has been a hotly debated topic since the turn of
the century. Wechsler took a more ecological approach and conceived of
intelligence as a multidimensional construct, one that manifests itself in
many forms. He originally defined intelligence as the “capacity of the
individual to act purposefully, to think rationally, and to deal effectively with
his environment” (1944, p. 3). He considered intelligence not only as a global
entity but also as an aggregate of specific abilities. Wechsler explained that
intelligence is global because it characterizes the individual’s behavior as a
whole. It is also specific because it is composed of elements or abilities that
are qualitatively different.
Wechsler maintained this definition of intelligence throughout his career.
He believed that intelligence should be measured by both verbal and perfor-
mance tasks, each of which measured ability in a different way and which
could be aggregated to form a general, global construct. However, particu-
larly later in his career, Wechsler began exploring “nonintellective” factors of
intelligence, including the abilities to perceive and respond to social, moral,
and aesthetic values (Wechsler, 1975). Wechsler was keenly aware that the
results of factor analytic studies accounted for only a percentage of the over-
all variance of intelligence, and he believed that another group of attributes
contributed to this unexplained variance. According to Wechsler, these attri-
butes are made up of basic human motivations, attitudes, and personality
traits, such as persistence, goal awareness, enthusiasm, and other conative
dispositions not tapped directly by existing measures of intellectual ability.
Wechsler also hypothesized that these attributes influence an individual’s
performance on such measures, as well as the individual’s effectiveness in
daily living and in meeting the world and its challenges.
The subtests Wechsler selected and developed tap many different mental
abilities, which together reflect an individual's overall ability. Some require
abstract reasoning, whereas others require perceptual skills, verbal skills,

2
 Concept of Memory and Learning

and processing speed. All of these abilities are valued to varying degrees by
our society, and all relate to behavior that is generally considered intelligent
in one way or another. None of the subtests by itself, however, was designed
to assess the entire range of cognitive abilities. For example, a subtest may
require the examinee to use primarily perceptual skills but not abstract
reasoning; another subtest may require the individual to recall specific
information but not to perceive spatial relationships.
Wechsler viewed his intelligence scales as clinical instruments that sample
an individual's abilities. He also believed that the abilities represented by
these tests are not always developed equally in most “normal” individuals.
Experience has shown that peaks and valleys are typical of the scores
obtained by most individuals, a pattern indicating that intellectual abilities
are developed in different ways and result in different profiles of cognitive
strengths and weaknesses.
Although tests of intellectual ability, such as the WAIS-III, provide a consid-
erable amount of information about an individual’s relative intellectual
strengths and weaknesses in a relatively short amount of time, the clinician
should view each examinee as unique and take into account nonintellective
factors and other life-history information when interpreting the test results.
Emphasizing the importance of this approach, Matarazzo (1972, 1990) re-
minded the clinician of the necessity of considering an individual's life his-
tory (e.g., social and medical history and linguistic and cultural background)
as part of a comprehensive assessment. Test scores, behavioral observations,
and life histories are critical sources of information in all diagnostic assess-
ments, but clinicians should keep in mind that they themselves are the
cornerstone of any assessment. Those who are responsible for interpreting
the results of intelligence testing must be careful to distinguish between
cognitive abilities, conative factors (i.e., personality traits, such as anxiety,
persistence, and goal awareness), and other nonintellective variables that
contribute to test performance (Wechsler, 1950).

Concept of Memory and Learning

The term memory has been conceptualized and used in many different
ways. The concept of memory is closely linked to learning because memory
is the natural outcome of learning. Squire (1987) provided an excellent defi-
nition of learning and memory: “Learning is the process of acquiring new
information, while memory refers to the persistence of learning in a state
that can be revealed at a later time” (p. 3).
A widely recognized view is that memory consists of a short-term system
and a long-term system (R. C. Atkinson & Shiffrin, 1968). In its most basic
Introduction

form, short-term memory refers to a temporary storage (usually from only

seconds to 1-2 minutes), whereas long-term memory refers to the perma-
nent or more stable storage of memories. The process by which information
is transformed into mental representations is referred to as encoding. The
process of bringing stored information into conscious awareness, Or remem-
bering, is referred to as retrieval.
In current psychological theories of learning and memory, long-term
memory is often categorized as either procedural or declarative memory.
Procedural memory effects a change in a person's behavior on the basis of
experiences without the person’s necessarily having conscious access to the
events that produced the change in behavior (Squire & Butters, 1992). These
behaviors, such as driving a car when one already knows how to drive or
knowing how to get back and forth to work or school, are considered to be
performed automatically. In contrast, declarative memory is the ability to
store and retrieve specific pieces or bits of information or knowledge (Squire
& Butters, 1992).
Declarative memory can be further divided into semantic and episodic
memory. Semantic memory involves memories for general facts and con-
cepts. Episodic memory involves information that is situation- and context-
specific. In view of this conceptual framework, the WMS-III is primarily a
measure of declarative episodic memory. That is, the information that is
presented is novel and contextually bound by the testing situation and
requires the examinee to learn and retrieve information.
Considerable research into the neurological basis for memory functioning
and impairment has accumulated. A thorough critique of the research inves-
tigating the neural circuitry of memory is beyond the scope of this Manual
(for such reviews see Squire, 1992, and Squire & Butters, 1992); however, a
brief synopsis is provided. The neural circuitry involving the cortical and
subcortical limbic structures of the medial temporal lobes, especially the
hippocampus, amygdala, and related diencephalic structures, have been
implicated as important structures for memory functioning (Squire, 1992;
Squire & Butters, 1992). Historically, hemispheric differences in memory
processing have been suggested, specifically that verbal memory may be
predominantly processed in the left temporal lobe structures, whereas visual
and perceptual memory may be processed predominantly in the corre-
sponding structures of the right hemisphere (Milner, 1968). However, evi-
dence for hemispheric specificity for auditory and visual stimuli is far from
consistent in the literature. Other brain regions may also affect memory
functioning. Lesions of diencephalon structures (e.g., the medial—dorsal
nuclei of the thalamus) may result in reduced memory functioning at a
different stage of the encoding-retrieval process (Squire & Butters, 1992).
 Concept of Working Memory

Lesions of the frontal lobe may also result in memory dysfunction. This
memory dysfunction differs from anterograde amnesia, both quantitatively
and qualitatively, and is associated with decreased learning efficiency due
to a failure to employ effective encoding and retrieval strategies, greater sus-
ceptibility to interference effects, problems monitoring recall for redundant
or incorrect information, and breakdown of recall for event order, time, and
source of information (Malloy & Richardson, 1994; Stuss, Alexander, et alee
1994; Stuss & Benson, 1984, 1986; Stuss, Eskes, & Foster, 1994). Differences in
patterns of memory performance have been found useful in discriminating
among Clinical groups with cerebral dysfunction or functional disorders
resulting from various neuropathological or psychological processes (Butters
et al., 1988; Delis et al., 1991; Massman, Delis, Butters, Dupont, & Gillin, 1992).

Concept of Working Memory

Working memory denotes a person's information-processing capacity.
Daneman and Carpenter (1980) credited Newell and his colleagues with the
term “working memory” (Newell, 1973; Newell & Simon, 1972) when they
developed their model of an information-processing system. The concept
of working memory has replaced (or updated) the concept of short-term
memory. This conceptual workspace is currently viewed as an active part of
the information-processing system as opposed to the traditional short-term
memory, which was viewed as the passive storage buffer. Therefore, the
concepts of working memory and short-term memory are similar because
both refer to the temporary storage of incoming information and because
both are limited in capacity. However, the two concepts differ in one key
aspect: Short-term memory is viewed as a passive form of memory, whereas
working memory is viewed as an active form. Traditional short-term mem-
ory refers to the passive storage of information while that information either
becomes encoded into long-term memory or is forgotten. Working memory,
on the other hand, serves as more than a temporary storage space for
incoming information. Rather, it is where calculations and manipulations of
information occur. Furthermore, as Baddeley and Hitch (1974) pointed out,
this component stores the products or output of these calculations and
transformations in addition to the original information.
The measurement of working memory dates back to the early experiments
conducted by Baddeley and Hitch (1974). Traditionally, this construct has
been measured by presenting a large amount of information to the exami-
nee. The examinee must first process or transform this information and then
retain the end product. The task of recalling the information may occur
immediately after the presentation of material, as in the Digit Span subtest.
Introduction

However, when tasks increase in complexity (e.g., more information is pre-

sented), the working memory system becomes increasingly taxed. Baddeley
(1986) stated,
if learning and/or retrieval were limited by the amount of available
attentional capacity, then requiring a subject to perform a second
attention-demanding task during learning or retrieval should cause
performance impairment. Furthermore . . . the greater the extent to
which a process was limited by available attention, the more susceptible
it should be to disruption by an attention-demanding task. (p. 39)
Tasks that have been developed to test the maximum attention span either
increase the amount of information that must be stored during a single
task or require the examinee to perform two tasks simultaneously. De Jonge
and de Jong (1996), building on the distinctions made by Turner and Engle
(1989), categorized these two types of tasks as simple span and complex span
tasks. The simple span tasks measure the storage component of working
memory because they deemphasize the manipulation of the material. The
Digits Forward part of the WAIS-III and WMS-III Digit Span subtest is an
example of a simple span task. In complex span tasks, both the storage and
processing of information are involved simultaneously. For example, the
examinee must perform two different types of mental processes at the same
time, such as reading sentences aloud while remembering the last word of
the previous sentence (see Daneman & Carpenter, 1980). A working memory
span task also becomes complex by increasing the amount of material that
must be manipulated. For example, the task requires the examinee to per-
form more extensive calculations on the material that has been stored in
memory, such as mentally solving arithmetic problems. Both the WAIS-III
Arithmetic subtest, which requires the examinee to perform somewhat
complex arithmetic calculations mentally, and Digits Backward of the Digit
Span subtest, which requires the examinee to reorder number sequences
mentally, are complex tasks because they require calculation or reordering of
the information. Despite the distinctions between simple and complex tasks
proposed by de Jonge and de Jong, their research has shown that these types
of working memory tasks are related and that both form a single dimension.
Working memory tasks have been distinguished according to visual and
verbal material (Baddeley, 1986; J. T. E. Richardson, 1996). Baddeley and
Hitch (1974) originally proposed a multiple-component system of working
memory. Building on this theory, some researchers believe that working
memory has three distinct components, which serve as “workspace buffers”
for information that is to be processed (Logie, 1996). The system comprises a
central executive processor and two “slave” systems. The two slave systems
are the phonological loop, where verbal material is stored and processed,
 Concept of Working Memory

and the visuospatial sketch pad, where spatial material is stored and
processed (Baddeley, 1986, 1992; Logie, 1995, 1996).
Although this three-component theory is popular, other researchers have
deemphasized the distinction between the verbal (phonological loop) and
the visual (visuospatial sketch pad) components (see J. T. E. Richardson,
1996). The WAIS-III and the WMS-III also deemphasize the distinction
between verbal and visual material. The WAIS—III measures working
memory with tasks in which the material is presented auditorily, whereas
the WMS-III Working Memory Index is equally weighted with one visual
task and one auditory task.
Though the specifics of a working memory model are disputed, most cogni-
tive psychologists agree that the core of any definition of working memory
involves the temporary storage of material that is in an active state. Carlson,
Khoo, Yaure, and Schneider (1990) have pointed out that there is a single
workspace, which is limited, and that this single-workspace model holds
whether or not the working memory is divided into subsystems (p. 195).
Recent literature has suggested that working memory is a key component of
learning (Kyllonen, 1987; Kyllonen & Christal, 1987, 1990; Woltz, 1988). An
individual-differences model of working memory predicts that the greater
the working memory is, the greater the attention and learning capabilities
will be. According to this theory, working memory is responsible for learner
differences in a wide variety of learning tasks. Building on E. H. Cooper
and Pantle’s (1967) “total time hypothesis,” which states that the amount
of information learned is a direct function of the amount of time spent
learning, Baddeley (1986) proposed that the crucial factor is not necessarily
time, but rather the amount of processing that can occur. Cognitive psychol-
ogists have posited that working memory is one of the important predictors
of the individual differences in learning, intellectual ability, and fluid reason-
ing (Kyllonen, 1987; Kyllonen & Christal, 1990; Sternberg, 1980). According
to Kyllonen (1987), working memory capacity, along with information-
processing speed, a declarative/factual knowledge base, and a procedural
knowledge base, underlies the individual's ability to learn new information.
Research has provided initial support for this premise of the relationship
between working memory and reasoning tasks (de Jong & Das-Smaal, 1995;
Fry & Hale, 1996; Jurden, 1995; Kyllonen & Christal, 1990). Although working
memory and reasoning appear related, there is also ample evidence suggest-
ing that the two constructs are not identical but are quite distinct (Carlson
et al., 1990; de Jonge & de Jong, 1996; Kyllonen & Christal, 1990). In sum,
the research indicates that working memory capacity is an important mod-
erating variable of learning.
Introduction

Antecedents of the
WAIS-Ill and the WMS-Iil
Wechsler’s original intelligence test, the Wechsler—Bellevue Intelligence Scale
(1939), was a milestone in the history of intelligence testing because it
incorporated both verbal and performance scales and yielded scores for
those scales in addition to an overall composite score. Further, the Wechsler—
Bellevue was innovative because it provided deviation IQ scores that were
based on standard scores computed with the same distributional character-
istics at all ages. Wechsler (1944) constructed the test by collecting a sample
that matched the population of the United States on several key variables
(e.g., age, sex, education level, occupation level) and then by normalizing the
scores. The Wechsler—Bellevue and its descendants, including the WAIS—III,
have each included a group of different subtests that contribute to global IQ
scores. These features and the structure of the test have remained intact
through the years since the Wechsler—Bellevue. With each revision, the
norms were updated, outdated items replaced, and scoring rules changed.
The WMS-III is the most recent revision of the original Wechsler Memory
Scale (Wechsler, 1945) and the Wechsler Memory Scale—Revised (WMS-R;
Wechsler, 1987). Like its predecessors, the WMS-III is an individually
administered, clinical instrument designed to assess important domains of
memory and learning in older adolescent and adult populations. Although
the WMS-III has maintained many aspects of its predecessors, significant
improvements have been made to the test in response to both current
research and theory and the needs of clinicians. With each successive ver-
sion, clinicians have been provided flexibility in the content and scope of
memory assessment, from a general screening to a more intensive, detailed
analysis of memory functioning.

Development of the Scales

The developmental phases of the WAIS-III and the WMS-III are very similar,
and both included the following five broad stages:
* areview of the existing items and development of new items
and subtests;
* pilot testing of the revised items and subtests to investigate psycho-
metric characteristics;
 Antecedents of the WAIS-III and the WMS-III

* anational tryout study to examine item difficulties, item bias, subtest

functioning, and factor structure (concurrent with the tryout, the scales
were administered to various clinical groups in order to investigate
clinical utility);
¢ a large national standardization study to collect normative information,
to investigate bias, and to make final item and subtest decisions; and

¢ multiple studies conducted concurrently with standardization to

determine the reliability, concurrent validity, construct validity, and
clinical utility of the test.
The following sections provide overviews of the developmental phases of
the WAIS-III and the WMS-III and provide the rationale for the revisions of
the Wechsler Adult Intelligence Scale—Revised (WAIS-R; Wechsler, 1981) and
the WMS-R (Wechsler, 1987).

WAIS-IlI
The decision to revise the WAIS-R involved the following issues:
¢ Updating of Norms. Because there is a real phenomenon of IQ-score
inflation over time, norms for a test of intellectual functioning should be
updated regularly (Flynn, 1984, 1987; Matarazzo, 1972). Data suggest
that an examinee’s IQ score will generally be higher when outdated
rather than current norms are used. The inflation rate of IQ scores is
about 0.3 points each year. Therefore, if the mean IQ score of the U.S.
population on the WAIS-R was 100 in 1981, the inflation might cause it
to be about 105 in 1997. Some of Matarazzo’s and Flynn’s suggested
causes of this IQ-score inflation in the general population are improve-
ment in the education system, improved nutrition, better health condi-
tions, and increased dissemination of information. Regardless of the
reasons for these changes in test performance, periodic updating of the
norms is essential; otherwise, average IQ scores will gradually drift
upward and give a progressively deceptive picture of an individual’s
performance relative to the expected scores in his or her own age group.
The normative data have been updated by the restandardization of the
instrument. The WAIS-III sampling plan included 2,450 individuals aged
16-89 years. The sample was divided into 13 age groups and stratified on
key demographic variables, including age, sex, education level, and geo-
graphic region according to the U.S. census data (U.S. Bureau of the
Census, 1995). A complete description of the obtained sampling matrix
is provided in Chapter 2 of this Manual.
Introduction

Extension of the Age Range. Individuals in the United States are living
longer. Current estimates place the average life expectancy at birth to be
over 78 years for women and over 72 years for men (Rosenberg, Ventura,
Maurer, Heuser, & Freedman, 1996). The WAIS-R provides normative
information for individuals only up to 74 years of age; therefore, it does
not address the significant population of adults over 74 years of age.
In the United States, the current population of adults aged 75 years
and older is approximately 15 million, or 6%, of the total population
across all age groups (U.S. Bureau of the Census, 1997). Furthermore,
the proportion of older adults is expected to increase.
Because of the growing population of older adults, the age range of
the WAIS-III extends through 89 years of age. Additionally, the stimuli
were modified to reflect this change: Artwork was redrawn, stimuli
were enlarged, and the use of bonus points for quick performance
was deemphasized. A new subtest, Matrix Reasoning, which does not
require manual manipulation or quick performance, was added to the
Performance scale. All of these features make the scale more appropriate
for an older adult population.
Modification of Items. In most assessment instruments, some items
become outdated over time. The WAIS-R items are no exception. In
the Information subtest, for instance, the content of some items is too
chronologically remote for younger examinees. Moreover, contemporary
methodologies for testing item bias were used for the WAIS-III item
selection.
Problematic items were identified and deleted on the basis of formal
reviews of the items and empirical data from statistical and bias
analyses. The formal reviews were conducted by experts in crosscultural
research, intelligence testing, or both. Reviews were collected at three
key points during the development of the test, with approximately 20-25
reviews obtained at each stage. During the very initial stages of the
project, all WAIS-R subtests and items were reviewed by internal and
external reviewers for potential bias, datedness, content relevance, and
clinical utility. Bias experts evaluated items in terms of content and
potential bias. During the tryout phase and again during the standardi-
zation phase of the project, content and bias experts reviewed the items
and identified those that were potentially problematic.
Along with these reviews, empirical data were used to test hypotheses
and to assist in the decision process. First, on the basis of item statistics
and item bias analyses of the WAIS-R standardization sample data,
biased or dated items were deleted or rewritten. The retained items were
then tested in three pilot studies (with sample sizes ranging from 113 to

10
 Development of the Scales

168 examinees). Once again, the empirical data (item difficulty and item
correlations with the relevant subtest total score) were used to select
those items to be tested in the nationwide tryout study. For the tryout,
446 participants composed the sample, which was stratified according to
age, sex, education level, race/ethnicity, and geographic region. Item
analyses based on the tryout data were performed for each subtest to
determine the item sets for standardization. Data from an oversampling
of 162 African American and Hispanic examinees helped the project
team detect and remove items that were potentially biased against either
of these groups. Results from traditional Mantel-Haenszel bias analysis
(Holland & Thayer, 1988) and item response theory (IRT) bias analyses
(Hambleton, 1993) provided further data on potentially problematic
items. During the standardization phase, the procedures were repeated.
Item analyses based on the standardization data were performed for
each subtest to determine the item sets for the final version.
Item bias analyses based on an IRT method require data from a large
number of examinees. Certain item parameters must be estimated
through an iterative process. This process requires responses to each
item on each subtest by a minimum of 200 individuals in both the focal
and comparison groups. Although the sampling data (see Tables 2.2-
2.13) indicate that an ample number of examinees were tested, not every
examinee necessarily completed every item because of discontinue
rules. Furthermore, estimations based on the data collected during the
tryout phase suggested that a sufficient number of examinees might not
complete the most difficult items. Therefore, an oversampling of 200
African American and Hispanic participants were tested without dis-
continue rules so that the item bias analyses could be repeated with
sufficient observed item scores for both of these groups. On the basis
of these empirical analyses and the content reviews, items that did not
meet acceptable criteria were removed.
Updating Artwork. Because much of the WAIS-R artwork has become
outdated and is not likely to be attractive to examinees, the WAIS-III art-
work has been made more contemporary. Moreover, some of the visual
stimuli have been enlarged so that individuals with visual acuity prob-
lems will not be at a disadvantage.
Several steps were taken to make the WAIS-III stimuli more attractive.
The Picture Completion items were redrawn, enlarged, and colorized.
Despite concerns that colorizing the artwork might change the nature of
the task, colored pictures were deemed more relevant and ecologically
valid (i.e., easily transferred to real-life situations). The Picture Arrange-
ment cards also were redrawn and modernized. Several WAIS-R items
had been derived from comic strips popular at the time the previous

11
Introduction

editions were developed, and these items were removed. Improvements

were also made to the Object Assembly subtest. The WAIS-III Object
Assembly Layout Shield is constructed of sturdy card stock so that it can
stand alone and includes the item instructions. The puzzle pieces fea-
ture numbers printed on the back to assist the examiner in laying out
the pieces in the specified arrangement. Finally, the Digit Symbol—
Coding subtest features more space between the key and the items to
prevent left-handed examinees from blocking the key from view as
they work.
Extension of Floor and Enhancement of Clinical Utility. The IQ scores
of the WAIS-R do not extend downward far enough to discriminate ade-
quately among examinees with mild to moderate mental retardation.
For the oldest age group (70-74 years), the WAIS-R Verbal IQ (VIQ)
scores extend only 2.67 SDs below the mean. On the Performance scale,
a 70-year-old individual who cannot respond correctly to even one
Performance item on any of the Performance subtests can still obtain a
WAIS-R Performance IQ (PIQ) score of 61 points. The restricted floor is
not limited to the IQ scores; the WAIS-R subtest scores are scaled to
about 3 SDs below the mean, but often there are not enough easy items
to permit accurate scaling to this level.
On the WAIS-III, the range of possible scores has been extended down-
ward. Easier items, which are administered if a basal criterion is not met,
were added to several subtests. In the WAIS-III, the Full Scale IQ (FSIQ)
scores extend downward to 45, the VIQ scores to 48, and the PIQ scores to
47. Data were collected on individuals diagnosed with mild or moderate
mental retardation according to the Diagnostic and Statistical Manual
of Mental Disorders—Fourth Edition (DSM-IV; American Psychiatric
Association, 1994) and the American Association of Mental Retardation
criteria (1992). Partially on the basis of these data, IQ scores in this range
were extrapolated downward to make assessment more feasible at this
lower end of functioning.
Additionally, new diagnostic features were included to help the examiner
test the limits of performance or to examine more closely the types of
errors that examinees make. These features make the WAIS-III more use-
ful in the field of neuropsychology. For instance, an optional procedure
for testing incidental learning following Digit Symbol—Coding adminis-
tration (Hart, Kwentus, Wade, & Hamer, 1987; E. Kaplan, Fein, Morris, &
Delis, 1991) was added to the WAIS-III.
Decreased Reliance on Timed Performance. Six of the WAIS-R subtests
have time limits, and many of their items include bonus points for
quick performance. Such time constraints are especially problematic for
older adults, whose processing speed is expected to decrease. Although

12
 Development of the Scales

processing speed is important and should be tested, measures of other

intellective processes should not be confounded by this factor. There-
fore, for the WAIS-III, a new nonverbal subtest (Matrix Reasoning) that
does not have time limits was created. Additionally, the number of items
with time-bonus points was decreased in the existing subtests.
To decrease the reliance of the PIQ score on quick performance and
subsequent bonus points, Matrix Reasoning, a nonverbal, unspeeded
subtest of abstract reasoning, was added to the Performance scale. It
replaces Object Assembly as a standard subtest. Object Assembly, which
relies heavily on bonus points for quick performance, is now an optional
subtest and is not required for computing WAIS-III IQ or Index scores.
Enhancement of Fluid Reasoning Measurement. Several theories of
cognitive functioning emphasize the assessment of fluid reasoning
(e.g., Carroll, 1997; Cattell, 1943, 1963; Cattell & Horn, 1978; Sternberg,
1995). Fluid reasoning is the “ability to perform mental operations, such
as manipulation of abstract symbols” (Sternberg, 1995, p. 437). Matrix-
reasoning types of tasks, for example, are considered typical measures of
this type of ability. The WAIS-R has been criticized for not having
subtests that sufficiently measure abstract, fluid reasoning. The new
WAIS-III subtest, Matrix Reasoning, has been added to enhance mea-
surement of this domain.
Strengthening the Theoretical Basis. Current research suggests that
cognitive functioning encompasses more than what is measured by VIQ
and PIQ scores. For instance, in their review of the literature of the
various factor analytic studies of the WAIS-R, Leckliter, Matarazzo, and
Silverstein (1986) showed that most researchers have found a model
with three factors (verbal comprehension, perceptual organization, and
memory/freedom from distractibility). In addition to the traditional IQ
scores, the WAIS-III includes Index scores, which are measures of more
discrete factors and domains.
From the beginning of the WAIS-III project, attempts were made to
include new subtests that would be related to a hypothesized third fac-
tor (Working Memory) and to a hypothesized fourth factor (Processing
Speed). These factors have been labeled “mediators” in cognitive func-
tioning because the component skills have been found to be important
to learning (Kyllonen & Stephens, 1990; Woltz, 1988). Kyllonen and
Christal (1990) have demonstrated the relationship between working
memory and g, or global intellectual ability. Kyllonen (1987) has also
advanced a formal theory of cognitive functioning in which working
memory and processing speed are core components in the acquisition
of new information. Therefore, two new subtests were developed for

13
Introduction

the WAIS-III. Letter-Number Sequencing was designed as a measure

of working memory, and Symbol Search was designed as a measure of
processing speed.
Factor analyses of the WAIS-III standardization data support a model of
four indexes: verbal comprehension, perceptual organization, working
memory, and processing speed. This organization of the WAIS-III sub-
tests into more discrete cognitive domains, or indexes, is especially
important because working memory and processing speed, which are
related to learning acquisition, can be distinguished from the verbal
comprehension tasks and perceptual organization tasks. This breakdown
can be especially useful for diagnosing learning disabilities, attention-
deficit/hyperactivity disorder (ADHD), and other cognitive deficiencies.

Statistical Linkage to Other Measures of Cognitive Functioning and

Achievement. The Wechsler Intelligence Scale for Children—Third
Edition (WISC-III; Wechsler, 1991) was co-normed with the Wechsler
Individual Achievement Test (WIAT; The Psychological Corporation,
1992). This linkage provides examiners more information about the
interrelationship of a broader spectrum of cognitive abilities. This
co-norming has also enabled examiners to “predict” an examinee’s
achievement scores on the basis of his or her intellectual ability score.
To allow for such predictions with the WAIS-III, the WAIS-III and the
WIAT were linked for 16- to 19-year-olds.
Additionally, because memory and intellectual ability in adults are
commonly tested concurrently, the WAIS-III and the WMS-III were co-
normed. A sample of 1,250 individuals took both the WAIS-III and the
WMS-III in a counterbalanced order. These data allow direct compari-
son of intelligence and memory through normative information.
Extensive Testing of Reliability and Validity. Finally, the developmental
phase of the WAIS-III included additional studies of psychometric
properties of the scale. In a study of the stability of the instrument,
394 examinees were retested from 2 to 12 weeks after the first testing.
Evidence of the concurrent validity of the WAIS-III was provided by
correlation studies between the WAIS-III and the following instruments:
the WAIS-R, the WISC-III, the WIAT, the Stanford-Binet Intelligence
Scale—Fourth Edition (R. L. Thorndike, Hagen, & Sattler, 1986), and
the Standard Progressive Matrices (Raven, 1976). Finally, for evidence of
the construct validity and the clinical utility of the scale, the WAIS-III
was administered to individuals with neuropsychological deficits
(e.g., Alzheimer’s dementia, traumatic brain injury), mental retardation,
psychiatric disorders, learning disabilities, and hearing impairments.

14
 Development of the Scales

WMS-Iil
In 1987, the WMS was revised for the first time (see Prigatano, 1977, 1978,
for critical reviews of the WMS). The WMS-R provided improved norms,
extensive scoring rules, additional subtests for measuring delayed recall of
information, and other new subtests with visually presented stimuli. These
additional subtests were developed by Wechsler, although, the final revision
was not published until after his death. The WMS-R has been the subject of
numerous research studies since its publication. Although the WMS-R pro-
vided clear advantages over its predecessor, several comprehensive reviews
have identified areas in which the scale could be improved (Chelune,
Bornstein, & Prifitera, 1990; Elwood, 1991; Loring, 1989).
During the initial phases of the WMS-III development, all WMS-R subtests
and items were reviewed for potential cultural and sex bias, appropriateness
of content, theoretical basis, and clinical utility. Expert bias reviewers evalu-
ated the WMS-R items, and approximately 20 clinical psychologists and
neuropsychologists evaluated the scale in terms of content, psychometrics,
and clinical utility. Additionally, a standing advisory panel met about twice
yearly to review the developmental research, to consult on technical and
content issues, and to provide direction regarding the standardization sam-
pling and the validation studies with clinical groups.
During the early stages of development, numerous small pilot studies were
conducted, and the results of these studies provided data on the various
psychometric characteristics of new subtests and for the revised subtests.
Additional pilot studies involving various clinical groups provided informa-
tion on clinical utility and on both examiner and examinee friendliness.
These studies resulted in additional refinements to content, administration
procedures, and scoring.
The next major phase of development consisted of a national tryout with
approximately 450 individuals who were administered both the WMS-III
and the WAIS-III. All items and subtests were evaluated for cultural and sex
bias, psychometric characteristics, administration procedures, clinical util-
ity, and underlying factorial structure. Scoring studies were performed for
the Logical Memory, Visual Reproduction, and Family Pictures subtests to
investigate their reliability and clinical sensitivity. On the basis of these
results, the subtests for the standardization of the WMS-III were selected.
In addition to the battery of memory subtests, the standardization version
included a number of other measures for the purpose of evaluating con-
current validity.

15
Introduction

In summary, the solicited reviews of the WMS-R, an extensive review of the

published literature, and the recommendations from the advisory panel
identified a number of ways in which the WMS-R could be improved. These
improvements fell into two broad categories: normative/psychometric
issues and content/score configuration issues. A summary of the identified
issues and how these issues were addressed and incorporated in the
WMS-III follows.

Normative/Psychometric Issues
Standardization Sample Size, Age Range, and Representativeness. The
WMS-R standardization sample consisted of six age groups, each of
which included approximately 50 individuals. The total sample size was
approximately 300 individuals who ranged in age from 16 to 74 years.
Although the standardization sample of the WMS-R was relatively small
compared to those of other contemporary Wechsler scales, it was com-
parable to the samples for the original WAIS and other contemporary
neuropsychological instruments. The concerns underlying a small
sample size are an increased likelihood of greater measurement error
and a less accurate estimate of population parameters. Also, the nor-
mative age range of the WMS-R, like that of the WAIS-R, is limited to
individuals younger than 75 years of age. Finally, because memory is a
subset of overall intellectual functioning, it is important for the individ-
uals in the standardization sample to be representative of the general
population in terms of overall ability level.
The WMS-IUII standardization sample included 1,250 individuals ranging
in age from 16 to 89 years. The sample was divided into 13 age groups.
The first 11 age groups, spanning ages 16-79, included 100 participants
each. The last two age groups, 80-84 and 85-89, each included 75 parti-
cipants. Within each standardization stratification variable, the WMS-III
was randomly administered to one half of the WAIS-III standardization
sample. In this way, the ability levels of the WMS-III standardization
participants would be representative of the general population.
Interpolated Norms. Normative scores for three of the WMS-R age
groups (18-19 years, 25-34 years, and 45-54 years) were interpolated on
the basis of the scores of the adjacent sampled age groups. To the degree
that age-based memory performance is nonlinear, interpolation of
norms may not fully capture the relationship. Although no definitive
evidence shows a nonlinear relationship for the nonsampled age groups,
it was recommended that each age group be sampled.
For the WMS—III, each age group in the standardization sample was
sampled. None of the normative scores for the WMS-III age groups were

16
 Development of the Scales

interpolated. Normative evidence from the WMS-III has demonstrated

that for comparable constructs between the two instruments, there is
indeed a linear relationship for the nonsampled age groups in the
WMS-R for which norms were interpolated.
¢ Scale Reliability. The reliability coefficients of some of the WMS-R
indexes and subtests are lower than desirable. In general, reduced
reliability decreases the scale’s clinical sensitivity and results in larger
confidence intervals. The restriction of scale range is one reason for
these low reliabilities in the WMS-R. For example, the Figural Memory
subtest has only four items, and the Mental Control subtest has only
three items.
Reliability coefficients for the WMS-III Primary subtests and Primary
Indexes are generally higher than those of the WMS-R. Average subtest
internal consistency reliability coefficients range from the .70s to the
.90s. With one exception, average composite reliability coefficients of the
Primary Indexes range in the .80s and .90s. The Auditory Recognition
Delayed Index has a reliability coefficient of .74.

Content/Scoring Configuration Issues

¢ Visual Memory Stimuli. There is little empirical evidence that the WMS-R
visual memory subtests are adequate measures of a hypothetical “pure”
visual memory system or that they are differentially sensitive for individu-
als with unilateral hemispheric lesions (Chelune & Bornstein, 1988;
Heilbronner, 1992; Loring, 1989; Naugle, Chelune, Cheek, Liiders, &
Awad, 1993). Research has also indicated that performance on the Visual
Reproduction subtest can be confounded by the effects of constructional
dyspraxia (Haut, Weber, Demarest, Keefover, & Rankin, 1996; Haut, Weber,
Wilhelm, Keefover, & Rankin, 1994), and that a direct-copy condition
would help in interpreting performance on this subtest (E. Kaplan, 1988).
On the basis of the reviews, two of the three WMS-R visual memory
subtests were deleted from the scale: Figural Memory and Visual Paired
Associates. Five experimental subtests of visually presented memory
stimuli were developed as possible replacements for these two. For
these possible replacement subtests, an effort was made to include
visual stimuli that would make verbal encoding difficult. Rather than
purporting to tap exclusively a hypothetical verbal or visual memory
system, the WMS-III distinguishes between auditory and visual memory
by the modality of presentation of the subtests. Preliminary studies of the
WMS-III (presented in Chapter 4 of this Manual) support the discrimi-
nant validity of the auditory and visual memory measures.

17
Introduction

Internal Validity. Factor analytic studies of the WMS-R have not yielded
a factor solution consistent with the index structure of the WMS-R.
Support for a two-factor solution—General Memory and Attention/
Concentration—when only immediate memory subtests are entered
into the analysis has been reported in several studies (e.g., Bornstein &
Chelune, 1988; Roid, Prifitera, & Ledbetter, 1988). The index configura-
tion of the WMS-III was not constructed on the basis of factor analysis.
Rather, the indexes were constructed on the basis of clinically meaning-
ful aspects of clinical memory assessment. Chapter 4 of this Manual
presents the results of confirmatory factor analyses for various concep-
tualizations of dimensions of memory. These results provide support for
the index structure of the WMS-III.
Recognition Versus Recall Memory. Because the WMS-R does not
provide standardized recognition trials, it is limited as a means of
identifying specific retrieval problems. It has been suggested that such
specificity may help to distinguish between clinical groups (Butters et
al., 1988).
Whenever possible, recognition measures were added to the WMS-III
and are administered immediately after measures of delayed recall.
The Auditory Recognition Delayed Index was added to the memory
indexes. This index can be contrasted to the Auditory Delayed Index,
which is composed of parallel recall measures.

18
 CHAPTER 2

Standardization and
Norms Development

Standardization
Description of the Samples
The WAIS—III and WMS-III normative information presented in this Manual
is based on national standardization samples representative of the U.S.
population of adults aged 16-89 years. A stratified sampling plan ensured
that the standardization samples included representative proportions of
adults according to each selected demographic variable. An analysis of
data gathered in 1995 by the U.S. Bureau of the Census provided the basis
for stratification along the following variables: age, sex, race/ethnicity,
education level, and geographic region. The following sections present the
characteristics of the WAIS-III and WMS-III standardization samples.
e Age. The standardization sample for the WAIS-III included 2,450 adults.
For the WMS-III standardization sample, 1,032 adults were tested. This
standardization sample was weighted to match the 1995 census data and
each age group was required to have an average full scale IQ score of
100. This weighting method yielded a sample of 1,250, which was used
for both the norming of the WMS-III and the working memory factor on
the WAIS-III, as well as for any related analyses reported in this manual.
Each sample was divided into 13 age bands: 16-17, 18-19, 20-24, 25-29,
30-34, 35-44, 45-54, 55-64, 65-69, 70-74, 75-79, 80-84, 85-89. These
samples improve on the WAIS-R and WMS-R samples, which included
1,880 and approximately 300 adults aged 16-74, respectively. Except for
the two oldest age groups, each WAIS-III group included 200 partici-
pants; the 80-84 age group included 150 participants, and the 85-89
age group included 100 participants. For the WMS-III, 100 participants
were included in each age group except the two oldest groups, which
consisted of 75 participants each.
In addition to the basic standardization samples, 437 individuals were
tested so that at least 30 participants for the WAIS-III and 20 participants

19
Standardization and Norms Development

for the WMS-III would be included in each educational level within

each age group. The data from these additional cases were treated as
oversampling data and were excluded from the basic standardization
samples. These additional data were collected for later research investi-
gating the relationships between cognitive abilities and education level.
¢ Sex. The standardization sample consisted of an equal number of male
and female participants in each age group from 16 through 64. The older
age groups included more women than men, in proportions consistent
with census data.
¢ Race/Ethnicity. For each age group in the standardization samples,
the proportions of Whites, African Americans, Hispanics, and other
racial/ethnic groups were based on the racial/ethnic proportions of
individuals within each age band within the U.S. population according
to the 1995 Census data.
An additional 200 African American and Hispanic individuals were
administered the WAIS-III without discontinue rules. This oversampling
provided a sufficient number of item scores across all items for item bias
analyses.
¢ Educational Level. The samples were stratified according to the follow-
ing five education levels based on the number of years of school com-
pleted: < 8 years, 9-11 years, 12 years, 13-15 years, and > 16 years. For
examinees aged 16-19, parent education was used.
* Geographic Region. The United States was divided into the four major
regions specified by the Census reports (see Figure 2.1): Northeast (NE),
North Central (NC), South (S), and West (W). The number of participants
from each region was proportionate to the population in each region.

North Centrai Northeast

Figure 2.1. Standardization Sampling Sites

20
 Standardization

Locating and Testing the Samples

The collection of standardization data was achieved primarily through the
use of marketing research firms in 28 U.S. cities in the Northeast, North
Central, South, and West regions. These firms used various methods to
recruit participants to fit the sampling plan matrix; these included random
telephone calls, newspaper advertisements, and flyers placed in senior cen-
ters and with various community organizations. Additionally, independent
examiners from most states recruited and tested examinees. All participants
were paid an incentive fee to participate.
All potential standardization participants were medically and psychiatrically
screened with a self-report questionnaire. Table 2.1 lists the criteria by which
individuals were disqualified from the standardization sample.

Table 2.1. Exclusionary Criteria for the Standardization Samples

¢ Color-blindness
¢ Uncorrected hearing loss
¢ Uncorrected visual impairment
¢ Current treatment for alcohol or drug dependence
* Consumption of more than three alcoholic beverages on more than two nights
a week
* Seeing a doctor or other professional for memory problems or problems
with thinking
*¢ Upper extremity disability that would affect motor performance (e.g. ability to put
puzzles together)
Any period of unconsciousness for 5 minutes or more
* Head injury resulting in hospitalization for more than 24 hours
* Medical or psychiatric condition that could potentially affect cognitive functioning,
such as
¢ Stroke
¢ Electroconvulsive treatment
¢ Epilepsy
¢ Brain surgery
¢ Encephalitis
¢ Meningitis
¢ Multiple sclerosis
¢ Parkinson's disease
¢ Huntington's chorea
¢ Alzheimer’s dementia
¢ Schizophrenia
¢ Bipolar disorder age
¢ Currently taking antidepressant, antianxiety, or antipsychotic medication
Ee ___.__________

21
Standardization and Norms Development

Representativeness of the Samples

The percentages of the U.S. population and the WAIS-III and WMS-III stan-
dardization samples according to race/ethnicity are presented in Figure 2.2.
Percentages according to geographic region are presented in Figure 2.3.
Tables 2.2-2.13 present detailed demographic information for the U.S. popu-
lation and for the WAIS-III and the WMS-III standardization samples. Tables
2.22.4 provide the proportions of the U.S. population, the WAIS—III sample,
and the WMS-III sample, respectively, according to age, race/ethnicity, and
education. Tables 2.5-2.7 provide the proportions according to age, sex, and
education. Tables 2.8—-2.10 provide the proportions by age, sex, and race/eth-
nicity, and Tables 2.11-2.13 provide the proportions by age, race/ethnicity,
and geographic region. These data indicate a close correspondence between
the two samples and U.S. Census proportions.

WMS-ili Weighted Sampling

Tulsky and Ledbetter (2000) described the weighted sampling methodology
used for the development of the WMS-III normative sample. An analysis of
the WAIS-III and WMS-III standardization data revealed that some of the
age groups did not have an average intellectual ability of 100 and that some
of the age groups did not fully represent the U.S. population according to
some of the stratification variables. The WMS-III normative sample was
therefore weighted so that each age group more closely represented the cen-
sus data and more closely approximated an average intellectual ability of
100. The actual WMS-III standardization sample consisted of 1,032 partici-
pants. From these, protocols were randomly selected within the stratification
parameters for duplication to derive the final WMS-III weighted normative
sample of 1,250.

Case weighting improves the overall representativeness of the sample.

Moreover, the weighting does not bias the results if the initial sample is rep-
resentative of the subpopulations in general. Compared to the samples for
previous versions of the scale, the WMS-III sample represents a marked
improvement. Furthermore, the WMS-III is the only commercially available
memory test to have been extensively normed on a stratified, random,
representative sample of the general population with concurrent collection
of intellectual ability.

22
 Standardization

100

White African American Hispanic Othe

Figure 2.2. Race/Ethnicity Characteristics of the U.S. Population and

WAIS-III and WMS-III Standardization Samples

40

Northeast North Central South West

Hicensus J wais-i J wus-ii

Figure 2.3. Geographic Region Characteristics of the U.S. Population and

WAIS-III and WMS-III Standardization Samples

23
 “‘uoneonpa yuared uo paseq SI UOleonpa
‘steak GT-9] pase saouTUeXe 10] {pajafduroo sieadJo JaquNu st UOHeONpY *(1oNGLsIq /IaoNnpotg) snsuaD ey} Jo nveing ‘S'f) -Od ‘Wo BUTYSeEM
‘SgGI ‘sNsuaD ay} Jo Neamng ‘s'q Aq [ary eyep a[qepeai-auTyeW]] G66T You ‘Aaaing uoryjndog JuaLing Woy are eyep uonendod ‘sf ‘ajo
SOO Seo O20. S20] Ol'0 SOO 90 IrO 692] 890 z2e0 860 920 SSh| OE lL EMI OVec ILE 00°92 | 68-S8
€l0
Sz0 sso Zz0 lro| eto 900 650 ero tec) Seo 490 OL ILZb Shy] 998 961 vELe ELbl LOCC y8-08
020
LO 6ztO 60090 «p20 6820) 910 Seo 990 260 0c] 170 20 Ov 602 ICE} LIL GrSL Z60E cOEl €0'91 | 6Z-SZ
Standardization and Norms Development

gz0 980 690 +re0 880] 920 Zeo S20 890 £2) veO 6£0 26l LOL LLC) vlcl P69! cece Zt ZL LL | vZ-OZ
zvo oso zeo iso] evo veo 6Lt 120 6LE| PLO S60 6re 6Ie 997} 89EL PII BELE Olcl g9°8 | 69-S9
660
650 90 S80 of0 Pe0| 0S0 re'O Zyl ego “Zle| e60 OO Ze Oc SLI} E02) S64 9G }e 628 6995 | 79S
lyk 90 860 SzO 620] 1Z0 vel ZBt Ol SGZ) BSL OY BOE GBL L90| reve GEIC VESC LEV cle yS-SP
Syl p80. 6780 «970 «6bZ0| 860 LE 99% SSl 182] 061 O9F SOF IO C0] 6eee cOee SIS2 chy Ycl | WYSE
oel 460 Ol ZO zlo| eck pre Ore SOc e8c| Zl SLE ZIG 9L1 Ie0| 960c C6OGL crse chy R080) ve ve
ost eel 2Z0 €20 Ozo| ozl woe 2 922 LOE] 802 OOF 09S E21 900|SE6l Bele ESec Lev cB0 | be-Se
250 612 620 8e0 200] ivO zee OOF! OOF 92] 920 ELE BOS ~SeC ~1e0) G06. /E1e 08le les “S80 | ve Ne
y60 960 80 €€0 960] 060 rel 908 962 QE] OBL Eth 68S ECE VEN} E78 OLG2 098 90h Bel | 6I-8l
Sik Ob OSO OS0 680] P80 60% OL2 99% LE] ZZ LBr EPS LEE CBO) ILL LLve crOe IVY ELL
9Ilz SI-€I ZI II-6 & | OTe SI-SI ZI II-6 9&8 | 9I= SI-SI @I II-6 8> | 91< SI-Sl cl IEG & a3sy
aio oruedsty UvILIBUTY UBL avTyM
uoneonpy pure ‘Ayormyyg /aoevy ‘easy Aq uoneindog *$*f) ay} Jo SaseyusdIag “2°Z BIGel

24
 ajqey"e'2 srydeiz0oulag Sost1 JO 9}9vI
24} eYD [TI-STpUeIS
VM UONEZIPse:ajduresg sadeyu
Aq adsag
‘asy
‘Aynor
pure
my)q/aoey uoneonpy
ou1UM UBILIY UbILIaUIY aruedsiy 1210
eae 1i-G ci ST-2i) 95 IL-6 ZI SI-eI 9Iz| 5 11-6 ZI SI-El 9I<| 85 TI-6 ZI SI-€I 9I<
LI-91 ee ee eee ne A
61-81 00@ Sey
eaG Oe2 /SSh NOL OF CEStie Serer SOPSerno
ly =: Rini
(90. OF St 70.
2-02 eee ee er oe i
62-SZ SRRS O ZINC E
S20 ogre Ot: S@> Sh |} — Of 0% 90 s0
ve-0€ ee en mer jammy gaya ge Gh” Se 02) OF Ok Sh Se OF 0} OF 60) 0¢ sO)
vy-Se GaerneRees TE TRO RRO NNO eNO eNO
NS Oe SO Sei Oe) OF | = S00 (Of ot
pS-Sb ae ne Leee hana h
RRs ae se erg See 08 Ob) OF $0 = $0) 6
ee een OF FO so
acemnc earner GameOgu eke O'heee Sty)| SO 0} S00 0] FO —
or ee e
ne HUUi M UJ H}}
es eee e
gee ee i Tena La
‘210N
NV= ‘oSGF'z UoNeoNpY
si Jaqunu
Jo sread ‘paja[duro
10J o S9eUTUILXA
pase GT-9T ‘sIead
Jou uoMeonpa
s} paseq
uo juared “Uoneonpa
s[[eD
Ul YSTYMB1ePore
payiodal a}eoIpuT
Jey] UONneZIprlepue}js
sjoI0}01d
IO] sUOSIad BUNVEUT
ay] JURAaIaI PLajI9B1aM JOU ‘P9}99[[09

25
Standardization
 age “We d1ydeiZ0UulEGg SoNst19}9eIeYD
JO 2} []]-SIM uoTezIprepueys
:ajduies sosejuediad
Aq ‘asy

N©
‘AyDouy7q/sey
pure uoneonpy

85 ip6eeo>
Ciel ti SI 5591s 8= | LI=6 Che SISEL” Sl=

eb
oyM
cl
gs

TI-6
I1-6
UPILITYy
SI-€I

SI-cie
oruedsiy

UBILIOUIY
61-81 OO! Oh in Ori Oe Oa OA Oe 0 OR ihe ree OO le Sea Oe =e =S 0'l sae Ol
ve-02 00 = we Gee ce Os| iO WS OF Co mas 0’? ar Oly Oy OL | OF OL aa 0G =:
6¢-S2 OOl OG, cae 00 OSC Ga O10 AlOr Tene Ore Le
0 Ce0 Oc: Oia Gee
0 OS CS 20 Oi Sse Sar Ol =x 0c
0Ol
vé&-0F Ors ae 08 OSCSs 0SSS OS)|) sa O08 Or O¢ OS| Wh OG ae Oe = Ot OL OF Oe Ob
vr-SE
OOL Ocge EeOU SC
0 OCsee SOLC = Orb 0a OE Ge0 Or bee
0 0c = sO Othe = OF = OG Scr: O'L
vS-Sr
00l Oi ee OFS COZ 0 SOE0 (OS ole Oe ORS = (ar Or Oe aa OL = Onl: Oca
2958s
00)s teed 0.0 Oca OO
nO a0.) OC O09Oa 06a Obes= Orbs Of = 0¢ r= 0} One aa
69-S9
00L aETL (ONSIL
§ (olins: Colello.
| Oa Oe MOS: OE MOS Ore Wi = = O'l a = aaa a ons
ue ChaoenO 0S Sie RO URO Ez =O Os OVS= ce == ee 02b Oi = 0} = ae = mo as
Standardization and Norms Development

0OL
—¥Z-02
62-SZ OOL O04 OE O'CEERS OC RP O'6C| a O02 Oe we 0 | Oe = a = “a ri; a = Ola Ol

‘a1ON Palysiam
N= ‘QS7‘T UoNeONpyY
si JIaqumu
Jo sreak ‘pa}atduroo
I0j SeauTUTeXd
pase G[—-9| ‘steak UONeONpea
si paseq
uO juared ‘UONeONpe
S[[aD
UI
YOTYMeyepare jou pajiodal 9}edIPUI
Jey} UONeZIpIepuUe}s
sjoo0}01d
IOJ SUOSIad SUTJA0UWU
BY} JURAVTAI PII9}1I9919MJOU ‘pajda][09
 27
Standardization

“uoneonpa Juared uo paseq st UoNeoNpa ‘sea GT-9] pase saduTUTexa 10j ‘paja;duioo sea Jo Jequinu
sI UOQRONpY ‘(1oNqISIG /IaoNpolg) snsuay au} Jo NeaIng "S' :OC ‘UOIUTYseM ‘GEEI ‘sNsUa_D ay} Jo Nesing
‘sq Aq [aTy eep ajqepeai-auryoeW|] S661 Youv ‘Aaaing uoyY{ndog JuaLing Woy are eyep uoVeEINdod “s"fy ‘aoNy

68'12geez9€'12 LLLt
Sesh8e°9lvseSOS

c9'6
Lc 91'9€c9'9€e9°€€ve'0€9e' 6r'bl

ee
68°21LL
ve

ee'lz

19’ $8'L 292 92'g Oz’ yS'6LebSoll

¥e'6 evi
89'S}

vee 89° vey SZ9 96'8 $08 626

cock
c0'el Srp
SEB Lesh
Sl
ve

90°4167°8166'12
LV'ceoe'Ze
8c vl'S¢98°0¢
S061 20°21
y0°8¢ LS

Ce c0°SE62'€€
c9'9¢ esl

ey

ayeulaj vivv0’!69'71BLSLeLk
86260°2¢
9e'Se 69°8

os’?
8"

60°6
cl ve

90°626€°8¢L9L€CL

GZS 6S'PSI's
r8'0l

ey
8201

€8°cecl
go’

ple 6z€
L8°1¢ 1S'°01

y8'ce
9e'l¢ y6'01

SI-€I
11-6 SI-€1 ZI 9I< SI-€I ZI

9I<

cal
[ei0L, arn
uoneonpy pure ‘xag ‘asy Aq uonendog *S’f 24) Jo sasejuad19g “GZ ajge]
 alge]="g"Z s1ydeizso0ulsqd SonsiiajoereyyD
JO 24} [[I-STVM UONLZIPIepueys

28
:ajdures sosejuad10g
Aq ‘asy ‘xag pure uoneonpy
ayeulay
[eo],
are
a3y Uu 88> Lis6 cE SiI-EL 9i= RBS cL RIE ice Some
cle le Sie O1=

ZI-9l 00g Ee Of Sel FS FOL

| We Or Wee Eel Ss| iv WOE Sse OES OOF
61-8t 00¢ i cee GOS
ae6 e065 0G Oh i SO OF GS SOLS 0ice Over OSE
ve-02
+ 00g el Se 24h Oe |O9 Ge SS O24 Ot2 OS| @O7 OW Wis OOr OR
6¢-S¢é 00g Orc OG Gl G7 CO 9 ae eS Gh SO te eG Ch SVS SO USS -S 00G
2 GiSG
vé-0F
+ 00¢ We Sip Gu CH Gil
| Gh OS OO! OL We SA? SG SHS “Site Ws
b-SE
+ 00¢ ic care
0 Oh C0 00 -Sic2 0:6= 0:S7 aeG:7 OS| Ge =20S OPE Sie O6¢>
vS-Sr
+ 00¢ Sy O89 @@ CL Get| Ge Gis GH! Ol |e WS Ge StS Wie We
79-SS 00g Re GL ee ONO ee OR Ce RO One asd Une Oks ate OR
69-S9 00¢ Wz CZ GW GY Ge Ohl Sk Ole 9G ce SA Sak Ors OE Oras
vl-0L 00¢ Od eS ere ee OG Ol ee ek el Ob RE Hoye ONS OA
Standardization and Norms Development

62-SZ 00¢ SS GL 6G Sin OM WG Se HO ele ONS OSE OAs PSST mS

emer PaO 0 eee CleatlO Ole ave CCOl b OC OOP eee CEs
0 CL) 280ste S=L0 Clb
Ooms OO is eO Oc Cie
O Seen COIS CTOOO EE CERO
0 50'S 016) HOEE “OiZEOZ
OU OSL

=
N
sI
si

Jo
uo

10J
pase

‘ION
sreak
GT—-9]
‘sIead
paseq

‘OSP‘Z
juared

Jaquinu
UONvONpa

UOneINpY
saouTureXe

/payadusoo

‘uoyeonpa
 aigel‘Le s1yde1s0ulag sonstiaj
JO 24}oe1eyD [TI-SIM uonezIprepueys
:ajdures sasejyuad
Aq ‘a8yi1eg ‘xag pure uoneonpy
aeulay yjevoL
11-6 ster [91
985 11-6 zi SI-€I 9IZ| 85

ore
11-6 zi SI-€I 9Iz

él
ZI-91 0Ol Ov 09 ost Joo — 06 OL OZ OZ| OF OSI OZ OZ Oz

OSL
61-81 OO 0% oe ODL Or| OF O9 OL OF OO!| OF 06 OZ OOF Ove
2-02 00k 0?

O'S}
og oz f|os o% OS O Oe OS| OF Ol OZ Ory OOl
62-S2 00 OL og or loz oF OS O8 OZ OZ| OF OO O9€ O92 Ove
ve-0E 0Ol OV 09 oz JO OF O8 OOF OFF OL| OS OFl O€€ 09% Oz
+b-SE 00

Ole Ost OZLL

0% 0% ot oer| O2 OF OL OS! OZ| OF 09 O26 082 ose
vS-Sr 00l 0? o's oz jos! Of O9 OS! OC Oe| OS OL O1€ OSZ 082
¥9-S¢ 00l OL 09 O86 OO| O9 OL O68 OF OL OL} OEL OLE O02 OL

0°020'91 Oe
69-S9 00! 0s OL oz O@]loOoOOL O12 O08 Ov OLL] OLL OLE OSL Ov
2-0 001 OL OZ

O'9oL
oO OOLlO906 Of 08 OS OLLI) OO O'lp OSL Ol
62-SZ 00k 06

O'8t
09 o8 Oloz06 O12 O06 OS OZ] OSL OSE OSL OZ
8-08 SL

O'VL
Ock £9 OF Lel|or€6 LBL 49 L9 Lee; O98 O08 LO Z0b

e'6
68-S8 SL oF EeszloRsoO Of O88 OV€]/08
LoL E12 OZ OO

es
‘alo parysiay
N= “0Sz‘l uoneonpyY
st 1aqumu
Jo srea ‘paja;duroo
10j saourutexa
pase 6[-9| ‘siead Uoneonpa
si paseq
uo juared ‘uoneonpe

29
Standardization
 a]ge]"9°Z saseUadI0g
Jo ay) °s’fQ uonendog
Aq ‘asy ‘xag pur Ayforuyyg/aoey

30
ore ayeulay [e}0],
ueoLyy UvoLyy UeoLyy
ay | ITM,
«= UeoLioury syuedstY
=: Jomo say ueseury dsuedsy{ PIO azTyM ueoaUTY dsuedsiy yO
JA\=8)! 6E°SE 662 se) 4 6L GV'ee Hed v9'S 98°} 78°89 9ZS GLb gore
6I-8L 9E ve 60°8 919 LLL le og 00°8 979 cL 98°29 60°91 ok 69 eve
ve-02 L6€€ €v9 evil S6L VE Ov 1QZ LZe9 00°2 89 ce york OLEL S6'e
6e-S¢e LOVE 919 cO'L Sle ve 8Y bel €v9 88h S069 LVEl Svel €0'r
vE-0E y9'SE GL svg QL} 9E 81 c6'9 8V'S €8"} LZ c8 L9°S1 tL €6 6S'°€
vy-SE PLING go's os’ BL} LE cy vS'9 vo'Vv L G8 EbGL 60°C S16 €9€
vS-SV €6°8€ Loy Lo€ €8'b 286 y9's Gee GL GL8L 920k cyl Lg'€
y9-SS 6E cv Lov c6C LOL lv ov eos ese SS} 78°08 S86 GZ9 9G’
69-S9 Le c8 ese LL?E 90°} LeoSv 01s 80°€ Z0°} 80°€8 €6'8 98°S el?
vl-OL Ze el ZVE ole v8'0 c0'8P 99° 9S°¢ | 0s’ G}'S8 Bl 897 ve?
6Z-SZ 9E EL c6S ech 98°0 cS'0S vy v0’? 66'0
Standardization and Norms Development

S228 72 as 9c'e S8k

8-08 Le 98° 68°C 8b 780 00°SS SOS vo? v8'0 98°98 vel os'€ 89°}
68-S8 Gok 02't 870 19 co Gc's 69'2 00°} €v'Z8 Oe 68°E 8b

‘azoxy ‘sq uoTefndod

eyepare wor jualiny uoynjndog
‘Aaains YouvpyG66T ajqepeal-auryoeyl]
elep [ayy Aq °S'¢ neaing
Jo ay} ‘snsuaD
‘SEET W;uTYseN
‘UO :Oq “SQ neaing
JO au} snsuaD /IeoNpoig) ‘(loINqINSIC
 31
Standardization

“po a
jou are eyep YOIYM UTI S|J9D “OSP‘'Z eae
JOU 319M PLa}119 JURAITAI BY} BUNVEUI SUOSIad OJ sjoo0j01d UONeZIPIepUr}s Jel] 9JBIIPUl peyiodai
os 016 Ol 0¢ Ov 019 a Or Ol 0°08 001 68-S8
ome Oe
28 0°88 20 0% Z9 Lys a £0 0% eee ost y8-08
£0 Lé
OL S68 Ol g0 Ge ges g0 ck ce 0'9€ 002 62-GL
G‘} 0¢
o'8 0'S8 Gh ce 09 0'SY = ct Se 0) 7 002 vl-OL
Gi os
S'6 g'S8 Ol 02 s'9 S'S a 0@ Oe 0'0V 002 69-S9
0} Or
GO S18 0 0 o's Sev 0 Ge cs 0'6€ 002 v9-GS
GZ gs
Ob g'8Z 0% GS? '9 0'6E Ol os cv S'6e 002 vS-Sv
oe Gl
Shh gZL st OV os S'6e ch Ov 9 0'8E 00¢ vy-SE
oe 08
Oel GSZ 02 Ge Gl 0'8e id Sv gs gle 002 vye-0e
Sv OL
Shh Orel Ghz 0 09 Sv G'9€ OL gs g'8 ose 00¢ 62-Se
Ov
Oe! Srl 002 S'0 s‘9 Gl GSE 0¢ 9 OL Sve 002 ye-02
G2
OL} 0°29 sh S's S'6 See 07 OL GZ See 002 61-81
ge Sel
Olt O'S S69 oe 0s gl Sve Gt 09 GL Ose 00¢ Zi-9t
Sv
§=a}TYM yO suedsty uRoeUry = ayTYM JQ «©uedstY«UROIeUTy «=aT u asy
JyIQ «=stuedsiyyYueopeury
ueoLyy uUeILyy UedLY
[e101 ayeulay ore
Ayoruyyg /eoey pure ‘xag ‘easy
Aq so8ejuao1eg :o[duiesg UOTeZIpsrepuUe}s [II-SIVM 94} JO SINSI19}9eIvY) asydeis0owlsg 9Ҥ'Z age
 “pajdeT[0o
jou
1AM BLIA}LID JURAITAI AY) SuTeaur suosiad 10] sjoo0}01d UOHeZIpIepUr}s }eY} a1eOIpUT pa}1odal jou are eyep YOTYM UT S]JaD ‘OGZ‘I = N PasysiaM “a1ON

0'9€LYVE 0'8¢
= Ov 08 0°88 a Le LQ 0°09 =o el car Gl 68-S8
cil Ov L9 0°88 ct Le Z9 ees a et oe 78-08
Gl
oe oe OL 0°28 oe = Ov 01S = Oe Oe 00! 6Z-GL

OLE
= Ov 06 0-28 re ae 09 0'0S oe Ov Oe 00 vl-OL

agM,OVE 0-980'9€ O'SE0'9€OLE 0°070'6€OlY

= 0S Ol 08 = Ov 08 O'eY = om Oe 69-99
00l
Standardization and Norms Development

Ov 09 OL 062 02 0¢ 09 0:0 0¢ Ov 0's 00t 9-SS

OV OL oxo} 0'6Z 0¢ 0€ 09 0'6€ 07g OV Ov 0Ot yS-Sv
Ov 06 O- O09 Ol 0S 0s 0'6€ Oe On 09 vy-SE
001
09 09 Oel ogz 07 0¢ OL 0'6€ Ov Ov OS 00! ye-0E
o€ Ove Ov 00L Oe 0S Oz Ose ae 08 OL OOL 62-Se
OV O'LL Ovk OL 07¢ 039 Oz OSE 0¢ 0's OL ve-02
00}
Ks Oc! O81 0°89 0¢ OL 06 O-¢E aS 0S 06 6I-8t

AVOTUY}W/aoey
(olay
o's 0°01 O91 0°69 O€ Ov 08 OSE 0¢ 09 08 001 Als}
yO duedsiY ueewYy day 1yIQ duedsyy ueoeUy ajTyM IQ duedsty ueooury uasy
uBoLIV UeOLy ueoLgy
[e10oL gjeulay ae

‘easy
‘xas
pure
Aq saseyued1eg :ajdures UONezIpepuryss [[I-SINM 24} JO SONST19}9v1eYD sIydeisoulsg “QZ ajgel
 33
Standardization

-(lOINqINSIG/JeoNpolg) snsuaD ay} Jo Neaing *S"f :Od “UOWuUTYseM,

“gf ‘aon
‘CEG ‘snsuaD au} Jo neaing °S’f Aq [aTy eyep a[qepeel-auTYoRIN]] S661 YAW ‘aang uounndog JuaLing Woy are eyep uoneindod
M S ON dN
UBILIOUITY UPILATY
uorsey sTydeas0e5y pure ‘AQoruyyq/aoey ‘ady Aq uoNefndog “sf ey} Jo sa3ejusdiod "LL" 3]gel
 “pa}de[[09
JOU d19M PLIAILID JURAGTaI BU) SUNVUI SUOSIed I0J sfooo}01d UONezIpIepue}s Jey} a}eoIpUT poy1odas jou are eJep YOTYM UT S[[9D “OSHC = N '270N
ce? aes — = Ol oz ae 0'¢ OOL 68-S8

Oe
eee = ne LOS = Le OSI 78-08

Ol Z0
es
en ec ae mee Be 0% : a cal Ge 00¢ 6Z-SZ

Gc Ol
on Ge = 002 yL-0L

Get
=n S10 Gt Oe
Standardization and Norms Development

G0, = 50 == Ov = O°} Oe 0°02 OLE GZ OLI 002 69-S9

OV
am

0} 5)
|
Oo CeEor Saw) 0:75. 0S = G¢é 0's O91 0°62 G02 091 00¢ y9-SS

0¢
ave
G0 <p GS) ai) Laas om 09 = Gc Gel 74 G'8e Gol 00¢ vS-SV

Ge Gt
ag 0c

0S
ET Sone gh 99 = Gc Oe 0°02 Ge Ge? Gol 002 vy-SE

SS) Ov Ge
rs
Fhe ashe ek S'0 G9 — gc 0°91 gle O'6t O'el 00¢ vE—-0E

aN Gc 0€ 0€ Ge Gc
= OL
Cae es sheet GO 2 08 22 ot GI Ge Gk 00¢ 6c-Se
b> = ae e el! ge MN ee Oe 00¢ ve—-0¢

Sv
0} Ov
SOc 0 oe om pee Ol SV 002 61-81

09 G8
Ope Sie) aS so 07) Gee OL GS 00¢ (A) sy)
S ON u asy
M S ON dN M ON

S
130 oruedsipyy UBDLIDUIY URIS
uolsay ITydeis0—ey pure ‘AQIOTUYA/a0eY
‘a8y Aq sa8vjuao1ag :ajdureg uoTezIprepue}s []J-SIVM 24} JO SoNsLiajoereyD s1ydeis0ulsd "ZZ ajqeL

sr)~s
 age] “¢1°z s1ydesZoureg sonsti
JO ajoere
4} WM yD [[]-S uOneZIpIepue
:ajdure
}ss sa8ejua
Aq d10g
‘asy
/aeYy ‘OT pue
A s1yde1Z005
uordsay
UBILyy UBOLIOWY sruedsipy

ON
L\i-91 001
61-81 001
ve-0¢ 00!
6¢-S¢ 001
vE-0E 00}
vb-SE 00l
vS-Sp 001
y9-GS 001
69-S9 001
bl-O0L 001
62-GL 001
v8-08 Se
68-8 GL

‘a]0N parySia,
N = “OSz'T ST[aD
UT YSIYMeyepare jou pawoda ayeoIpul
yey uoNeziprepsjoo0}01d
urys
10} suosiad BuNaeur
ay} JUPAg[al PHAIH9o19M
Jou *pejdaT[oo

35
Standardization
Standardization and Norms Development

Quality-Control Procedures
The quality-control procedures used in collecting the WAIS-III and WMS-II
standardization data were designed to facilitate proper test administration
and to ensure that the test responses were accurately scored.

Examiner Qualification Procedures

One of the first steps taken to ensure the quality of the administration was
to recruit examiners who had extensive testing experience. Before being
selected to participate in the standardization, examiners completed a
detailed background questionnaire that asked for information about their
education and professional experience, WAIS-R or WISC-III administration
experience, WMS-R administration experience, certification, and licensing
status. Those selected were very familiar with individual assessment prac-
tices and with WAIS-R, WISC-III, or WMS-R administration. The majority
were certified or licensed professionals working in private or public facilities.
Although the overall test format of the WAIS-III does not differ significantly
from that of the previous edition, a number of content and procedural
changes could have affected correct test administration. Moreover, several
of the examiners were unfamiliar with the WMS-R. Therefore, all potential
examiners were required to study a training manual and to receive training,
either in person or by videotape, on the correct administration of both the
WAIS-III and the WMS-III. The majority of the examiners also attended a
training workshop conducted by a member of the development team.
After training, each examiner submitted WAIS-III and WMS-III practice pro-
tocols. All practice protocols were evaluated within 48 hours according to
detailed guidelines that focused on the completeness of administration and
accuracy of scoring. Examiners received detailed written and oral critiques
of their practice protocols, the majority of which were correctly adminis-
tered and scored. To address the few frequently occurring administration
and scoring errors, the development team sent a newsletter to all examiners,
alerting them to potentially problematic areas.
During the data collection, the development team carefully checked all
protocols within 72 hours of receipt, according to protocol check-in forms
(one for the WAIS-III and one for the WMS-IID. Protocols were evaluated
for completeness of administration and recording and accuracy of scoring.
If necessary, feedback was provided to the examiners.

36
 Standardization

Scoring Studies
To refine the scoring criteria of those subtests for which many acceptable
responses are possible (Vocabulary, Similarities, Information, and Compre-
hension on the WAIS-III and Logical Memory, Family Pictures, and Visual
Reproduction on the WMS-III), the development team conducted several
scoring studies.
For the WAIS-III, two scoring studies were conducted. For the first study, the
development team used all 446 WAIS-III tryout protocols and the first 527
standardization protocols. Using the tryout protocols, the team designed a
preliminary coding system for the responses to the four target Verbal sub-
tests. For this coding system, the team expanded and refined the criteria of
the 0-, 1-, and 2-point scoring categories. The team then assigned a unique
code to each unique response or group of similar responses (i.e., responses
with the same salient elements). As standardization protocols were received,
responses were coded and categorized. If a response did not fit an existing
category, a new code was added. Two team members independently coded
each response, identified discrepancies between the code assignments, and
resolved the differences so that each response had only one code. At this
point, the team members had to agree on the grouping of responses and the
assignment of codes but not on what score value to assign to a code. As
appropriate, new codes were added and “redundant” codes removed.
After the codes were assigned, the team evaluated the quality of responses
and assigned a score value (0, 1, or 2) to each code on the basis of the accu-
racy of the response. The overall subtest performance and item-total corre-
lations were then calculated. For this purpose, items that had low item-total
correlations, poor IRT fit statistics, or other indicators of poor psychometric
suitability (e.g., bias) were not included in the total subtest score.
The remainder of the scoring study was an iterative process. The scoring cri-
teria for all items were continuously reviewed, with particular focus on those
items that had lower item—total correlations. Several iterations were per-
formed and resulted in changes either in the coding rule or in the items
making up the total score. After each iteration (i.e., changing a score value or
adding or dropping an item), item statistics were recomputed. Only one
change was made at each iteration. If the change in the scoring or in the
inclusion or exclusion of an item improved the item-total correlations and
the reliability of the subtest, the change was retained. When the final itera-
tion was completed, three individuals with extensive knowledge of Verbal
subtest scoring rules reviewed the final scoring rules, made minor modifica-
tions to eliminate any remaining areas of confusion, and deleted or modified
inappropriate and redundant sample responses. Additionally, the develop-
ment team reviewed the final item sets to ensure that content validity had
been maintained.
37
Standardization and Norms Development

The second scoring study was designed to replicate the results of the first
study, with a larger number of scorers. Using the final codes from the first
study, a group of trained scorers coded Vocabulary, Similarities, Information,
and Comprehension responses from an additional 1,038 standardization
protocols. Two scorers scored each protocol. When their codes did not agree,
a third scorer resolved discrepancies so that, in the end, each item response
was assigned only one code. (Interscorer agreement data are presented in
Chapter 3.)
The codes were then translated into score values according to the same scor-
ing criteria developed in the first scoring study. Analyses focused on overall
subtest reliability coefficients, item-—total correlations, and item bias. Results
between the first and second scoring studies were very consistent. Only a
few minor changes were made in the scoring rules on the basis of scorer
comments, and, in a few instances, new codes were developed for some
of the more “novel” responses. The final step was to drop the codes and
to return to the traditional score points. The remaining standardization
protocols were scored according to the traditional score points rather than
the codes.
For the WMS-III, multiple scoring studies were conducted during both the
tryout and standardization phases for the Logical Memory, Family Pictures,
and Visual Reproduction subtests. The scoring studies involved both expert
content reviews and empirical analyses. The empirical scoring studies
included evaluation of internal consistency, test-retest stability, intrascorer
agreement, and clinical sensitivity and specificity. A variety of alternate scor-
ing schemes were evaluated on the basis of expert content reviews and the
empirical findings.

Quality Assurance of Scoring and Data Entry

For all of the subtests on the WAIS-III and the WMS-III, the following
procedure was used to ensure the quality of scoring and data entry for the
standardization protocols. Each protocol was scored and the scores entered
by two well-trained scorers working independently. A daily discrepancy
analysis between the two sets of scores was run and discrepancies resolved
by a third scorer. Scorers received feedback on scoring errors and additional
training if needed. Over the course of the data entry, the average agreement
between scorers on the nonverbal subtests exceeded 97%, and the average
agreement on the four Verbal subtests (Vocabulary, Similarities, Information,
and Comprehension) exceeded 90%.
To prevent scoring drift, scorers did not discuss scoring rules with other
scorers and discussed responses about which they were unsure only with the
members of the development team. Most important, anchor protocols were

38
 Norms Development

used. An anchor protocol was an actual standardization protocol that had

been scored by the development team or by a well-trained scorer. Selected
protocols were compared to the scores of the anchor protocols. If the two
scorers made the same scoring errors on a protocol, then the comparison to
the anchor protocol revealed the scoring drift. Scorers received feedback
immediately to prevent the repetition of the errors and to correct for the
scoring drift. (An interrater agreement study was conducted to evaluate the
scoring performed in clinical practice. This study is described in Chapter 3.)

Other Quality-Assurance Procedures

In addition to the quality-assurance procedures just described, several other
procedures were employed to ensure the consistency of data handling. For
example, the computer program accepted values only within specified
ranges. After all protocols were double-scored, the scores double-entered,
and discrepancies resolved, the development team computed the frequen-
cies of all variables. A special “data clean-up” team further checked for any
out-of-range values accidentally entered into the data bank. Additionally,
subtest score outliers were statistically evaluated and carefully scrutinized.
The development team also randomly scored multiple protocols and
compared the results to the final data file. All of these quality assurance
procedures guaranteed the integrity of the WAIS-III and WMS-III standard-
ization data.

Norms Development
This section summarizes the procedures for deriving the subtest, IQ, and
Index scores for the WAIS-III and the WMS-III and provides the rationale for
the procedures.

Derivation of Subtest Scaled Scores

For most subtest scores, each age group's raw scores were converted to per-
centiles and then to a scale with a mean of 10 and a standard deviation of 3.
This conversion was accomplished by preparing a cumulative frequency dis-
tribution of raw scores for each age group, normalizing these distributions,
and calculating the appropriate scaled score for each raw score. For each
subtest, the progression of means, skewness, and variance values across the
age groups were examined, and minor sampling fluctuations smoothed.
The progression of scaled scores within an age group and from age group to
age group was then examined, and minor irregularities were eliminated by
further smoothing. The WAIS-III and WMS-III administration and scoring
Standardization and Norms Development

manuals present scaled-score equivalents of subtest raw scores. For some

of the supplementary subtests, the raw-score distributions were clearly
nonnormal; for these raw scores, the normative information is presented
as cumulative percentages.

Conversion of Scaled Scores to 1Q

and Index Scores
In the WAIS-R, the scaled score for each subtest is based on the scores of a
nonimpaired reference group composed of examinees aged 20-34 years.
Wechsler corrected the scores at the IQ level but thought that the subtest
scaled scores for adults should be based on the performance of such a refer-
ence group. Wechsler chose examinees 20-34 years old because he believed
that optimal performance tended to occur at these ages. In the WAIS—III,
however, the scaled score is based on the scores obtained by the examinee’s
same-age normative group. The WAIS-III, therefore, represents a shift in
how the subtest scaled scores and IQ scores are determined. There were
two reasons for this shift.
First, optimal performance does not occur consistently at ages 20-34 or
across subtests. For instance, Kaufman, Reynolds, and McLean (1989)
demonstrated that for some subtests, the age effect is not as important as
other variables. For the Performance subtests, especially for speeded tasks,
performance rapidly and steadily declines with age. For the Verbal subtests,
however, after the age of 30, the age effect decreases when education level is
controlled. In general, the Verbal subtest scores are less related to age than
are the Performance subtest scores.
Second, and perhaps more important, the method used in the WAIS-R often
leads to interpretive errors. Such misinterpretations are due, in part, to the
low subtest scaled scores that older individuals typically receive when their
scores are compared to the scores of a younger reference group. Compared
to the scores of their same-age peers, the performance of these same older
individuals might be average. This point is emphasized by recent research.
Recent data clearly show that the decline in cognitive functioning after age
50 and especially after age 75 can be quite severe (Powell, 1994). Powell
demonstrated that on a test of neurocognitive functioning, the performance
of highly educated adults (medical doctors) older than 75 years was 26%
lower than that of younger doctors. Storandt (1994) and Nettelbeck and
Rabbitt (1992) described the sensory and psychomotor changes that accom-
pany the aging process and concluded that many functions, including hear-
ing, visual acuity, response speed, and visual processing speed, all greatly
decline in the older adult. Most telling, however, is the work by Ivnik et al.
(1992) and by J. J. Ryan, Paolo, and Brungardt (1990). Their work involved the

40
 Norms Development

collection of normative data for the WAIS-R and WMS-R for individuals
older than 74 years. J. J. Ryan et al. (1990) found that the mean Performance
subtest score was approximately 1.5 SDs lower than the mean of the refer-
ence group for individuals 75-79 years old and even lower for individuals
aged 80 and older. Even for the Verbal subtests, the mean ranged from 0.5 SD
to 1 SD lower than the reference-group mean. According to these data, older
adults may appear significantly impaired or “abnormal” in subtest scores
when their performance is compared to that of a young reference group.
Their subtest scores, however, may actually be average when compared to
that of their same-age peers. Though Wechsler was aware of this difference,
he viewed the age-corrected subtest scores as optional (Wechsler, 1955,
1981) and corrected scores at the IQ level.
Ivnik et al. (1992) deviated away from the traditional method of reference-
group comparisons in calculating the performance of these older adults.
Instead of using the reference-group scores, they compared each individual’s
performance to the performance of others of similar age. If compared to the
reference group's scores, an examinee’s scaled scores would have appeared
extremely impaired even though the examinee’s performance was average
for the examinee’s age. By using a method similar to that used for the
Wechsler children’s scales (e.g., the WISC-IID), Ivnik et al. adjusted the scaled
scores for normal aging differences.
Similar problems have been avoided in the WAIS-III because the norms were
calculated in the same way as they were for the Wechsler children’s scales.
The correction for age occurs at the subtest scaled-score level, and these
subtest scaled scores are summed to calculate the IQ and Index scores.
(Comparisons to the reference group [20- to 34-year-olds] can be made, but
these scores should not be summed to calculate IQ or Index scores.)
The WMS-III also represents a shift in the way Index scores are constructed.
In the WMS-R, subtest scaled scores were not derived. Instead, raw scores of
the contributing subtests were summed, and that sum was converted to an
age-adjusted Index score. In some cases, raw scores were weighted before
the sums of raw scores were obtained. For the WMS-III, age-corrected sub-
test scaled scores were first derived and then summed to construct the Index
scores. The WMS-III therefore parallels the WAIS-III in the manner in which
Index scores are derived. This method of deriving Index scores ensures that
an Index score is equally weighted by its component scaled scores. As with
the WAIS-III, the score correction for age takes place at the scaled-score
level.

The addition of reference-group scaled scores at the subtest level in the

WMS-III is also new. Reference-group subtest scores are intended to be
supplementary and should never be summed to obtain Index scores.

41
Standardization and Norms Development

Derivation of !Q and Index Scores

The sums of subtest scaled scores for the WAIS-III IQ and Index scores and
the WMS-III Index scores were formed by summing each individual’s actual
age-corrected scaled scores on the relevant subtests. For each age group, the
means and standard deviations of the sums of scaled-score distributions
were calculated and are reported in Appendix A of this Manual.
The data on these sums of scaled scores demonstrated a high degree of
similarity within each of the scales and indexes. An analysis of variance
revealed no statistically significant variation by age group in the mean
scores for the scales and indexes. Moreover, the results of Bartlett’s test for
homogeneity of variance applied across the age groups indicated that the
variance did not differ significantly by age. Consequently, the age groups
were combined to construct the tables of IQ and Index score equivalents
of sums of scaled scores.
For each scale, the distribution of the sums of scaled scores was converted
to a scale with a mean of 100 and a standard deviation of 15. This conversion
was accomplished by preparing a cumulative frequency distribution of
actual sums of scaled scores for each scale and index, smoothing and
normalizing these distributions, and then calculating the appropriate IQ
or Index score equivalent for each sum of scaled scores. Successive adjust-
ments were based on computerized smoothing and visual inspection of
the distributions. Tables 2.14 and 2.15 list the subtest scores that contribute
to each of the WAIS-III IQ scales and indexes and the WMS-III indexes,
respectively.

Age-Adjusted and Reference-Group Norms

As discussed previously, two sets of normative scores were derived for the
WAIS-III and the WMS-III. The first set of normative scores is based on
age-corrected subtest scores. In general, when clinical questions dictate
comparisons of an individual's performance to that of his or her age peers,
the age-corrected norms should be used. Moreover, the age-corrected scaled
scores are summed to obtain the IQ and Index scores for both the WAIS-III
and the WMS-III.
The second set of norms is based on the performance of a reference group
that consisted of the participants in the standardization sample between the
ages of 20 and 34. In terms of key demographic variables (sex, race/ ethnicity,
education level, and geographic region), the reference group is representa-
tive of the U.S. Census proportions for this age range. In general, when
clinical questions dictate comparisons of an individual’s performance to
that of a reference group, these norms should be used. The reference-group
subtest scaled scores should not be summed to derive IQ or Index scores.

42
 ajqeL°2
L “h T]I-SIVMysayqng uoNT
Josodu
OJ i0D
pue Xapu] Se109§
jeuondgo
1sa}qnS Old OIsd IOA 10d INM

>
oY
ysaj}qns
ainyolg

ISd
UOHE|dWOD
Aseinqeso,n,

N
1161q Bulpo9—joquifs
SONUE|IWIS

as
yoo\g ubiseq
HOW

>
S
xuyeyy Buluoseay

S
y161q ueds
UONEUOJU]

Sis
> p>
anid juewebueHy
uolsueyesdwioy

SS
>
joqwks Yeas
2
Jequin\—Jee7
bulouenb
, as
j991qQ Alquessy
,
Wi

, aul joquiks ‘yoIeEs Jequinn—seye7‘Bulousnba

pue g }0aIqO Ajquiessy s}sejqns
UeD aJNNSGns
40) 484} s}se}qNsJepUN UIEUSO “Se0UR}SWUN
J8J9
0}OJIO
eu |I|-S/Y/M uOHeNSIUI
puke LUPYBuljoog /enue~
JO}B ‘UO!SSNOSID

43
Norms Development
 a a er a aa a
8109S |e}0] |/e08Y

|yseu1U0DJse]U0D
1 day siA

Z
adojs Bulwee7
a109S |e}O] |e98y
2100S |e}O| |/P99Y IS}
| S}SI7] PIOM
| Seinjaidg Ajiwe4y
. adojs Bulwee7

||2O9Y
.@109S
|P}OL
IS}
y) /- 8100§ |E}OL (2094
| 20ssy padied jequa;
Standardization and Norms Development

Vo . @109S |e}0} UoIUBOoeY

ewe)
.a100S
|e}O]|seoe4
, aha Soc ormes,
. G10 |e}O] |JEOSY 1S}

Asowayy
je016074
Ye J 8109 Je}O] ||208Y

UOI}EJUIO
|
. 9109S |e}OL

UOTIsodur0D
d}Isodui0D YOjU]
$9109S uolUua}ay eAaleY —adojs Hulusee] wef wal 994 jad \eq \eq wa wu wu| ysayqns

ITI-SIWM
jeyuawajddns jelsl-ajburs Huluseey BM uas) pny SIA pny wu] SIA pny

3s91qNS
$3109¢
xapu]
3/Gel
"GL'Z
sa}Isoduio’) A10}Ipnyssad01g Areullig saxapu]

pure
3 ~J
JO
 3iqel “GL'Z [II-SWM 3Sa3qQN¢ UOHIsodu10D
JO xapu]pue a\IsoduIoD
$d10Ig (panuyjuod)
Areullig soxopuy A10j}Ipnyssa001g sayIsoduio’)
pny SIA wy = pny SIA pny ua) BYM jel4l-ajuis
= Hujwee]
yse}qns Www] wu] wel 19q leq 984 Jaq wa) Waly Hujuses) adojs
© |eAaljay uoluajey
N-7 Burouenbes
- $9109S
|e}O). @109S Va
jeyuawejddns

jenedsueds
Puenuo4
|e}O] 9409S
Puemyorg
|e}]O) 1090S
> Ss

|e}O). 1090S Wp
[e}UE/ JO1jU0D

|e}OL
8100S

1161
ueds
|e}O]
{|
@109S
1/2084[210] 8109S , a

jeo1607
Asowayy
UO!|UB}OY JUdDEq ver
dewey]
Je}O, 9100S
607 qlap/way
Jed dossy
Jp
pe
B109E
pny
98y
|9q
|2}OL
|eASIOY
Je}O] 8100S Wo

|)
UoN!JuBodeY
[BJO] 1095 . .

jeqle,
pailed
DOSS
I]

JUGWE
|;eo8y|e}O] @109S ,. .

se0e4 UOIJUS}JOY
UONUB}JOY JUBDOd Ws
Norms Development

AS
 a1geL "GL'Z [II-SINM 1Sa}qNs uoTIsoduio7
Jo xapu]pur a}yIsOduioyD
S2810I¢ (panu1JUOI)

46
Areullig soxopuy A10}Ipnyssa001g sajIsoduio7))
pny SIA wiw| pny SIA pny uas) BM jell-ajbuis Huluseey jeyuawiajddns
yseyqns wu| Ww wel Esa] 19g 990y |2q Wea waif Hurusee7 adojs jeAejey uonusjey $8109S
Ajjwe4 Saunjoig
I]
ee t—C ee
a
JUsDEg UONUEJOY .
PAOM SISIT
IL
peAejaqjjeoay Ya

JUBDIBd UONUSIEOY a

si, Je}O]
11day 8109S
j/e0ey
Standardization and Norms Development

Je}O)
9109S
SSIS

UOIFEUILUOSIG

UoNIUBoDeY
||SSP SS1NS

JUDIE
Adoy UO!|US}OY
 CHAPTER 3

Reliability and Score Differences

The statistical properties of the WAIS-III and the WMS-III presented in this
chapter determine the confidence examiners can have in the accuracy of
obtained scores. The psychometric properties that are critical for the inter-
pretation of scores are reliability and stability coefficients, standard errors of
measurement, confidence intervals, statistical significance of the differences
between scores, and frequency of score differences (i.e., base rates). This
chapter reports and discusses these statistics as they relate to the quantita-
tive interpretation of scores on the WAIS-III and the WMS-III.

Reliability
The reliability of a test refers to the accuracy, consistency, and stability of
test scores across situations (Anastasi & Urbina, 1997). Classical test theory
posits that a test score is an approximation of an individual’s hypothetical
true score, that is, the score he or she would receive if the test were perfectly
reliable. The difference between the hypothetical true score and the individ-
ual’s obtained test score is measurement error. A reliable test will have rela-
tively small measurement error and consistent measurement results within
one administration and on different occasions. The reliability of a test
should always be considered in the interpretation of obtained test scores
and differences between an individual’s test scores on multiple occasions.

Reliability Coefficients
The reliability of each WAIS-III subtest (except Digit Symbol—Coding and
Symbol Search) was estimated from the item scores from a single adminis-
tration. The items of each subtest were first rank-ordered on the basis of
item response theory (IRT) difficulty estimates. The subtest items were
divided (by an odd-even split) to form two half-tests. The variances of the
two half-tests were also compared to ensure that there was no significant
statistical difference. The reliability coefficient of the subtest is the correla-
tion between the total scores of the two half-tests corrected by the
Spearman-Brown formula for the full subtest (Crocker & Algina, 1986).

47
Reliability and Score Differences

Because Digit Symbol—Coding and Symbol Search are speeded subtests,

the split-half coefficient is not a proper estimate of reliability. Therefore,
test-retest stability coefficients were used as the reliability estimates for
these subtests. These stability coefficients were based on the scores of four
age groups (16-29, 30-54, 55-74, and 75-89) participating in the test-retest
study described later in this chapter. The stability coefficient is the correla-
tion between the scores on the first and second testings corrected for the
variability of the standardization sample (Allen & Yen, 1979; Magnusson,
1967). For each subtest, the overall mean coefficients across the 13 age
groups were calculated with Fisher's z transformation. For each of the
normative age groups, the reliability coefficient is the one for the broader
age band that includes that normative age group.
For the WMS-III Primary subtest scores, the split-half internal consistency
methodology just described was used to estimate reliability coefficients. For
a number of WMS-III supplemental subtest scores, however, internal consis-
tency is not an appropriate measure because of a variety of factors, such as
item-presentation format and dependency between items. For supplemental
subtest scores, therefore, reliability estimates were computed according to
generalizability theory. Generalizability coefficients, first introduced by
Cronbach, Rajaratnam, and Gleser (1963), can be considered analogues to
traditional reliability estimates (e.g., Brennen, 1983; Cronbach, Gleser,
Nanda, & Rajaratnam, 1972; Franzen, 1989). With this methodology, sources
of variance were partitioned according to the familiar analysis of variance
(ANOVA) method. A repeated-measures ANOVA method and the scores
obtained by participants in the standardization sample who completed the
WMS-III on two separate occasions were used to calculate the coefficients.
For these analyses, this subsample was divided into two age bands, 16-54
years (n = 141) and 55-89 years (n = 156).
The reliability coefficients of the WAIS-III IQ scales and indexes and the
WMS-III indexes were calculated with the formula recommended by
Guilford (1954) and Nunnally (1978). Table 3.1 presents the internal consis-
tency reliability coefficients for the WAIS-III subtests, IQ scales, and indexes,
and Table 3.2 presents the reliability coefficients for the WMS-III Primary
subtest scores and Primary Indexes. Table 3.3 presents the generalizability
coefficients and standard errors of measurement for selected WMS-III
supplemental subtest scores. For the WAIS-III and the WMS-II, the overall
reliability coefficient for each subtest, scale, and index is the average coeffi-
cient across the 13 age groups calculated with Fisher’s z transformation.

48
 Reliability

As the data in Table 3.1 indicate, the average reliability coefficients of most
of the WAIS—III subtests (except Picture Arrangement, Symbol Search, and
Object Assembly) range from .82 to .93. For the Vocabulary, Digit Span,
Information, and Matrix Reasoning subtests, the coefficients are extremely
high (=.90). The coefficients for Arithmetic, Comprehension, Letter-Number
Sequencing, Picture Completion, Digit Symbol—Coding, Similarities, and
Block Design range from .82 to .88. Symbol Search has an average test-retest
coefficient of .77, which is relatively high for test-retest reliability. The co-
efficients for Picture Arrangement and Object Assembly are lower, .74 and
.70, respectively, but are equal to or greater than the reliability coefficients
obtained for the WAIS-R subtests. The low reliability of Object Assembly for
older adults contributed to the decision to exclude this subtest from the
computation of IQ and Index scores.
The average reliability coefficients for WAIS-III IQ scales and indexes range
from .88 to .97 and are generally higher than those of the individual subtests
that compose the IQ scale or index. This difference occurs because each
subtest represents only a small portion of an individual’s entire intellectual
functioning, whereas the IQ and Index scores summarize the individual's
performance on a broader sample of abilities. Therefore, the extremely high
reliability coefficients (in the .90s) for the WAIS-III IQ scales and indexes are
expected. The somewhat lower reliability coefficient for the WAIS-III Pro-
cessing Speed Index score (r = .88) is expected because of the relatively small
number of subtests composing this index. Overall, the WAIS-III has higher
reliability coefficients than does the WAIS-R.
The WMS-III reliability coefficients for subtest scores that contribute to the
Primary Indexes are presented in Table 3.2. The average reliability coeffi-
cients across age groups for these subtest scores range from .74 to .93, with a
median reliability of .81. The average reliability coefficients for the Primary
Indexes range from .74 to .93, with a median reliability of .87. As expected,
the reliability coefficients for the Primary Indexes are generally higher than
those for the individual subtest components that compose them. All of these
coefficients represent a substantial improvement in reliability compared to
the reliability of the WMS-R scores. The reliability coefficients for the sup-
plemental subtest scores that were calculated from the generalizability study
are shown in Table 3.3. These average coefficients range from .72 to .87, with
a median reliability coefficient of .77. The average supplemental subtest
scores generally have lower reliability compared to that of the subtest scores
that contribute to the Primary Index scores.

49
 ‘papodas ase sjUa191J900 Ajyigeljal JSe}e/—js9} ‘YoIeaS joquAS pue Bulpog—joquids 161g 404 gq
“UOIPEWOJSUBI] Z SJOUSI-4 UM peyejnajeo asm sjUa!dijJo0o Ajigelje1 eHeISAY ,

1OdIMISd
28° 88" 68° 68° 88 68° 06° 06° 06 68° 68 ZS" 98" 98" 98°
ce ve’ 16" ve 6° v6 ve ve" G6 €6 ve €6° io). So €6
€6 06° c6 6 ve Se G6 €6 ve €6° G6 v6 Aon co

IDA
ve"
96° 96° 96° 26 96" 96° LO 26 96° 96° 96° 96° 96° 96° ve
OISA
86

86° Z6 86° Z6 86° 86° 86° 86° 86° 86° 86 86° Z6 LO

Old
v6

v6 ve" ve €6" G6 96" S6° ve v6 €6 sé c6 €6 €6:

DIA
el LO

Z6 96° Ae) Z6° Z6 86° 86° Z6° LE 26 L6° Z6 LE 96°

vO
OL 0S v9 6S 89° He: ol BL LL GL Ee EL O75 ce

qaSS
cL His

GL
ye
6L
c8
vL
vl
ol
ce
La
Yay
(oye
VL
vl
Le
GL
08°
WL
OL
6L
99°
Lele
Hehe
6L co co 68° 68

06° ve 68° 06 Lo ve €6 68° L6 88° Lo 88° 68° Z8°

98° eye LS 6L es Z8° 88 Ge 88 88 06° 06" 88 88°
vs Z8° ZS) Z8 98° 98° 98° v8 ve vs Le 18 Le LS
Reliability and Score Differences

€8 cs os cs Z8° 88 28; +8 8° 6L 98° Le SVE co

c8 Z8° 98° GL 88° BL 6L 98° SZ c8 Le LE 88" jij,
G8"

v8 Cz 98° Lo 98° Z8° 08 es vs 98° Ach 98° 98° vs

xepul/ereosAseiqns
eo ve 88° Le co

Lo 6 6 Aon 06° ee c6 68° 68° €6" Leo 6" 06° 68°

06° vs Z8° 06° 06° €6 eo 6 Aon 06° c6 06° Le 06
88° ee, Z8° 98° 68° Lo 6” 06 06° 88° 68° ZS 88° 88°
98° 68° 68° 98° Ss Z8° 88 68° 88° ie} 98° cs So Le
6 €6 ve €6 €6 sé v6 Aon €6" ve Ace ve €6° 06°

aousiajay
dnoiy

ZI-9L
abelaAy
68-S8
H9-SS
69-S9
pl-0L
62-SL
8-08
PE-OE
Ph-SE
PS-Sh
6I-8L
2-02
62-S2
ed
aby
dnoiy
dnory Aq a8y
saxapu] OI
TII-STVM ‘S}S93qNS pure
‘sayeag JO 24}
syuaTIyye0D ATIqeIeYy age “LE
 alge]"Ze AWTGeyeYy syuarayja075
Jo 2up [I]-SIWM [Link] }sajqng sa10og
pue Areullig saxapuy
Aq a8y dnoiy
SSS
EEE

aby dnoiy aoualajoy

obelony
xapuyjseiqns I-91 G6I-Bl HZ-0Z G62-SZ vE-0E pr-SE HS-Shb H9-SS 69-S9 pl-OL 62-SL 8-08 68-S8 oul dnoiy
je01607. Asowayy
|
|;eo8YJe}O] @109S 06°
88°
88"

98°
Le

98"
L6

98"
SB

88°
06°
88°
68°
seoe4
|

88"

248°
uoUBooeY
|e}0) 8109 (fA BL So" OL yl 9L°

62
6L
gL
LL

ZL
08"
89°

69°
69°
jeqion paileg Seyeloossy
|
|;e984|e}O] @109S €6° £6"
€6"
c6

6"

L6"
16"
Ajjwey

sé
Saunjoid

S6
|

£6"
v6"
€6"

v6"

v6"
96°
|}e08Y|e}O] 8109 S8 €8" 1s" v8"

gL
el
LES
88°
EL
eZ

98°
£8"
ak:
BL’
98"
Jequiny-19}}07
Bulsuanbas
|P}OL 8109S 88° c8 48° cs 6L

bsih
jeneds

Pas
ueds

62

LES
8Z

98"

SL’
Gh

88°
|e}OL Q109S 98"
3" Se" LL 6L Se"

€8
SB
kL

ce

SVE
el
gL
6L

Lg”
LL
(291607 Asoway
1)
|e08Y Je}O] 8109S SZ €8 6L gL

LL
9L

GL
el

c8
se0e4

bl
BL
|i
£8

LAs
08"

bsLid
UoNUBooeYy
|e}]O, @109S ce vl 69° pl gL’

kL
OL
18

LL
kL

£8
cL
ag)

69°
89°

jequap,, paued Sayeioossy

||
|ye098y|e}O] 8109S c8 88" 9Z
es"
98"

el
62
ce

ee"
98"
Ajjwey Seanjoig
I

88"
48°

v8"
28°
08"

|e08y e}O] @109S £8 le" v8 78°

cL
ce
BL

c8
62
c8
62

Ss"
98°
v8"
06°

Auoyipny uoniubooay
paAejag
|e}O] 8109S OL c8" pl gL

BL
69°

OL
eZ
62

BL’
Auoyipny ayeipawiu| sé 6" 6 v6"

62 ve
06" Se
v6"
S6"

jensi, ayelpawiuyy C8 Se v8 cs" 3"

BL

vs"
6L

v6" 98"
ayelpawiuyMioway ley Le 6” ce

6" €8" 6”

v9" €6" 08 L6”

98°

v6
Le
Asoypny pefejag 88° Se 28° Ls

Se
Se
oF 68°

vl 6" SB c6 06"
28°
jensi, peAejaq €s 98" ee es"
co" 98" 8°

62

Ze
v8" 6 06" Se

08"
28°
Asojipny uouBbooey
peAejeq 0L 8 17 gl

cs v9
OL

69° 8Z
6L’

jeseue5 Kiowa Le” 6 Le ce

Ze
+9" co
Bupon Aioway £8 98° 88°

Le
S8 8L c6" 8°
bl £6" Se
c6 vs

€8°
eZ 16" cs

OL €6" 1s 6 Se Ls ol 06° ZL
£6" 88

68°
6L°

c6 88° c8 62 c6 28°
IL c6" 62 68° 28° 08" LL 06" 88°

, eBereny Ayijiqeijes sjusio1ye0d

a1eM Pajeingjed
YIM SJOYSI4
Z "UONEUUOJSUBI}

Lo
Reliability

=
Reliability and Score Differences

Table 3.3. Generalizability Coefficients and Standard Errors of

Measurement of Selected WMS-III Supplemental
Subtest Scores
Generalizability Standard Error
Coefficient of Measurement
Ages Ages Ages Ages Average
Subtest 16-54 55-89 Average? | 16-54 55-89 SEmu
Logical Memory |
Thematic Total Score he) 78 BUT 1.50 1.41 1.45
Word Lists |
Recall Total Score 76 .82 19 1.47 1.27 137
Visual Reproduction |
Recall Total Score ae. .84 19 1.59 1.20 1.41
Spatial Span
Forward Total Score ne LAS) 74 1.59 1.50 1.54
Backward Total Score 72 afl WE 1.59 1.62 1.60
Mental Control
Total Score 85 89 87 146 0.99 1.08
Digit Span
Total Score .87 84 .86 1.08 1.20 1.14
Logical Memory Il
Thematic Total Score 78 19 19 1.41 Loe 1.39
Word Lists Il
Recall Total Score 19 .80 .80 137 1.34 1.36
Recognition Total Score Til 81 76 1.62 13 1.47
Visual Reproduction Il
Recall Total Score 70 82 sift 1.64 Lee 1.47
Recognition Total Score .65 82 Aras) ELT Uy F/ 1.54
Copy Total Score 72 74 76s 1.59 ids! 1.56

“The average was computed with Fisher’s z transformation.

52
 Reliability

Standard Errors of Measurement and

Confidence Intervals
The standard error of measurement (SE,,) provides an estimate of the
amount of error in an individual's observed test score. Because the standard
error of measurement is inversely related to the reliability of a subtest, the
greater the reliability is, the smaller the standard error of measurement is,
and the more confidence the examiner can have in the precision of the
observed test score. Measurement error is commonly expressed in terms of
standard deviation units; that is, the standard error of measurement is the
standard deviation of the measurement error distribution. The standard
error of measurement is calculated with the formula,

SE = SD 1l-ry,

where SE), represents the standard error of measurement, SD is the standard

deviation unit of the scale, and r,, is the reliability coefficient of the scale.
Comparisons between the standard errors of measurement of the subtest
scaled scores and the IQ and Index scores should not be made because they
are based on different standard deviation units. Because the standard devia-
tion is 3 for the subtest scaled scores and 15 for the IQ and Index scores, for
both the WAIS-III and WMS-III, the standard errors of measurement of the
subtest scaled scores usually appear smaller than those of the IQ and Index
scores. In fact, the IQ and Index scores are actually more accurate measures
than any of the individual subtest scaled scores.
The standard errors of measurement for the WAIS-III subtests, IQ scales,
and indexes are shown in Table 3.4, and for the WMS-III, in Table 3.5. The
standard errors of measurement for those subtest scores that were used in
the generalizability study are provided in Table 3.3.
The standard error of measurement is used to calculate the confidence
interval, or the band of scores, around the observed score in which the indi-
vidual’s true score is likely to fall. Confidence intervals provide another
means of expressing the precision of test scores. The examiner can use confi-
dence intervals to report an individual's score as an interval that is likely to
contain the individual’s true score. Confidence intervals also serve as a
reminder that measurement error is inherent in all test scores and that the
observed test score is only an estimate of true ability. For example, if a 64-
year-old examinee obtained a WAIS-III FSIQ score of 102, the examiner can
be 95% confident that the individual's true IQ score falls in the range of
98-106 (because the 95% confidence interval is 102 + 1.96 SE,,, where the
SE, is 2.07), and 90% confident that the individual’s true IQ score is in the
range of 99-105 (102 + 1.65 SE,,). Confidence intervals based on the standard
error of measurement are calculated by the following formula:

53
 "\INSO1 a4} JO Joo! arenbs ayy Hulureyqo pue dnobh afe yowe 10) s/Z4g paienbs ay} jo wns ay} Buibesene Aq payeinojeo aiom S39 abelene OU! ,
‘JUaWeINSeeW JO SJOe PuepUe}S ay} ayndwoo
0} pasn a1am (Se100S xEpU] PUL H] 9} 10} G} PUe S}se}qns au} JO} E ‘“9"!) UOIWEIAOP PuepUe}s UOe;Ndod ey) puke "Ee aIqeL UI UMOYS S}UEIDIJJe09 AjI|IGel|o1
SY] ‘S9JOOS XAPU| PUL H} AU} JO} S}lUN xapUl/H] Ul PUB S}Sa}qNs eu} 410} S}JUN aJODS-pajeds U! peyodel ale JUBWAINSeAW JO SJOUE PJePUR}S BU ‘BJON
9e'S els LOS =80'S- OS S60 Zip S88 SZP™ S6ipe 00S OVS m00Cenc aS OSS ISd
yee yee Chr 22ZL6e privy. cVSES S698 LCR Si Gah nr 8 Cin [Link] OG Campi Ole mee ence OOne, INM
6LE G6 Ly Liv? ley OLE. vee “SEC Z26'CR Wo Cm™ SO re SV Ge acoCmeec caer cOliz lOd
c0'e LOE €6¢ -8l¢: €6¢ che 69 —“L2i'c- WE. Oe S8e S08 MOLE Sone tee IDA
OlSA
CSIC
cicmumCciCumm
0g"?
60°C
NEC
Scice
BLOC”
0GE
Chics
Lvie
lee
9G
NO
Com
CR
purl
OOry
oOlr,
GLE
29
Old
LEE
69
OV
10h.
Caco
Ser
10
JC
S75
cS
Gi
2
C=
C2
Sa
Cer
mm
©
ae
ose gcc vOE cGdc= GSie VOCo (Ole MSeiCae SiGe SZ Vice OL Ce Meyce OTIC Ov CameoG OIA
9S" 99 Clic O8ihe 66 02 Tare OSs ee Lah Coe OSE Sc Opa Siete on eee ae VOR
SVL evit OSH: © OSEE™ OSI ZSi Ee Ze hee Catan cates aeCec eee Go me Ts en Com ern Ss
Oot ec i Pech 6G ee G7e be VV ev Gab ee OSE be OSs eee Va bee beer ed COR emia Vd
660 Z6'0 Oc | = 66:0" 96:0 (06:0 "82105 S640 Se C60 R06OF 0rls 060 seeOs ee OG OmenCOnE YW
860 vi 2b hee SLeake sale SOs ae Ose Oe iO eee0te O10 eeO10 ee On ame ORT: dd
Zot 6b L 80'l = 80'be 80} chit @Clit CE (Odi Ee SOc ee OGals tae C: Perc seme cea ca ate)
2 Ze k Sol 2G eCLGibee LO knee COS VOsh a COU Sbee ie Os rmieZ Chena Cihe PaeeetCae mei emer Od
= 6e'} Oe l 80iIL = Gkls OS} LO Sty SSZesE Cele OS) Pi meZ ci ayy eerie eee On meena NI
5 JE Lob LE Oh Vee oh bees eOrhe Verkan VCnes OG RnN Chub NeCalan Eat mana ten OGRE =D)
= Z8°0 L6°0 06:0 — 060 S80 S6i0% 8620) SS8i0D GEOF 66iI0RNGZA0) §06 OR SOGiON meSGOmGGI0 |
= c6'0 v6'0 Ogre: 908k S60: [Link] 620-5620 06: 002558Os 56102 S807 G60 ae OG SE SO0 Sd
5 vO'L SO'L wee SO) Chil G60) O6i0 OGIO S6i0s S60 70 66 Oi oOr eeOs Pe 0nr Vv
DB 6b ok k 66,02 66.006 29 bess COtbeas VOLE 6602 eeVOilaa nS he onc hiene? Cale ba eemereoar S)
E ZL0 620 64:0 —~£0 620 ~ 6210) “29105 EZ0= SSi0m 62 ONE On SiO mc 7 OmmO7Ome Gol0 A
2 dnoiy e"ZS 68-S8 8-08 62-SL bl-0L 69-S9 9-SS HS-Shb HH-SE VE-OE 62-SZ 2-02 GI-B8l LI-9L xapulejeospasayqns
aouaiajay
oabeliany
aby
dnoiy
=
©
Po dno1y 33V Aq S9d100S xXo9puy pue Ol “ys97qnS ITI-SIVM 294} JO JUDUTOINSva|] JO SIOIIG Prepuejg “HE ajigel

34
 ‘JUBsWSINSeew
\INSB1 By} JO JOO1 aseNbs ey) Huiule}go pue dnoih abe yore 10} sV4Sg paienbs 9yj Jo Wns au} BuiBesane Aq pajejnojeo aiom SV/Zg efelene oy %
Reliability

JO SIO PYePUE]S BU} a]NdWOD 0} pesn a1aM (SE10NS XEPU| BU} JO} GL PUB S}SEe}qGNs ay} JO} E ‘*8"l) UOIVEIASP Psepue]s UOHe|ndod au} pue Z's aIGeL Ul
UMOUS $]UBIDIJ909 AjI|IGel|e1 SUL “S@XEPU! BU} 10} S}UN XEepU! U! PUB S}sa}qns au} JO} S}1UN aJ09S-pajeds Ul payodei ae JUBWSINSeOW JO SIOIE PYePURSs OY] “AJO//

Bupom
Kioway
uoljiubooey

Aioway
jesoues
08"} LL'9os'yLBS96900°6 os'y
8's46°¢00'S819co8os’L6°V

29°}bly9e°9L9°Sgel Se°800°9819
00'S Z8°9verLys

volevLvs

Aojipny
les 69'S Lys 02'S 189 02'S 9e°9 Lys 00°9 18'S

peAejeg
Lys
ce'y 9S'7 L6€ ZOE veov bly OS’ bly vev Z6'E

pafejag
969 189 vO'L 80°8 6LL 28 228 Go'L

jensi
Or'Z WR

Aioyipny
ayeipawiwy

pafejag
28°9
L9°800°9vey 9¢°9

18S

Aioway,
Ae) 029 L9°S 18'S LL9 LZ9 ys'9 ys'9 Lys Lys

Sye|paui]
os’18S
L's Lg°¢ 18'S bly 167 Ly’s 19'S 18'S Ly’s rly

ayeipewiuy

jens!
Aioypny
vev vvy os'y LEE Os’ L60 vev 0s'y LOE vev

ver819

uoiiubooey
00°9 Leg 00'9 00°9 18°9 18°9 v0'L pS'9 19'S 18'S

A
L6°E
See

@100S

Auoyipny
LUE 88" LEE See See vev 249'€ L6°E L9°€ Z6'E

Ajjwe4peAejag
Je}OL
Zel Ze'h
v9}

Lvl
Srl 7S} cae Ze yh 29 t 9g} p9'L rot eS

|/E08Y |/E99y
ISaanyoig
8109
|B}0]
06'0 60 1o 02'L 22k Z€'\ ra Olt

yh
a vz}

SayBj90ssy

UOoNluBooeY
81095
|P}O|
seoe4 jequa,
pale
vi Sol Ly} vel 80°} 80°} Leh 02'} Ze} vO'l clk

@109§
I]

|e}O)
Shh 7S'| 19°} OL'| GZ‘ 19'| 29'| 69'| Leh vz'|

|/e99y
@109S
Asoweay
|B}O|

jenedsje01607
I]
6h" 1 Ov} 2' 80'L ve'l by’ LY’ €s'} lz k 02'L

sequiny—198}907]

8109S
||

|e}OL
Bujouanbas
Fa 6€' yb'| LET, 29h os'| 29't Le" 9S'} 22k

ueds
1095

|e}OL
6e'} Oe'l 80'} Ae os't y0'l Ly'h LE't A! OS'L

|jeE084
8109S

|saunyoid
|e}O|

Ajwe4
cok ve'h vol clk 9S'L byt Ly't byl vel clk

Seye}oossy

||e08Y
UOIIUBOOeYy
|P}0| @109S
|e}O|
jeqle/,
pale,
€8°0 820 620 09'0 620 €Z°0 €2'0 249'0 £20 29'0

8109S
|

|seoe4
97} ys} v9" LL} OZ't L491 L9°} 29} veh byl

xepuljseyqns
|/eO8Y
@109S
Auoway
|P}O|

jeo1607
v0'} L0'} S60 70'l 060 Leh 80'1 9th 66°0 vO'l S60

JUaUaInsvay
|
dnoiy

dnoiy
eZ 68-S8 8-08 62-SL tl-OL 69-S9 9-SS HS-Sh Hr-SE HE-DE 62-SZ 2-02 GI-Bl I-91

aby
aouaiajay abeloay

prepuels
[[I-SINM
AreUuILIg
AreUuILd
saxapu]
}sa}qng
sa10og
S1O11q
age]
dnoiy Aq a3y

pue
“G'E
9)
Jo
JO
Reliability and Score Differences

Confidence Interval of p = Observed Score + zp(SEy),

where p is the confidence level, such as 90% or 95%; and zp is the z value
associated with the confidence level, which can be located in the normal
probability tables.
The confidence intervals provided in Tables A.3-A.8 in the WAJS—III
Administration and Scoring Manual and in Table E.1 in the WMS-III
Administration and Scoring Manual were derived by a slightly different
method. The 90% and 95% confidence intervals for the WAIS-III IQ and
Index scores and the WMS-III Index scores are based on the estimated true
score and the standard error of estimation (SE;) according to the method pro-
posed by Dudek (1979) and Glutting, McDermott, and Stanley (1987). The
estimated true score is obtained by the formula,

Estimated True Score = 100 + r,, (X — 100),

where X is the observed composite score, and r,,, is the reliability coefficient
of the related composite scale. The standard error of estimation is derived by
the formula proposed by Stanley (1971):

bp S15 Wl reat

where SE; is the standard error of estimation, 15 is the standard deviation of

the composite score, and r,,is the reliability coefficient of the related com-
posite scale. This method centers the confidence interval on the estimated
true score rather than on the observed score, and in turn, results in an asym-
metrical interval around the observed score. This asymmetry occurs because
the estimated true score will always be closer to the mean of the scale than
will be the observed score. Therefore, a confidence interval based on the
estimated true score and the standard error of estimation is a correction for
true-score regression toward the mean. Because the reliability of the
WAIS-III and WMS-III composite scores are relatively high, the confidence
intervals calculated with the standard error of measurement centered on the
obtained score and those calculated with the standard error of estimation
centered on the estimated true score will be very close. Also, confidence
intervals calculated by either method are interpreted in the same way.

Test-Retest Stability
The stability of scores of the WAIS-III and the WMS-III was assessed in sepa-
rate studies. For each instrument, participants were tested twice, with a
test-retest interval ranging from 2 to 12 weeks. For the WAIS-III, the mean
 Reliability

retest interval was 34.6 days, and for the WMS-III, 35.6 days. The WAIS-III
sample included 394 participants, with roughly 30 participants from each of
the 13 age groups. The sample had the following composition: 50.3% female
and 49.7% male; 77.4% White, 13.5% African American, 6.1% Hispanic, and
3% of other racial/ethnic origin. The education-level composition of the
WAIS-III sample was 11.9%, <8 years; 13.2%, 9-11 years; 37.2%, 12 years;
22.2%, 13-15 years; and 15.5%, 216 years. The WMS-III test-retest sample of
297 participants consisted of approximately 10-30 individuals in each of the
13 age groups. The WMS-III test-retest sample had the following demo-
graphic composition: 48.1% female and 51.9% male; 78.8% White, 13.5%
African American, 6.1% Hispanic, and 1.6% of other racial/ethnic origin. The
education-level composition of the WMS-III sample was 12.1%, <8 years;
13.8%, 9-11 years; 37.7%, 12 years; 20.2%, 13-15 years; and 16.2%, >16 years.
For the WAIS-III, the test-retest stability coefficients were calculated for four
pooled age groups: 16-29, 30-54, 55-74, and 75-89 years. The WMS-III stabil-
ity coefficients were calculated for two pooled age groups: 16-54 and 55-89
years. The WAIS-III mean scores, standard deviations, and uncorrected and
corrected coefficients for each of the four age groups for both testing occa-
sions are presented in Tables 3.6—3.9. The average test-retest stability coeffi-
cients across all age groups are included in Table 3.9. The corresponding
WMS-III reliabilities and descriptive statistics for the Primary subtest scores
and Primary Indexes are presented in Tables 3.10 and 3.11, respectively.
The average stability coefficients across all age groups are included in Table
3.11. The retest stability coefficients were corrected for the variability of the
standardization sample (Allen & Yen, 1979; Magnusson, 1967).
As the data in Tables 3.6-3.9 indicate, the WAIS-III scores possess adequate
stability across time and for all age groups. The average stability coefficients
for Vocabulary and Information are excellent (in the .90s); the stability
coefficients of Similarities, Arithmetic, Digit Span, Comprehension, Digit
Symbol—Coding, and Block Design are very good (in the .80s); those of the
other subtests are fairly good (in the .70s). As the data also indicate, the
mean retest scores are higher than the scores from the first testing. These
differences, mainly due to practice effects, are about 2.0-3.2 points for
the VIQ score, 3.7-8.3 points for the PIQ score, and 3.2-5.7 points for the
FSIQ score.

As the data in Table 3.11 show, the average WMS-III test-retest stability coef-
ficients range from .62 to .82 (median = .71) for the Primary subtest scores
and from .70 to .88 (median = .82) for the Primary Indexes. Although three of
the Primary Indexes have adequate reliabilities in the .70s, the majority of

a7
Reliability and Score Differences

Table 3.6. Stability Coefficients of the WAIS-III Subtests, IQ Scales,

and Indexes: Age Group 16-29

First Testing Second Testing Corrected

Subtest/Scale/Index Mean SD Mean SD I42 r?
V 10.6 ZO 10.8 AS 85 .89
S 10.3 3.0 10.9 3.0 13 74
A 10.2 2.6 10.8 25 .80 .86
DS 10.2 PANS) 10.7 eae, BThe) 83
| 10.1 Qf 10.7 2.9 Pp .94
Cc 10.5 2.4 10.9 2.3 .67 18
LN 10.6 2.3 10.7 es, 48 70
PC 10.6 2.9 VAs PATE .66 .67
CD 10.2 3.0 11.4 Zo .80 81
BD 10.3 2.6 11.3 2.7 silt 83
MR 10.2 at 10.3 2.6 .70 48
PA 10.3 PA tes 3.0 .60 .67
SS 10.1 ah ialeu 2.8 .69 74
OA 10.2 220 12'5 3.6 64 74
VIQ 101.4 Wile 104.6 12.6 91 94
PIQ 101.6 12.2 109.8 i2a7 83 88
FSIQ 101.7 lee 107.4 12.4 91 42)5)
VCl 101.7 ee 104.2 13.3 .89 2
POI 101.6 13.3 108.9 13.3 ag) .83
WMI 101.9 12.5 104.8 14.5 82 .87
PSI 100.7 14.6 106.7 14.6 83 84
Se
Note. N = 100. For the Lette--Number Sequencing subtest and the Working Memory Index, N = 70.
* Correlations were corrected for the variability of the standardization sample (Allen & Yen, 1979;
Magnusson, 1967).

38
 Reliability

Table 3.7. Stability Coefficients of the WAIS-ILI Subtests, IQ Scales,

and Indexes: Age Group 30-54
a SS lS
First Testing Second Testing Corrected
Subtest/Scale/Index Mean SD Mean SD "2 i
Vv 9.9 3.0 10.0 3.0 .93 94
S 10.1 ARS 10.4 3.0 85 .88
A 10.0 3.2 10.3 3.2 .87 .87
DS 9.6 BaF, 10.0 2.9 79 83
| 10.0 2.8 10.6 2.8 .93 94
C 10.1 3.0 10.2 29 .80 81
LN 9.9 2.9 10.6 3.0 74 18
PC 10.3 3.0 12.7 3.2 TAS) oS
CD 9.6 2.9 10.7 3.2 84 84
BD 10.2 Pals 10.9 3.0 86 88
MR 10.1 2.6 10.4 3.0 .69 15
PA 10.0 2.8 11.2 3.2 70 73
Ss 10.2 20 10.7 Pail .80 82
OA 10.0 3.0 11.6 3.6 78 18
VIQ 99.3 14.4 101.3 14.9 295 96
PIQ 99.9 13.8 108.2 16.6 .88 .90
FSIQ 99.6 14.3 104.7 15.7 .96 .96
VCl 99.8 14.1 101.9 15.1 95 95
POI 100.7 14.3 108.1 16.2 .86 .88
WMI 99.4 15.2 102.5 15.6 .90 .90
PSI 99.2 14.1 103.8 14.6 87 88

Note. N = 102. For the Letter-Number Sequencing subtest and Working Memory Index, N = 64.
* Correlations were corrected for the variability of the standardization sample (Allen & Yen, 1979;
Magnusson, 1967).
Reliability and Score Differences

Table 3.8. Stability Coefficients of the WAIS-III Subtests, IQ Scales,

and Indexes: Age Group 55-74

First Testing Second Testing Corrected

Subtest/Scale/Index Mean SD Mean SD feo r
V 9.8 2.7 10.0 2.8 2 .93
S 10.1 2.9 10.5 2.6 .84 85
A 9.6 3.0 9.9 3:4 .86 .88
DS 9.7 2.6 10.1 2.8 85 .89
| 10.2 3.0 10.7 2.9 .93 .93
C 10.1 29 10:2 2.6 83 .85
LN 10.1 2S 10.4 2.9 silt .80
PC 10.0 2.8 11.6 2.9 82 85
CD 10.2 3.1 11.0 3.0 85 .86
BD 9.9 2.8 10.1 2.8 rdae .80
MR O25 2.8 Sh 249 18 81
PA 10.0 29 tee 3.4 .62 .67
SS 10.2 29 10.7 3.0 as TE
ye OA 10.0 2.6 11.0 3.0 76 82
VIQ 99.0 14.1 101.1 14.2 97 yi
PIQ 99.1 14.2 104.8 oun nO .92
FSIQ 99.0 14.3 102.9 15.0 .96 Sif
VCl 99:9 14.6 101.8 14.0 .96 96
POI 98.3 13.5 102.3 14.8 188) 92
WMI 99H 12.8 101.3 13.7 .90 93
Psl 100.9 14.8 104.7 15.3 89 90

Note. N = 104. For the Letter-Number Sequencing subtest and the Working Memory Index, N= 81.
* Correlations were corrected for the variability of the standardization sample (Allen & Yen, 1979;
Magnusson, 1967).
 Reliability

Table 3.9.
Stability Coefficients of the WAIS-III Subtests, IQ Scales,
and Indexes: Age Group 75-89
LS St
First Testin Second Testing Corrected Santen
Subtest/Scale/Index Mean SD Mean SD Ny 7" Groups
Vv 9.8 ee 10.2 2.8 85 .88 91
SS) 9.6 2.9 10.3 3.0 82 83 .83
A 10.1 3.0 10.6 3.1 .83 84 .86
DS 9.7 2.6 9.6 2.7 .69 73 .83
| 10.1 are 10.7 3.1 .94 .94 .94
C 9.9 2.6 10.2 3.0 75 19 81
LN 10.0 3.1 10.5 3.4 Tf) 11 15
PC 10.1 3.1 11.0 3.4 82 82 79
CD 9.6 2.8 10.2 3.3 91 91 .86
BD 9.7 2.9 10.0 3.2 76 77 82
MR 10.2 2.6 10.1 3.1 72 76 ll
PA 10.1 3.2 10.8 3.4 7A 68 .69
Ss 10.1 3.1 9:9 3.5 80 .80 19
OA 9:9 2.9 10.8 3.2 .65 68 76
VIQ 98.9 13.0 101.3 14.7 94 95 .96
PIQ 99.4 15.2 103.1 18.7 .93 93 91
FSIQ 99.0 14.1 102.2 16.3 .96 .96 .96
VCl 98.9 13.4 102.1 14.8 93 :95 95
POI 99.6 14.1 1023 17.6 .89 .90 .88
WMI 100.2 14.2 107;5015:9 85 85 .89
PSI 99.0 14.9 100.3 17.4 92 ‘92 .89

Note. N = 88. For the Letter-Number Sequencing subtest and Working Memory Index, N = 67.
@ Correlations were corrected for the variability of the standardization sample (Allen & Yen, 1979;
Magnusson, 1967).

61
Reliability and Score Differences

Table 3.10. Stability Coefficients of the WMS-III Primary Subtest

Scores and Primary Indexes: Age Group 16-54
ee Eee
Primary
SlibtectSeores! First Testing Second Testing Corrected
Primary Indexes Mean SD Mean SD I aa
Logical Memory |
Recall Total Score 10.2 B12 12.4 3.1 Teil 74
Faces |
Recognition Total Score 10.5 2.8 13.2 3.3 64 70
Verbal Paired Associates |
Recall Total Score 10.4 2.7 11.8 3.0 ef. 81
Family Pictures |
Recall Total Score 10.5 2) 12.4 2.9 61 .63
Letter-Number Sequencing
Total Score 10.3 2.6 10.7 2.9 61 od
Spatial Span
Total Score 10.2 2.7 10.4 2.6 65 72
Logical Memory II
Recall Total Score 10.2 3.1 12.5 3.0 tt 76
Faces Il
Recognition Total Score 10.6 2.8 12.8 2A 58 63
Verbal Paired Associates Il
Recall Total Score 10.5 2.8 11.1 Xs ao dah,
Family Pictures II
Recall Total Score 10.4 3.0 12.6 2.9 .67 .68
Auditory Recog Delayed
Total Score 10.1 2.9 11.4 3.0 .60 62
Auditory Immediate 101.7 14.4 111.5 16.3 84 85
Visual Immediate 102.8 13.7 117.1 16.2 74 aT
Immediate Memory 102.8 14.5 117.6 17.9 84 85
Auditory Delayed 101.9 14.1 110.6 13.9 82 83
Visual Delayed 103.0 14.7 117.4 15ez .74 75
Auditory Recog Delayed 100.6 14.4 106.9 15.0 .60 .62
General Memory 102.5 14.6 115.4 15.9 87 87
Working Memory 100.8 12.4 103.1 14.3 70 19
re
Note. N= 141.

* Correlations were corrected for the variability of the standardization sample (Allen & Yen,
1979; Magnusson, 1967).

62
 Reliability

Table 3.11. Stability Coefficients of the WMS-III Primary Subtest

Scores and Primary Indexes: Age Group 55-89

Primar y . - pvaiege
Subtest Scores/ _FirstTesting Second Testing Corrected Shenk
Primary Indexes Mean SD Mean SD Ne ea r
Logical Memory |
Recall Total Score 9.6 2.8 11.3 3.1 ora .80 a,
Faces |
Recognition Total Score 10.1 3.1 12.1 3.8 64 .63 67
Verbal Paired Associates |
Recall Total Score 9.6 3.1 11.1 3.6 84 .83 82
Family Pictures |
Recall Total Score 9.7 3.1 11.1 3.1 .70 .68 .66
Letter-Number Sequencing
Total Score 10.0 3.0 10.3 3.2 75 hth 14
Spatial Span
Total Score 10.0 3.0 10.0 3.1 .69 pyA0) fil
Logical Memory II
Recall Total Score 9.9 2.9 12.1 3.0 74 76 76
Faces Il
Recognition Total Score 10.4 3.0 122 3.8 .63 61 62
Verbal Paired Associates II
Recall Total Score 9.6 3.0 10.8 3.3 79 79 78
Family Pictures Il
Recall Total Score 10.0 3a 11.3 3.1 13 73 Ais
Auditory Recog Delayed
Total Score 9.6 2.7 10.8 Si2 16s 76 70
Auditory Immediate 97.3 14.8 107.2 17.9 85 85 85
Visual Immediate 99.3 16.3 109.6 18.8 76 73 AB
Immediate Memory 98.0 16.9 110.2 19.8 84 82 84
Auditory Delayed 98.5 14.6 108.7 IPA 84 85 84

Visual Delayed 101.1 15.4 111.3 18.9 76 76 76

Auditory Recog Delayed 98.2 13.7 104.0 16.2 VA 76 .70
General Memory 99.1 15.7 110.6 18.7 88 88 88
100.0 14.8 101.0 15.7 .80 80 .80
Working Memory
ne EE

Note. N = 156.

@ Correlations were corrected for the variability of the standardization sample (Allen & Yen, 1979;
Magnusson, 1967).

63
Reliability and Score Differences

the indexes have stability coefficients in the .80s. With the exception of the
working memory subtests and the Working Memory Index, the WMS-III
mean subtest scaled scores and mean Index scores increased by roughly 0.33
SD to 1 SD from the first to second testings. Smaller gains in retest perfor-
mance would be expected with test-retest intervals of relatively longer dura-
tion. In general, the older pooled age group (i.e., 55-89 years) show smaller
retest gains than do the 16-54 age group.
Stability coefficients for the WMS-III supplemental subtest scores were
evaluated by two methods. First, many of the supplemental scores were
evaluated in a manner similar to that used for the Primary subtest scores,
that is, by test-retest correlation coefficients. A second method of evaluating
score stability was used because many of the WMS-III supplemental subtest
scores have relatively small raw-score ranges, and, therefore, some of the
score distributions are highly skewed (e.g., Information and Orientation
Total Score). Because the test-retest correlation coefficients for many of
these scores may be artificially low due to restriction of range, a decision-
consistency methodology was used. With this approach, the subtest scaled
scores are divided into ranges, and the consistency of the decision range,
or classification, from test to retest is assessed. The decision-consistency
reliability indicates the concordance of the decisions in terms of percent
correct classification.
Cutpoints were established for WMS-III subtest scaled scores based on
standard deviation units from their respective means. The scaled scores
were divided into four groups: <3, 4-5, 6-13, and 214. Decision-consistency
reliability was then assessed by comparing the classifications of the first
testing to the classifications of the second testing.
Table 3.12 presents the stability coefficients and descriptive statistics for the
WMS-III supplemental subtest scores for two age groups: 16-54 (n = 141)
and 55-89 (n = 156). Those stability coefficients reported as percentages
represent decision-consistency coefficients, and the stability coefficients
reported as decimal fractions are test-retest correlation coefficients. The
decision-consistency coefficients range from 50% to 100%, and the average
test-retest stability coefficients range from .56 to .80. In general, those sub-
test scores that are derived from differences between two scores (e.g., Word
Lists I Contrast 1 and 2) or the ratio of two scores (i.e., percent retention
scores) demonstrate the lowest reliabilities.

64
 Reliability

Interscorer Agreement
Because the scoring criteria for most of the WAIS-III and WMS-III subtests
are simple and objective, interscorer agreement is very high, averaging in
the high .90s. However, some of the subtests of both scales require more
judgment in scoring. Therefore, for both the WAIS-III and the WMS-III,
special studies were conducted to evaluate subtest score agreement
between scorers.
The subtests targeted for the WAIS-III study were three of the Verbal sub-
tests—Vocabulary, Similarities, and Comprehension. A subsample divided
into two groups of 60 was randomly selected from the standardization sam-
ple. This subsample had the following composition: 53% female and 47%
male; 81.5% White, 12.6% African American, 4.2% Hispanic, and 1.7% of
other racial/ethnic origin. For each group of this subsample, three raters
independently scored each of the 60 protocols. The interrater reliability
coefficients for the three Verbal subtests were very high: .95 (Vocabulary),
.93 (Similarities), and .91 (Comprehension).
For the WMS-III study, 10 protocols from each of the age groups were ran-
domly selected from the standardization sample and were independently
scored twice. The interscorer reliability coefficients for Logical Memory I
and II, Family Pictures I and II, and Visual Reproduction I and II (i.e., the
WMS-III subtests requiring the most scoring judgment) were all greater
than .90.
For both studies, the interrater reliability coefficients were calculated
according to the appropriate intraclass correlation procedures (Shrout &
Fleiss, 1979) to account for scorer leniency. These results show that although
these subtests require more scoring judgment, they can be scored very
reliably.
 ee EE Tee oan ote en einen nr nt TT de EEEEEEEEEEI Ie
%C9 6'¢ ora’ Le EOP V9 9'¢ BLL Lo €or UONUS}OY JUS
89° Oe LE 6? 66 69° VC 6 LL 6¢ O'OL Q109S |e}O] Ijewsy|
|] Asowayy je01607
08 Ge 0°01 8c 26 LL Go 0'OL 9°¢ 86 8109S |e}OL
JO1]U05 je}Ud|
c9 ee c Ol 6¢ 66 6S" 9¢ 9°01 O€ co OL 9109S |e}0] premyoeg
Ach Oe 66 6c €Ol 09° 6'¢ (S(O) 6'¢ LOL 9JOIS [e}O] PAeMIOF
|eneds
ueds
Ez Cis GLk Le LOk 09° 9¢ care 8c VOL 8109S |e}O0] |J2B99Y
| UoNONpoiday jensi~p
%OL 6c 6 OL Lag, 66 %BY 6'¢ 8OL Oe LOL c $se1JU0D
%IG ve 98 Oe bee SVG Ge 08 Cas LG } JSesJUOD
%LB 82 96 O'e LOL %LL 92 Z8 62 16 edojs Hulweey
el ce O'LL OG= —66 19° ee O'LL oe €'0l 9109 |B}0] |/209Y
%QL Ze pth 6a =66 %09 ee SLL 82 e'0l 8100S [BJO] |/200Y 3S}
| S}SI"] PAOAA
BL (EG 6 ON. 66 VOL Go v8 iG c OL ado|s Hulwee]
VL JES; SLL 8c Z6 ROL cae Gol Oe c Ol B109S Je}OL |/BOSY IS}
|
jeqia,
paued
Sayeis0ssy
Reliability and Score Differences

vo eS G kk (0)18; 86 19° Ore O?cl Ee. O'OL 8109S |e}OL Ijewsy]

SbLL Oe 96 OS € Ol Sol 9¢ (316 Oe LOL adojs Bulwee7
REL ce GL 60 G6 %II ate: Vol es, LOL 8109S |e}OL |[B99y IS}
1201607
|Asowsyy
9'0
ZAG
Z0
BEL
G0
%OOl
|e}OL
MAS
%00
qe400g
JETS
i210)
|b
UOIJEWOJU|
9
UOI}EJUSIIO

Augers
Hulsey
Bulsay
Hulsey
PCTS as ueey as uean ,jwueIyjeep 6 g@S Ss as urea

puoses
ysaiqns

ysul4
ysui4
UIIGQeIS

dnoiy
dnoly
68—Gs
vS-9L
aby
$3109¢ Ss9IqQNs [e}UsUIa|ddng []]-SIWM 9) JO SyUsTIIyJa0D AWTIGeIS "ZL"E ajqey
 BIGeL “ZL"E ANTIGeIS syuaTayJa0
JO eYp [TI-SIWM [eJUaUIaTddn¢g
1Sa}qn¢ Sa109g
(panu1juo9)
aby dnoiy vS-91 aby dnoly 68-SS
ysui4 Bursa puosas Buljsal Aunders ysuig Burlysay
= puoseg Busey
Augers
ysayqns
ueey as as JUBIDIB0D
UBS, as ueey gS= 2lUal91ye09

ues

|]
JUBDEg UONUd}OY Le OlVv Le %68

seoey
cOl

Oe
Sc
Ajwiey Saanyoig
I]

re}
JUsDEq UONUB}BY 0€ Ol O€ %98

6?
Sc

SO
jequa, pasied

001
Sayeis0ssy
|]
JUBDEd UONUd}BY Oe O'll 0'€ %bl

8c
ar
PAOM SYS]I]

Ol
60l
|;eoeyJe}O] 8109S Oe 601 ce L9°

Le
uoIUuBOoeYy
|e}O] 8109S be 00} 0'€ a7
JUBDEd UOHUB}OY 62 Ol

66 €Ol
ke %v8B

Le Le ce

26
ve ee
jensiA UoNDNpoday
|]

ol Z01alk
001001
|feoe8y|P}O] 8109S eb Le Sth ce LL

ee
9c
8° O'?k
LOL
UuoNUBoo8Yy
|e}0) 9090S kok 8c el

oe
Ado |e}0) 9100S Oe LOL 6c 6S"

cOl cOl
66

8° 8¢
6c

VOLLOLLOL
uoeuIWOSIG
|B}OL q2409SZS 89 si9 80 o9 8°0 %001

S°0
JUsEg UOIUS]aY Olc 8k SOL Oe LY Le v9

9°0 82
ve
‘aION=N [yl 10) aby dnoly ‘pS—-9}
= 9g) 10} aby dnoid ‘68-Sg
p s}UaloijJe05
payodeal
se sabejusoied
aie Aous}sisuod-uoisioap
‘seij!|igel|e1
esou) payodes
se JeWIOep SUONOBJ
a1e }SE}e1-]S9} “SUONE|S1109
g SOISIEIS
Jk Paseq
UO MBJ "S8J0NS

67
Reliability
Reliability and Score Differences

Score Differences
Differences Between IQ Scores
and Index Scores
An important consideration in interpreting the performance of individual
examinees is the amount of difference between the IQ scores and between
the Index scores on the WAIS-III and between the Index scores on the
WMS-III. The issue of score differences has two quite different aspects—the
statistical significance of the difference and the base rate, or frequency, of
the difference in the population. These aspects can best be described as two
questions: Is the difference real and not due to measurement error? Is the
difference clinically meaningful?

Statistical Significance of IQ
and Index Score Differences
A statistically significant difference between scores, for example, between
the VIQ and the PIQ scores, refers to the likelihood that obtaining such a dif-
ference by chance is very low (e.g., p < .05) if the “true” difference between
the scores is zero (Matarazzo & Herman, 1985). The level of significance
reflects the level of confidence the examiner can have that the difference
between the scores, called the difference score, is a true difference.
The difference between scores required for significance is computed from
the standard error of measurement of the difference (SEy,,,). This statistic
provides an estimate of the standard deviation of the sampling distribution
of the difference between the two obtained IQ or Index scores. Multiplying
the standard error of measurement of the difference by an appropriate z
value yields the amount of difference required for statistical significance at
any given level of confidence.
The differences between WAIS-III VIQ and PIQ scores and between any pair
of Index scores required for statistical significance are presented in Appendix
B of the WAIS-III Administration and Scoring Manual. The critical values
are provided for the .15 and .05 levels of significance for the 13 age groups.
Appendix F of the WMS-III Administration and Scoring Manual presents the
analogous tables for differences between WMS-III Primary Index scores and
between Primary and supplemental subtest scores. Although these differ-
ences vary slightly from age group to age group, average values for all of the
age groups, which are given in the last row of each table, are generally suffi-
cient for the purpose of assessing differences in an examinee’s abilities.
 Score Differences

Frequency of IQ and Index Score Differences

The base rate of the difference between two scores refers to the prevalence
or frequency of such an observed difference in the general population. Often
the difference between an individual’s VIQ and PIQ scores is significant in
the statistical sense but is not at all rare among individuals in the general
population. The statistical significance between scores and the rarity of the
difference are two different issues and consequently have different implica-
tions for test interpretation. (For a detailed discussion of the distinction
between statistically and clinically meaningful differences between scores,
see Payne & Jones, 1957, and Silverstein, 1981.)
The frequencies of various differences between the WAIS-III VIQ and PIQ
scores and between Index scores that occurred in the WAIS-III standardiza-
tion sample are presented in Table B.2 of the WAIS-III Administration and
Scoring Manual.' The table provides the percentage of examinees whose
scores differed by the given amount or more, regardless of the direction of
the difference. For example, about 17.6% of the examinees obtained VIQ
and PIQ scores that differed by 15 or more points; 23.1% of the examinees
obtained VCI and POI scores that differed by 15 or more points, whereas only
about 8.8% of the standardization sample obtained VCI and POI scores that
differed by 22 or more points. Table F2 in Appendix F of the WMS-IIT
Administration and Scoring Manual provides analogous information (i.e.,
frequencies of score differences of various magnitudes in the standardization
sample) for the Primary Index scores.

Differences Between a Single Subtest Score

and an Average of Subtest Scores
Very often the performance of an individual on the subtests of a Wechsler
intelligence scale varies across different subtests. An evaluation of the vari-
ability helps the examiner identify the strengths and weaknesses of the
examinee’s cognitive functioning. As with differences between the IQ and
Index scores, the interpretation of a particular subtest score as especially
high or low should take into account the statistical significance of the
observed difference and estimates of population base rates.

score discrep-
' Analysis of variance indicated no significant difference in the average VIQ—PIQ
age groups.
ancies or in the average discrepancies between various pairs of Index scores across
es vary as a func-
Matarazzo and Herman (1985) have demonstrated that VIQ-PIQ discrepanci
supplemental
tion of ability level. Similar findings were obtained for the WAIS-III; therefore,
D of this Manual. The user is cautioned that when an
tables have been included in Appendix
has sustained neuropsych ological damage and current ability functioning may be
examinee
& Johnson, 1978; Parsons
compromised, such differences may be misleading (Heaton, Baade,
& Prigatano, 1978).
Reliability and Score Differences

A common procedure for evaluating an individual's cognitive strengths or

weaknesses involves the comparison of an individual's single subtest scaled
score to his or her average scaled score on a group of subtests or all subtests.
For instance, an examinee’s scaled score on the Information subtest may be
compared to his or her average scaled score on the 6 or 7 subtests of the
Verbal scale or to an overall average of 11, 12, or 13 subtests. A single score
that is significantly greater than the individual's own mean score may reflect
a relative strength, whereas one that is significantly less than the mean may
indicate a relative weakness. The procedure for testing such differences for
statistical significance was originally suggested by Davis (1959). Silverstein
(1982) refined the procedure to account for the fact that several comparisons
are being made simultaneously.
The interpretation of the difference between a single subtest score and the
individual’s own mean score is an intraindividual comparison. Strengths and
weaknesses identified in this way are strengths and weaknesses relative to
the individual’s own general ability level. Therefore, a subtest scaled score
that is high in absolute value, that is, well above 10, may still represent a
relative weakness for an individual of extremely high general ability; on
the other hand, a fairly low scaled score may indicate a relative strength for
an individual of generally limited ability.

Significance of Differences Between a Single Subtest

Score and an Average of Subtest Scores
For the WAIS-III, the minimum differences between an individual’s subtest
scaled score and the average scores of various groups of subtests are pro-
vided in Appendix B of the WAIS-III Administration and Scoring Manual.
Table B.3 provides this information for comparing a single subtest scaled
score to the average of various sums of scaled scores. These minimum differ-
ences are the differences required for statistical significance at the .15 and
.05 levels of significance. For example, as indicated in the table, a Vocabulary
scaled score that is at least 1.70 points greater or less than an individual’s
average of scaled scores on the 6 Verbal subtests is significantly different
from that mean score at the .15 level, whereas a difference greater than 1.99
points is significant at the .05 level.

Frequency of Differences Between a Single Subtest

Score and an Average of Subtest Scores
As with the differences between the composite scores (e.g., between IQ
scores or between Index scores), a difference between a single subtest scaled

70
 Score Differences

score and the average of a group of subtest scores may be statistically signifi-
cant but not especially unusual in the population. Thus, the fact that an
individual's scaled score on a single subtest, for example, Information, is
significantly less than the individual’s average scaled score of the Verbal
subtests does not necessarily indicate that the difference is clinically mean-
ingful. For this purpose, the estimated base rate of the difference, which is
the frequency of such a difference occurring in the standardization sample,
can be very useful, because it indicates whether the difference is rare or
common in the general population.
Table B.3 in Appendix B of the WAIS-IIT Administration and Scoring Manual
provides data on the frequencies of differences for the entire WAIS-III stan-
dardization sample, which are the estimated base rates of the general popu-
lation. According to this table, for example, a difference of 3 or more points
between the Information subtest scaled score and the average of the 6 Verbal
subtests occurred in only 5% of the entire standardization sample.

Differences Between Subtest Scores

Statistical Significance of Differences
Between Subtest Scaled Scores
Very often the difference between scaled scores on a particular pair of sub-
tests may be of interest for a variety of reasons. For example, an individual's
score on Similarities may be 4 scaled-score points greater than the score on
Vocabulary, and the examiner should know whether such a difference is sta-
tistically significant before interpreting it. Also, because most of the WMS-III
indexes are made up of two subtest scores, the examiner may want to evalu-
ate the significance of the difference between the scaled scores on the two
subtests that compose an index.
As mentioned previously, an individual may not perform at similar levels
across different subtests. Differences of any given amount between two sub-
test scaled scores may occur either because of chance fluctuation or because
of a true difference in abilities. As with the difference between IQ or Index
scores, the standard error of measurement of the difference between two
subtest scores is used to determine the minimum difference required for sta-
tistical significance. Table B.4 in the WAIS—III Administration and Scoring
age
Manual presents the average of these minimum differences across all
pair of
groups at the .15 and .05 levels of significance for every possible
of 2.68 points or
WAIS-III subtests. For example, a scaled-score difference
at the .05 level. In
more between Vocabulary and Similarities is significant
difference is signif-
other words, the examiner can be 95% confident that this
tion and
icant. Similar information is provided in the WMS-III Administra
between
Scoring Manual. Tables E3 and F4 show the minimum differences
71
Reliability and Score Differences

subtest scaled scores at the .15 and .05 levels of significance across all age
groups for the Primary subtests and the supplemental subtests, respectively.

Intersubtest Scatter
Intersubtest scatter is the variability of an individual’s scaled scores across
the subtests (Matarazzo, Daniel, Prifitera, & Herman, 1988). Such variability
has frequently been considered diagnostically significant. Although various
measures of scatter are possible, the scatter index used in this Manual is the
easiest to obtain: the simple difference between the individual's highest and
lowest subtest scaled scores.
Before interpreting the scatter exhibited in a particular test record, however,
the examiner should consider how common such a scatter is in the popula-
tion. The cumulative percentages of intersubtest scatter within various
WAIS-III scales (e.g., Verbal, Performance, Full Scale, and indexes) are
reported in Table B.5 of the WAIS-III Administration and Scoring Manual.
The percentages are based on the data from the entire WAIS-III standardiza-
tion sample so that examiners can estimate whether a particular scatter is
common or rare in the general population. For example, only 3.7% of the
entire standardization sample obtained a scatter of 9 scaled-score points
or more within the 6 Verbal subtests. On the other hand, 71.2% obtained a
scatter of 6 or more scaled-score points on the 11 subtests contributing to
the FSIQ score.

Discrepancies Between Digit Span Forward

and Backward
The WAIS-HI and WMS-III Digit Span scaled score is based on the combined
raw scores for Digits Forward and Digits Backward. Digits Forward and
Digits Back-ward, however, may involve different cognitive processes, which
may be differentially impaired in certain clinical groups (E. Kaplan et al.,
1991; Lezak, 1995). For example, a study by Rudel and Denckla (1974) found
that children with developmental disorders involving putative right hemi-
sphere deficits had impaired performance on Digits Backward relative to
their Digits Forward performance.
In order to facilitate comparisons of Digits Forward and Digits Backward
scores, Table B.6 in the WAIS-III Administration and Scoring Manual pro-
vides the cumulative percentages for the longest forward and backward digit
spans for each age group and the entire WAIS-III standardization sample.
The means, standard deviations, and medians are also provided. Table B.7
provides the cumulative percentages of the differences between the longest
Digits Forward and Digits Backward spans for each age group and the entire

72
 Summary

standardization sample, along with the means, standard deviations, and

medians of these differences. Tables E5 and E6 in the WMS-III Administra-
tion and Scoring Manual provide identical information.

Summary
This chapter has presented information not only on the reliability of scores
but also on differences between a variety of derived scores, including statis-
tical significance and frequency. The accompanying tables provide data nec-
essary for the interpretation of relative strengths and weaknesses. Seemingly
large discrepancies between two scores, for example, may not necessarily be
statistically significant, and some statistically significant differences may not
be rare in the related population. The examiner, however, should keep in
mind that measurement error contributes to part of observed-score differ-
ences and that the errors in the differences between composite scores (e.g.,
between VIQ and PIQ scores) will be smaller than the measurement errors in
the differences between subtest scaled scores. This pattern occurs because
the IQ scales and indexes are more reliable than the subtests. As always,
when interpreting scores, the examiner should integrate relevant informa-
tion from a variety of sources, including the individual’s life history, educa-
tional background, and other test scores, in addition to the information on
the statistical properties of the WAIS-III and WMS-III test scores presented
in this chapter.

73
 * ered

r ¢ wticl
' ‘ (= 4 hems Vets aie

: pal | Pn ad .
ty ry Mad 1) ld irre? i! eel) Doheny
wis iy eset wa lerwwn
' fit) yet Sy8] rk / eS! « F08 NAS,

: pial, , yer ah. eg ath yates See irs Saad

ib lela

oi ‘ CL WAOTTY! Qi A err Te S47 yh “ae 92 iaa ema part

TT Fe a) elay nl S47 a ey. er pe tieet aN
poem Ari. i vA Samiti > eee Sadr ath meron b aT
"a lebsatten govem (UAW be TOM ar) 40 elie
ie Gag ine po .
cag pe
‘welt alt
* am
Oo0rus t he Seieeay ‘al Baw wal | ee
ws decker ~~ %
Ese Wn®-
Slyes e090 A ting
BAP 49 54 My Vatsed ine nr _

my a
As
7 matt eee. a oe
=e | aw 7 *
ee
wr ow
 CHAPTER 4

Evidence Base for Validity

of the WAIS-III and WMS-III

Multiple research studies were conducted as part of the WAIS-III and

WMS-III standardization. These studies provide evidence for the validity and
applicability of the WAIS-III and WMS-III and yield important information
for clinicians regarding their use. In the 5 years since the original publication
of the two instruments, subsequent research studies have yielded data on
the WAIS-III and WMS-III in the context of the diagnosis, treatment, and
exploration of the pathological processes of specific neurological, psychi-
atric, and developmental disorders. This chapter integrates the findings
from these studies with those of the previous studies published as part of
the original Manual. The discussions are organized around specific research
hypotheses and clinical applications. The data are provided as ongoing
evidence of validity for the WAIS-III and WMS-II] and with the expectation
that continuing research will expand and refine our knowledge of these
assessment instruments.

Evidence Based on Internal Structure

In their classic article, Campbell and Fiske (1959) presented a theoretical
methodology for interpreting the patterns of correlations in a multitrait—
multimethod matrix to provide evidence of convergent validity and discrimi-
nant validity. Their original methodology was based on the relationships
between similar constructs measured by different methods as well as appar-
ently dissimilar constructs and the examination of correlational patterns in
a matrix where some variables are predicted to be correlated (convergent
validity) and other variables are predicted to have less relationship (discrimi-
nant validity). The relationships between correlation coefficients were also
interpreted relative to the reliability of the various scales. Data supporting a
priori hypotheses about the pattern of the relationships provide evidence
of construct validity. Evidence of convergent validity is provided when two
variables that were expected to be correlated are, indeed, related. Low corre-
lations between two variables that were not expected to be related provide
evidence of discriminant validity.

75
Evidence Base for Validity of the WAIS-III and the WMS-III

Because evidence of the validity of a test can be found in the correlations of

its subtests with each other and with groups of subtests, the relationships
among the various subtests, scales, and indexes of the WAIS-III and of the
WMS-III were examined.

WAIS-III Intercorrelations
For the WAIS-III intercorrelation studies, several assumptions were made.
First, it was assumed that there is a general trait of intelligence, or a g factor,
and that because of the g factor, all of the WAIS-III subtests, even those mea-
suring apparently different abilities, would have some degree of relation to
one another. The importance of the g factor has been a much-debated topic.
However, from Spearman's (1904, 1932/1970) original work to Carroll’s (1993)
more recent work, results of study after study have indicated a relation
between all ability measures (Brody, 1992; Gustafsson, 1984; Neisser et al.,
1996; Sternberg, 1980). For example, in an extensive review study, Carroll
(1993) investigated the factor structure in over 450 data sets and found that
a general intelligence factor was present throughout those data sets. For the
WAIS-III, it was assumed that all subtests would have at least low to moder-
ate correlations with each other.
Second, the subtests contributing to the VIQ scale were expected to have
higher correlations with each other than with subtests composing the PIQ
scale. For example, the correlations between Vocabulary and Similarities
would be much higher than correlations between either of these two sub-
tests and any of the Performance subtests (e.g., Picture Completion, Digit
Symbol—Coding, and Block Design). Similarly, correlations between two
Performance subtests were expected to be higher than correlations between
a Performance subtest and a subtest of another scale.
Third, according to previous research (e.g., J. Cohen, 1952a, 1952b, 1957a,
1957b; Wechsler, 1991), intellectual functioning is broken down into more
discrete domains of ability: verbal comprehension, perceptual organization,
working memory (freedom from distractibility), and processing speed. The
subtests contributing to a domain were predicted to have higher correlations
with other subtests within the same domain than with subtests measuring
other abilities.
Fourth, evidence from previous studies has indicated that some subtests are
more related than others to the general intelligence factor. For instance,
Picture Completion, Vocabulary, Similarities, Block Design, and Information
tend to be more related to the g factor than are the other WAIS-III subtests
(see Kaufman, 1975, 1990, 1994; Sattler, 1992). The following predictions
were made on the basis of this evidence. Those subtests that have been
shown to have high “g loadings,” even those subtests contributing to

76
 Evidence Based on Internal Structure

different IQ scales (i.e., Verbal and Performance) or different indexes, would

have relatively high correlations with each other. However, two subtests with
high g loadings and on the same IQ scale or index (e.g., Vocabulary and
Information) would have higher intercorrelations than two subtests with
high g loadings but on different scales or indexes (e.g., Vocabulary and
Block Design).
Finally, studies of previous Wechsler scales have indicated that two sub-
tests—Picture Arrangement and Arithmetic—have split loadings across
different factors. Picture Arrangement has significant loadings on both the
Verbal and Performance scales. Arithmetic has been shown to split between
the Verbal scale and the Working Memory (Freedom From Distractibility)
Index (Wechsler, 1991). Therefore, the WAIS-III correlation matrix was
predicted to show these splits, with these two subtests having apparently
similar correlations with other subtests from both of the respective factors.
The results of the WAIS-III intercorrelations averaged across all 13 age
groups are presented in Table 4.1. More detailed results of the WAIS-III
intercorrelation analyses are presented in Appendix A of this Manual, which
includes 13 intercorrelation matrices, one for each of the 13 age groups.
Each table includes the correlations of each subtest with each of the other
subtests and with the sums of scaled scores for each of the IQ scales and
indexes. The correlation of a scale with one of its contributing subtests
(e.g., Performance scale with Picture Completion) was corrected by the
removal of that subtest score from the sum of the scaled scores in order to
control for inflated correlations. These corrected coefficients appear in the
main body of each table, and the uncorrected coefficients appear in the
shaded area on the right. The intercorrelations of the IQ scales and indexes
appear in the lower right portion of each table. Some of these correlations
are inflated because the two correlated scales have subtests in common
(e.g., Verbal with Full Scale). The correlation matrix for the overall sample is
the average correlation across all 13 age groups computed with Fisher's z
transformation (see Table 4.1).
The high magnitude of the intercorrelations among the majority of the sub-
tests supports the premise that a general intelligence, or g, factor is present.
As the data in Table 4.1 show, most of the subtests tend to correlate with
each other at least at a moderate level. Statistically, all intersubtest correla-
tions are significant. This tendency of high intersubtest correlations sup-
ports the notion of a g factor, which pervades several different types of
abilities. The pattern of WAIS-III intercorrelations is very similar to that
found for the WAIS-R and other Wechsler intelligence tests, in which most
of the subtests have significant correlations with the other subtests.

7
 3]ge]
L'7
* SUONP[IIIOIIOJU
jo ITI-STVM
T 3S91qNsg PgfB9s S3I0IS
PUP SUINS
JO PsfBIs SBIOIS
IO} OI sofBISg
pue

78
:sexepu]
[IV sasy
Asaigns
[leg
oD Km E E meeuecee eye isan 91 =9= Nim Sd dd” Ga° YN] Vd SS VO IPeAT “Hed liMd 9A Od WM Sd
A 68" 8" c6
S gL gg ie 06
V 09" lg 08 BL eg"
sd S87; m0ys 2G: G9 19 es"
| es Tome Seo) 07 98’ oie
% esr 0m SiS. Sees OLF v8 og
NI mang Seve SoSh, 21> lV, S ty )
Od wey were BOrs Bice = “Or > OF > ly OL 4g Lg
a) wary, Sore eps Boge Ge Se 20 ~ +7 2 Go 8g’ 19 16
ag comms amemyc oem (Spe Gy" = or. Be 2G > le br Ble ge
YN mer Recaroacre eo Por 2h OPTS > OF - 09 er GL eg
Wd meccery ec Sarg 0S BE Ge Le «Gh OS Peer
Ss Sy Pe ra We ee ME co 16°
vO gy PT Pes ame PO en ee cee en a
ee) Oe ice 6 OL ES es re aS es dg" og Bac): cee G6 HZ eC bG2
‘ed consco pa Ee 00) Se SG" 09" = 08" 00's 90 09" 69" v8 eee, 12. hoi Soo Gi
ling cremaee reeceSi a 08 Ce OS ON EO 90 eg 205" fe Stas OFM
OA es fy fee Tr Ye Mer eeee) a rn a: a| 20" co eo
Od ec epee Spc ae eS GG | By SO
Evidence Base for Validity of the WAIS-III and the WMS-III

eo 6S 0g cg €9"= 09
NM a) ocaume mr agg Nizce eG) eG po” = Gy OF 6S" ty 8S" ve’ gg
Sd ecgecg lichen ey” Se Ge = Bh Se G0 ee Bh OY - 90 ve
ueeW OOF OOF OO OOF OL LOL OO OO OOF +OF OOF OOF 10F O'OL 209 10S EO 10E 10E e0& O02
rs. ecg cc: ccwe Tene 0s “OER Wel WE DE Of O€ Lyl Ell vee ¢8 v2 G2 SG
 Evidence Based on Internal Structure

Second, the Verbal subtests do have higher correlations with other Verbal
subtests than with Performance subtests. A similar, but less distinct, pattern
is observed between the Performance subtests; that is, most of the
Performance subtests correlate more highly with each other than with the
Verbal subtests. The pattern is not as distinct, however, because some of the
Performance subtests (generally those with higher g loadings, such as Block
Design and Matrix Reasoning) tend to have reasonably high correlations
with the Verbal subtests that also have high g loadings.
The magnitudes of the correlations within the Verbal scale and within the
Performance scale show some variability. The patterns indicate that the sub-
tests within a specific ability or domain (e.g., Working Memory or Processing
Speed) intercorrelate more highly with each other than with other measures
with higher g loadings. For instance, the subtests of the PSI tend to have
higher intercorrelations with each other than with subtests on other indexes:
Digit Symbol—Coding has a relatively high correlation with Symbol Search
but smaller correlations with other Performance and Verbal subtests.
Similarly, the intercorrelations between Arithmetic, Digit Span, and
Letter-Number Sequencing of the Working Memory Index range from .52 to
.57, which, except for the Arithmetic subtest, are much higher than the inter-
correlations with other Verbal and Performance subtests.
The exceptions to the patterns of high within-factor correlations detected
in previous Wechsler scales are also found in the WAIS-III. For instance,
the Performance subtest that tends to be related most to verbal skills is
Picture Arrangement, for which examinees can “talk out” the solutions to
the problems.
In general, the patterns of correlations hold true throughout the age groups.
More important, they provide evidence of convergent and discriminant
validity (Campbell & Fiske, 1959), and they parallel the typical relations
found in the WAIS-R. These data support the expectation that subtests of
similar functioning correlate more highly with each other than with tests
measuring different types of functioning. Furthermore, the intercorrelations
between the subtests on the WMI and between the subtests on the PSI sup-
port the inclusion of the Letter-Number Sequencing and Symbol Search
subtests in their respective indexes. In all, these correlations are evidence of
convergent validity of both the IQ and Index scores.
d
Lower correlations between variables that are not expected to be correlate
are evidence of discriminant validity. For example, this pattern occurs
the third
between Letter-Number Sequencing and the subtests not related to
is found in the relativel y
factor. Additional evidence of discriminant validity
domains. For example, the
lower correlations across domains than within
are higher
intercorrelations between Verbal subtests (ranging in the .70s)
from the .40s to the .50s).
than those between VCI and POI subtests (ranging
79
Evidence Base for Validity of the WAIS-III and the WMS-III

WMS-III Intercorrelations
Similar to the intercorrelation studies for the WAIS-III subtests, scales, and
indexes, there were a number of a priori expectations for the WMS—III inter-
correlation studies. First, because the WMS-III is designed to measure mem-
ory functioning, it was assumed that all of the memory subtest and index
scores would show at least low to moderate intercorrelations. Second, it was
expected that the immediate and corresponding delayed measures would
generally be most highly correlated. Third, the visually presented subtests
and auditorily presented subtests were generally expected to correlate more
highly with their modality-specific counterparts. Subtests within a particular
index were also expected to correlate more highly with each other than with
subtests of other indexes. Finally, because some WMS-III subtests yield mul-
tiple scores and because these scores may contribute to different Primary
indexes, relatively high correlation coefficients between scores that contain
the same subtest components would be expected. For example, a relatively
high correlation coefficient between Logical Memory I Recall scaled score
and the Auditory Recognition Delayed scaled score would be expected
because the recognition component of Logical Memory II partially con-
tributes to this latter score.
The intercorrelations of the WMS-III Primary subtest scaled scores and of
the Primary Indexes and Auditory Process Composites based on the entire
standardization sample are shown in Tables 4.2 and 4.3, respectively. The
means and standard deviations for each subtest and index are also provided.
The correlations for the overall sample are the average correlations across
all 13 age groups computed with Fisher’s z transformation. The intercorrela-
tions, means, and standard deviations for the WMS-III Primary subtests,
Primary Indexes, and Auditory Process Composites for each of the 13 age
groups are presented in Appendix A of this Manual. Additionally, Appendix A
presents similar information for the Primary subtest scaled scores and
selected supplemental subtest scaled scores for three age bands.
As expected and as shown in Tables 4.2 and 4.3, the intercorrelation coeffi-
cients of the Primary subtest scores and the Primary Indexes range from low
to very high. Intercorrelations of the immediate and delayed conditions of
the Primary subtests are also in the expected range, generally showing the
highest degree of association. The Letter-Number Sequencing and Spatial
Span subtests generally show the lowest correlations with the other memory
subtests. This same pattern also occurs between the Working Memory Index
and the other memory indexes. Overall, the auditorily presented subtests
intercorrelate highly with other auditorily presented subtests. The visually
presented subtests, Faces and Family Pictures, are not as highly correlated as
 81
Evidence Based on Internal Structure

O€ 0'€ Oe O€ Oe Oe Le O€ Oe oe oe as
0'0l 00} 001 0'OL 001 0'0l 0'0L 0'0L OOL 001 OOr uee|\|
SE" vv co 89° Go Zo ce" ov 8h OL |8q 99y pny
9" 8c Ov" LE: 82 16° 9¢" oe" ev’ J299Y |] Wd wey
oo Ov’ 02 Se ce 13° 6h ev 298 |] VWdA
oa. or 02 0¢" Lo 19° co) uojuBooey || seoe4

jeneds
LG 6¢° by os” 02° g8" E99Y |] WW

ueds
15)
Le
€o
02
Sv
Ka ge oa Ly Hulouanbes N-7

uoliuBooayY
ITE98Y
ve oe

vl VdA
Ov" JB98Y wey Wd|

12994
|W Sy
|se0e4
|8
120 jyeoey 1]2994 Booey \|2084 ueds Hulsuenbeg ~—s|jed0y [ye9ey fooay [je00y -ysayqns
dey PNY — || Ad WeF Il VWdA || se0e4 IW jeneds N-1 | Wd we4 | WdA | seoe4 |
sa8y [TV :S8109§ payeas }sa}qns Areullsg [II-SINWM JO SUOTRI9I110910}U] *Z"y BIGeL
82
a1qel"e'7UT SUOTje[a1109190}
Jo [JI-SINM soxopU]
pure ssayIsoduioy
ITV Sosy

Auewiid saxepu| Asoypny ssad0ig se}isodwoy a”

pny SIA WU] pny SIA pny uey By= fel-ajBuig
= Bulwesy
wu wu] wal| jad Ea 994 Jaq wal wal Huluseey adojs uolUa}oY jeAa}yey
Arewlid saxapu|
Aoypny ayeIpaww)
/eNsi/, ayeIpewiy| Be"
eyelpewu|Mowe) v8" rach
Aoypny pekejag 88° ov 6L
jensi, pahejag ov 78° GL or"
pny 984 leq 8s’ og eg’ lg ee
je1eue5 Mowe 1s” 89° 06° 88° el’ 99°
Buryon Nowayy ov 87 ey oy ee" ce" 9p"

-a[6uls
jeu) Bulusee7 06° ve GL GL 9° eS" OL sp
Burweeyadojg oa St ee ve 6L 6k 62 a Gl.

ssed0jg
Asoyipny
uoua}oy rach ve rAch BL’ Qe" 9e° 99° 62" er’ ak

sayisodwoy
[PABLO 60'- 70'- 80'- - Gz €0'- Sy €0' 20" rate QL Les
ueey\ 0°02 0°02 0'0v L'02 102 0°OL LOS 0°02 0°02 102 0°02 0°0
Evidence Base for Validity of the WAIS-III and the WMS-III

ads os 6v v8 ES) 8P Ove 9°01 oS 6v vv LY 9¢

 Evidence Based on Internal Structure

expected. The slightly higher correlations between Family Pictures and the
auditorily presented subtests may suggest the role of verbal mediation for
this visually presented subtest.
Table 4.3 also includes the intercorrelations of the Primary Indexes with the
Auditory Process Composites. The Single-Trial Learning Composite is highly
correlated with the auditory indexes, whereas the Learning Slope Composite
has relatively lower correlations with all of the indexes. The Retention
Composite is most highly correlated with measures of delayed memory,
whereas the Retrieval Composite generally has low correlations with the
indexes. The Retrieval Composite has an inverse relationship with the
Auditory Delayed Index (which is based on recall) and a correspondingly
positive relationship with the Auditory Recognition Delayed Index. The
direction of this relationship is expected because the Retrieval Composite
represents the difference between auditory recognition and auditory recall,
with positive scores indicating that recognition is higher relative to recall
and with negative scores indicating the opposite pattern.

WAIS-II! Factor Analytic Studies

The application of factor analytic techniques to the Wechsler intelligence
and memory scales has yielded important information regarding the
underlying constructs assessed by these measures. The results of factor
analytic studies on previous versions of the these instruments influenced
the development of subtests and the composition of indexes for the most
recent revision.
According to Wechsler’s theory (see Chapter 1), the subtests of a multifaceted
scale, such as the WAIS-III, can be aggregated into two domains—Verbal and
Performance—which can be further aggregated into one score, the FSIQ
score. Not until the 1950s was solid evidence of additional factors demon-
strated (see J. Cohen, 1952a, 1952b, 1957a, 1957b). Examiners began using
factor-based scores and began finding clinical utility in breaking out the
domains into more discrete units of functioning (Kaufman, 1990, 1994).
Researchers and clinicians have made various suggestions about how to use
these additional scores clinically (see Kaufman, 1994; Sattler, 1992; G. E.
Smith et al., 1992). Furthermore, some of the initial clinical research has
shown that examinees with different clinical conditions or diagnoses are
more likely to exhibit certain patterns of scores (Prifitera & Dersh, 1992;
Wechsler, 1991). On the basis of data supporting a four-factor model in the
WISC-III, the WAIS-III project included plans for new subtests that would
tap more discrete functioning, specifically working memory and processing
were
speed. The Letter-Number Sequencing and Symbol Search subtests
a priori
added to enhance these additional indexes. Table 4.4 illustrates the
hypothesis about the predicted factor structure for the WAIS-III.

83
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.4. Predicted Factor Structure of the WAIS-III

A SS SS See ee

Verbal Perceptual Working Processing

Comprehension Organization Memory Speed
Factor Factor Factor Factor

Block Design
Vocabulary Digit Span Digit Symbol—Coding
Matrix Reasoning
Similarities Arithmetic Symbol Search
Picture Completion
Information Letter-Number
Comprehension
Object Assembly Sequencing
Picture Arrangement
ee
ee

Numerous factor analytic methods were applied to the WAIS-III standard-

ization data to test the factor structure. Both exploratory and confirmatory
factor analyses were performed, and the results of numerous extraction and
rotation methods were compared and contrasted. The factor analyses fell
into five broad steps. Initially, comparability with the WISC-III was tested
with factor analysis. The second step was to test whether the addition of the
new subtests strengthened the four-factor structure. Third, the stability of
the factor structure was tested across different age groups. Fourth, the four-
factor model was tested against alternative models in a confirmatory analy-
sis. Fifth, the final composition of the indexes was developed.

Exploratory Factor Analysis

Exploratory analyses were run in different ways. When possible, data from
the entire standardization sample (N = 2,450) were used. For other analyses,
data from those in the WAIS-III standardization sample who also completed
the WMS-III (weighted N = 1,250) were used. Sometimes the standardization
sample was split into five age groups as a way of checking the consistency
across age: 16-19 years, 20-34 years, 35-54 years, 55-74 years, and 75-89
years. For other analyses, different sets of variables were added and sub-
tracted from the analyses to test the effects of the inclusion or exclusion of
certain subtests. Throughout most of these analyses, a four-factor solution
was supported.
The initial step in the examination of the factor structure of the WAIS-III was
an exploratory analysis to determine if a pattern of results similar to that
obtained for the WISC-III could be replicated in the WAIS-III. For this com-
parison analysis, two sets of subtests were used. The first set included the 12
subtests in common between the WISC-III and the WAIS-III, that is, the 11
traditional WAIS-R subtests and Symbol Search. The second set included the
13 primary subtests of the WAIS-III (excluding Object Assembly). A principal
axis method (principal axis methodology is also called common factor
analyses) was used, and the factors were not restricted to an orthogonal
rotation. When factors are correlated, as they were in this case, the model

84
 Evidence Based on Internal Structure

may be “overfitted” and produce “Haywood” cases if multiple iterations are

allowed for estimating communalities. Therefore, each analysis was restrict-
ed to two iterations (see Gorsuch, 1983, for a review)
Gorsuch (1983, 1996) and others have suggested that factor solutions be
evaluated not only according to empirical criteria but also according to the
criterion of “psychological meaningfulness.” Therefore, the results presented
here have been interpreted in light of the research literature discussed ear-
lier, the statistical criteria, and the model of intelligence presented in the
WISC-III (see Wechsler, 1991). For these analyses, four factors were specified
to be retained.
The factor pattern loadings based on the first set of subtests are presented
in Table 4.5. In general, these results replicate the four-factor solution that
was found in the WISC-III. Not surprisingly, the Vocabulary, Similarities,
Information, and Comprehension subtests all load most strongly on the
first factor, the Verbal Comprehension factor. The results also indicate a
strong second factor, Perceptual Organization, and two additional factors,
Processing Speed (third) and Working Memory (fourth). These results are
very similar to those found for the WISC-III.

Table 4.5. WISC-III Subtest Set: Exploratory Factor Pattern Loadings

for Four-Factor Solutions
Verbal Perceptual Working Processing
Comprehension Organization Memory Speed
Vocabulary .89 Sy. 01 .06
Similarities 76 A -.02 01
Information 82 -.03 .09 -.03
Comprehension 19 .09 .01 -.06
Picture Completion ake 56 ain ie
Block Design -.04 .70 19 -.03
Picture Arrangement 32 39 ~.05 .07
Object Assembly ~.01 .80 -.04 em fe)
Arithmetic .26 .04 ay. .04
Digit Span uit z.0% 45 15
Digit Symbol—Coding 02 S05 .02 73
Symbol Search -.04 a .07 65
i CLUE EEEEE EEE
Note. The interfactor correlations range from .63 to .73, magnitudes indicating that the amount of
shared variance between any two factors is equal to or less than 53%.
Evidence Base for Validity of the WAIS-III and the WMS-III

The second step was to determine if the new subtests strengthened the four-
factor structure, as anticipated. For this analysis, the 13 primary subtests
from the WAIS-III were used, with Object Assembly, the optional subtest,
excluded. As with the previous analyses, the correlations for the overall sam-
ple and a principal axis method with an oblique rotation were used. Also as
before, the analysis was restricted to two iterations.
The results of this analysis, shown in Table 4.6, also support a four-factor
solution for the WAIS-III. The first factor is again the Verbal Comprehension
factor, with the highest loadings by the Vocabulary, Information, Similarities,
and Comprehension subtests. The second factor appears to be Perceptual
Organization, with the highest loadings by the Block Design, Matrix
Reasoning, Picture Completion, and Picture Arrangement subtests. The third
factor is defined by the highest loadings by the Digit Span, Letter-Number
Sequencing, and Arithmetic subtests. In fact, the inclusion of the new
Letter-Number Sequencing subtest made this factor more salient because
the pattern of the factor loadings appears stronger, and more variance is
explained by this third factor. The fourth factor, Processing Speed, is made
up of the Digit Symbol—Coding and Symbol Search subtests.

Table 4.6. WAIS-III Subtest Set: Exploratory Factor Pattern Loadings

for Four-Factor Solutions
Verbal Perceptual Working Processing
Comprehension Organization Memory Speed
Vocabulary .89 a0 .05 .06
Similarities 76 10 -.03 .03
Information 81 .03 .06 -.04
Comprehension .80 .07 On -.03
Picture Completion 10 .56 713 SiMe
Block Design -.02 =A .04 .03
Matrix Reasoning 05 61 21 -.09
Picture Arrangement 27 47 -.09 .06
Arithmetic ee 15 51 -.04
Digit Span .00 -.06 mf .03
Letter-Number Sequencing 01 02 .62 als
Digit Symbol—Coding .02 -.03 .08 .68
Symbol Search -.01 16 .07 63
—_—_—————— eee
Note. The interfactor correlations range from .60 to .77, magnitudes indicating that the amount
of
shared variance between any two factors is equal to or less than 60%.
 Evidence Based on Internal Structure

The purpose of the next analysis was to test the stability of the factor
struc-
ture across age. For this analysis, the standardization sample was divided
into five age bands: 16-19, 20-34, 35-54, 55-74, and 75-89. Because there
were from two to three age groups in each band, the subtest correlations of
the age groups were averaged according to Fisher's z transformation. As
before, a principal axis method with an oblique rotation was used, iterations
were limited to two, and four factors were specified to be retained. Tables
4.7—4.10 provide the results of this analysis.

Table 4.7. WAIS-III Factor Pattern Loadings for Verbal

Comprehension Factor by Five Age Bands
ee eee
Age Bands
Subtest 16-19 20-34 35-54 55-74 75-89
Vocabulary 90 .89 92 88 86
Similarities .69 83 .74 .74 68
Information 79 83 71 «f0 .83
Comprehension .82 81 84 76 TAs
Picture Completion -.014 .00 ale 23 PE
Block Design .08 .00 ~.01 ~.05 Oi
Matrix Reasoning aif 10 13 .02 .06
Picture Arrangement .26 31 22 .26 .30
Arithmetic 2 .30 16 16 PH
Digit Span -.06 -.02 02 02 16
Letter-Number Sequencing ~.04 -.03 AG .07 .06
Digit Symbol—Coding .05 02 -.02 -.01 -.06
Symbol Search -.06 -.02 05 .02 1S

Note. For the youngest four age bands, the interfactor correlations are between .57 and .80. Lower
interfactor correlations were obtained for the oldest age band, ranging from .48 to .67.

87
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.8. WAIS-III Factor Pattern Loadings for Perceptual

Organization Factor by Five Age Bands
e
wo ee
ait ee
Age Bands

Subtest 16-19 20-34 35-54 55-74 75-89

Vocabulary “08 =O TAO ae -.06

Similarities 16 .04 07 18 04
Information .07 -.01 16 .02 05
Comprehension .02 .06 .00 13 07
Picture Completion 52 .67 49 53 02
Block Design 57 59 .68 fie) 39
Matrix Reasoning 49 46 .67 63 42
Picture Arrangement AN 56 56 50 .05
Arithmetic 202 .09 .30 19 21
Digit Span 01 sal ib a -.04 .01
Letter-Number Sequencing __.01 aii 02 s05 -.03
Digit Symbol—Coding 7.09 -.09 01 03 -.06
Symbol Search 19 We .09 22 04

Note. For the youngest four age bands, the interfactor correlations are between .57 and .80. Lower
interfactor correlations were obtained for the oldest age band, ranging from .48 to .67.

Table 4.9. WAIS-III Factor Pattern Loadings for Working Memory

Factor by Five Age Bands
Age Bands
Subtest 16-19 20-34 35-54 55-74 75-89
Vocabulary On 10 ~.04 10 .08
Similarities Ais: -.04 .07 -.06 02
Information OM 04 We 18 01
Comprehension 01 -.06 mOS .00 .03
Picture Completion OG .00 2 =i Sal
Block Design .02 21 Oi .03 ~.06
Matrix Reasoning 24 “oll LOZ .26 .02
Picture Arrangement 04 wale ~.04 ~.06 7.05
Arithmetic 45 47 1 .56 44
Digit Span .76 79 se .68 58
Letter-Number Sequencing _.79 .70 71 .60 62
Digit Symbol—Coding .04 .03 .07 .07 39
Symbol Search AC .04 .04 .07 37
ee
Note. For the youngest four age bands, the interfactor correlations are between .57 and .80. Lower
interfactor correlations were obtained for the oldest age band, ranging from .48 to .67.
 Evidence Based on Internal Structure

Table 4.10. WAIS-III Factor Pattern Loadings for Processing Speed

Factor by Five Age Bands
eee
Age Bands
Subtest 16-19 20-34 35-54 55-74 75-89
Vocabulary .09 ~.01 10 .08 .00
Similarities -.03 01 02 .02 18
Information .03 ~.01 = ails ~.08 .07
Comprehension ~.06 02 .07 -.02 =.03
Picture Completion 19 On 21 16 .62
Block Design .00 01 13 .07 51
Matrix Reasoning ~.05 OE 01 OV, .30
Picture Arrangement ~.08 .04 B02 .04 A7
Arithmetic 14 .05 -.01 =03 -.09
Digit Span 01 .06 02 .00 -.09
Letter-Number Sequencing -.01 -.01 .08 18 alli
Digit Symbol—Coding 74 72 .70 .68 52
Symbol Search 2 65 .66 59 .59

Note. For the youngest four age bands, the interfactor correlations are between .57 and .80. Lower
interfactor correlations were obtained for the oldest age band, ranging from .48 to .67.

In general, the factor structure resulting from this analysis confirmed the
previous results. For the four youngest age bands, the pattern found for the
overall sample is nearly identical. For the oldest age band, the subtest load-
ings on the Verbal Comprehension and Working Memory factors are consis-
tent with the results for the four other age bands. The subtest loadings on
the Perceptual Organization and Processing Speed factors, however, are less
clear. For this age band, the Picture Completion, Block Design, Picture
Arrangement, Digit Symbol—Coding, and Symbol Search subtests all have
high loadings on the Processing Speed factor (see Table 4.10). Matrix
Reasoning and Block Design have relatively high loadings (2.39) on the
Perceptual Organization factor as well (see Table 4.8). Therefore, the subtests
that loaded on the Perceptual Organization factor in the previous analysis
(i.e., Picture Completion and Picture Arrangement) no longer load on the
expected factor. Instead, these two subtests load on the Processing Speed
factor. Matrix Reasoning, an untimed measure of abstract reasoning, loads
on the Perceptual Organization factor. Block Design, which requires some
abstract problem-solving skills, has a secondary yet significant loading on
this Perceptual Organization factor.
Evidence Base for Validity of the WAIS-III and the WMS-III

It is unclear whether this pattern is caused by chance fluctuations or age

effects on the subtests composing the POI and PSI. The pattern may also
occur because most of the Performance subtests have time limits, and,
therefore, the processing speed on these subtests was heavily weighted for
this oldest age band. In spite of these results, the factor structure is quite
consistent with the global results. Moreover, results of the confirmatory
analyses and clinical validity studies support the use of the four Index scores
for an older adult population.

Contirmatory Factor Analysis

Based on theory, research, and exploratory analyses, a factor model can be
derived for a cognitive battery such as the WAIS-III and confirmed with
structural-equation modeling (Bentler, 1980; Bentler & Wu, 1993; Bollen,
1989; J6reskog, 1993). Confirmatory factor analysis is similar to exploratory
factor analysis. Both are methods of data reduction by which variables are
grouped into a smaller number of underlying related factors. A subtest that
is a measure of a factor is said to have a high loading on (or correlation with)
that factor. Confirmatory factor analysis differs from an exploratory
approach, however, because the grouping of subtests is made a priori rather
than being generated by a computer algorithm. Instead, the examiner pre-
dicts, on the basis of theory and previous research, how the data may be
grouped into factors. The specific relations between the variables (in this
case, subtests) and a latent underlying factor (in this case, the WAIS-III
indexes) are specified in a model, and that model is tested to determine if
the correlations between the variables support this a priori structure. In the
following analyses, different models were tested and compared with one
another so that the WAIS-III structure could be determined. Confirmatory
factor analyses were performed on the data for the entire standardization
sample and for the five age bands described earlier: 16-19, 20-34, 35-54,
55-74, and 75-89. For each of these confirmatory sets (the overall sample
and the five age groups), the following four structural models were tested.
Each of these structural models was compared to a general, one-factor
model. As with the exploratory analyses, the Object Assembly subtest was
not included in these analyses.
* Model 1 (One Factor): All 13 subtests on a general factor
¢ Model 2 (Two Factors): 7 Verbal subtests and 6 Performance subtests

¢ Model 3 (Three Factors): 5 Verbal Comprehension subtests, 4 Perceptual

Organization subtests, and 4 Attentional subtests (i.e., Digit Symbol—
Coding, Digit Span, Letter-Number Sequencing, and Symbol Search)
on Factor 3
 Evidence Based on Internal Structure

* Model 4 (Four Factors): 4 Verbal Comprehension subtests, 4 Perceptual

Organization subtests, 3 Working Memory subtests, and 2 Processing
Speed subtests
* Model 5 (Five Factors): Similar to a model suggested by Woodcock
(1990, 1997) and Flanagan and McGrew (1997), with the Verbal
Comprehension/ Knowledge factor (4 subtests), Perceptual
Organization/Visual Processing factor (4 subtests), Processing
Speed factor (2 subtests), Memory factor (2 subtests), and Quantitative
Ability/Numerical Ability factor (1 subtest—Arithmetic)

Successive factor models were evaluated according to a variety of goodness-

of-fit indexes, with emphasis given to those that are less sensitive to sample
size (N) or to the number of degrees of freedom (df) (see Bollen & Long,
1993; Marsh, Balla, & McDonald, 1988; Tanaka, 1993). Thus, the chi-square
index (x) divided by degrees of freedom (x2/df) was used to calculate the
Tucker—Lewis Index (TLI; Tucker & Lewis, 1973), which has been shown to be
particularly robust to differences in sample size and degrees of freedom. The
goodness-of-fit index adjusted for degrees of freedom (AGFI), from Jéreskog
and Sérbom (1993), was also used. The fit of the data to each of the five
models was also evaluated according to the root mean squared residual
(RMSR) index, a measure of the degree of reproduction of the covariance
matrix from the model estimates. Finally, the successive improvement in
model fit, moving from one to five factors is shown by the x? difference. The
TLI shows the comparative fit of each model to the one-factor model. The
results of each of these goodness-of-fit analyses are presented in Table 4.11.
The results shown in Table 4.11 confirm that the four-factor model best fits
the data for the total sample and for most of the age bands. The results show
that the fit improves as the number of factors increases. Significant improve-
ments in “model fit” were found on each fit statistic from a two-factor to a
three-factor solution and again from a three-factor to a four-factor solution.
Also, both the four-factor and five-factor solutions are roughly equivalent
overall, with slightly better fit statistics for the four-factor solution for most
of the age bands. When the data for the total sample were analyzed, a slight
yet significant improvement in the ? statistic and slight improvements in
the RMSR were found in the five-factor solution. However, there was no
improvement on the AGFI, x*/df, and TLI statistics. Overall, these confirma-
tory analyses support a four-factor solution.

91
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.11. WAIS-III Goodness-of-Fit Statistics for Confirmatory

Factor Analysis
Goodness-of-Fit Index Improvement
Model x? df x7/df AGFI RMSR x? df TLI

Total Sample (N = 1,250)

One-Factor 1159.8 64 18.1 782 Sei
Two-Factor 741.6 63 11.8 867 483 418.1 1 0.37
Three-Factor 473.4 61 7.8 913 PTL 268.2 2 0.60
Four-Factor 238.2 58 4.1 .954 221 235.3 3 0.82
Five-Factor 222.1 54 41 954 202 16.1 4 0.82
Ages 16-19 (n = 200)
One-Factor 220.1 64 3.4 182 5/5
Two-Factor 184.6 63 2.9 817 571 35.6 1 0.21
Three-Factor 123.7 61 2.0 877 394 60.9 2 0.58
Four-Factor 82.2 58 1.4 .910 .307 ates 3 0.83
Five-Factor 1foll 54 1.4 .908 .280 Bet 4 0.83
Ages 20-34 (n = 300)
One-Factor 342.4 64 B.S! .730 614
Two-Factor 253.9 63 4.0 813 612 88.5 1 0.30
Three-Factor 144.5 61 2.4 .900 379 109.4 2 0.67
Four-Factor 81.4 58 1.4 .936 .280 63.0 S 0.91
Five-Factor TASS) 54 14 935 257 49 4 0.91
Ages 35-64 (n = 300)
One-Factor 203.5 64 Se. 176 529
Two-Factor 138.9 63 ee 854 .463 64.7 1 0.45
Three-Factor 101.8 61 ik 887 361 Sle 2 0.68
Four-Factor 59.5 58 1.0 935 236 42.3 3 1.00
Five-Factor 57.8 54 ile 931 PEN iLafi 4 0.95
Ages 65-74 (n = 200)
One-Factor 289.5 64 4.5 .790 463
Two-Factor 184.1 63 2.9 .873 SSY/T/ 105.4 1 0.46
Three-Factor 158.1 61 2.6 882 334 25.9 2 0.54
Four-Factor 98.9 58 ier 924 245 59.2 3 0.80
Five-Factor 97.5 54 1.8 920 .240 ide 4 0.77
Ages 75-89 (n = 250)
One-Factor 306.7 64 4.8 724 651
Two-Factor 187.4 63 3.0 836 1557; 119.2 1 0.47
Three-Factor 142.7 61 2.3 .870 476 44.7 2 0.66
Four-Factor 96.7 58 lst 913 324 46.0 3 0.82
Five-Factor 85.4 54 kG 916 .286 11.4 4 0.84

92
 Evidence Based on Internal Structure

For the oldest age band, the five-factor solution is slightly, yet insignificantly,
better than the four-factor solution. Both of these models are better than the
three-factor solution that had been in question for this oldest age band in
the exploratory factor analysis. Nevertheless, the four-factor solution was
determined to be a more parsimonious and clinically useful solution than
the five-factor model. The four-factor solution comprises Verbal Compre-
hension, Perceptual Organization, Working Memory, and Processing Speed.
This solution has empirical support across the age ranges. It is also an
improvement over the five-factor model in which the fifth factor is defined
by only one subtest (Arithmetic). This model is clearly a superior solution
to a one-, two-, or three-factor solution and more parsimonious than a
five-factor one.

Composition of the Index Scores

The final step of the WAIS-III factor analyses was to determine the composi-
tion of the Index scores. The 13 subtests were not automatically included
in the four indexes. Instead, analyses were conducted to determine if all
13 subtests were needed, specifically, if the Vocabulary, Information,
Comprehension, and Similarities subtests were needed for the VCI and if the
Block Design, Matrix Reasoning, Picture Completion, and, perhaps, Picture
Arrangement subtests were needed for the POI. A procedure similar to that
employed by G. E. Smith et al. (1992, 1994), that is, hierarchical regression
analysis, was used to determine if three subtests would suffice for each of
these indexes. The results of regression analyses indicated that three sub-
tests are sufficient to adequately measure the ability tapped by each of
these indexes. Therefore, for the VCI, Comprehension, which requires a
longer administration time and more judgment in scoring than do the
Vocabulary, Similarities, and Information subtests, was omitted from the
VCI. Likewise, Picture Arrangement, which tends to have a significant sec-
ondary loading on the Verbal Comprehension factor, was omitted from the
POI. The POI is therefore composed of Picture Completion, Block Design,
and Matrix Reasoning.
Because at least two or three variables are needed to define a stable factor,
the WMI and PSI were left intact. The third factor (Working Memory) is
defined by Digit Span, Letter-Number Sequencing, and Arithmetic, and the
fourth factor (Processing Speed) is made up of Digit Symbol—Coding and
Symbol Search. All four of the Index scores have a mean of 100 and a stan-
dard deviation of 15 (see the WAIS-III Administration and Scoring Manual
for further discussion).

93
Evidence Base for Validity of the WAIS-III and the WMS-III

Saklofske, Hildebrand, and Gorsuch (2000) reported results of the factor

analytic structure of the WAIS-III in the Canadian standardization sample
(N = 1,105). The authors replicated the four-factor structure reported for the
U.S. standardization sample. The confirmatory models resulted in factor
loadings that were very similar between the two samples for all of the sub-
tests and their respective factors.
Arnau and Thompson (2000) evaluated the WAIS-III standardization data
utilizing a second-order confirmatory factor analytic model. Their analysis
yielded support for the four-factor structure reported for the standardization
sample with a second-order general ability factor that correlated with the
first-order factors. The authors noted that a modified model that allowed for
Picture Arrangement and Arithmetic to load on Verbal Comprehension and
allowed for covariance of error terms between Block Design and Object
Assembly and Digit Span and Letter-Number Sequencing resulted in a
slightly better model. Arnau and Thompson (2000) noted that complex load-
ings of these subtests are supported in previous research. Ward, Ryan, and
Axelrod (2000) reported similar findings to those reported here and in the
original WAIS-III-WMS-III Technical Manual but suggested that although fit
statistics are better for more complex models, the models with fewer factors
have good fit statistics and are more parsimonious.
Factor analytic studies support the configuration and structure of the
WAIS-III Index and IQ scores. In general, these factors have similar subtest
loadings across studies and across age-bands. For the oldest age group
(75-89 years), the factor analytic results of the standardization sample data
indicated a greater overlap among measures of processing speed and per-
ceptual organization. Further research is needed to replicate this finding in
older adults; however, clinicians and researchers should note this shift in the
relationship among subtests when interpreting patterns of test results for
older populations.

Development of a Short Version of the WAIS-III

J.J. Ryan, Lopez, and Werth (1998) administered the WAIS-III to 62 patients
and reported administration times considerably longer than those for
healthy individuals composing the normative sample. The extended testing
time for administering the full WAIS-III has resulted in the development of
several short forms for estimating IQ scores (Blyler, Gold, Iannone, &
Buchanan, 2000; Pilgrim, Meyers, Bayless, & Whetstone, 1999; J.J. Ryan,
1999; J.J. Ryan & Ward, 1999). These studies use Ward’s (1990) seven-subtest
model, or a variation of that model, for estimating intellectual functioning.
Ward's model is composed of Picture Completion, Digit Symbol-Coding,
Similarities, Block Design, Arithmetic, Digit Span, and Information; the

94
 Evidence Based on Internal Structure

variation includes the same subtests except Block Design, for which Matrix
Reasoning is substituted (J.J. Ryan, 1999; J.J. Ryan & Ward, 1999). The exam-
iner is cautioned that the development of short forms based on selected,
specific subtests reduces the reliability of a composite score. Axelrod (2002)
reported short-form reliability coefficients of .92-.96 compared to those for
the IQ and Index scores based on all contributing subtests, .94-.98.
Therefore, the published confidence intervals for the IQ and Index scores do
not apply to short forms. Published tables for the determination of signifi-
cant differences and base rates of difference scores between IQ and Index
scores and between IQ and memory index scores do not apply to short
forms. These require the reliability coefficient of the composite scores, and
base-rate tables are affected by the relative score distributions of the com-
posites and the correlation between the composites. Studies of the effects of
administering the selected short-form subtests separately from the full
WAIS-III are needed to determine whether performance on the subtests is
equivalent in the context of the short forms and full battery.
Finally, a composite score should not be reported if there are significant
differences in performance among the subtests that make up the composite.
The use of a short form may occlude an examinee’s profile variability
that would have been evident in his or her performance on the full battery
of subtests.

Development of a General Ability Index

Tulsky, Saklofske, Wilkins, and Weiss (2001) developed an alternative index of
general intellectual ability based on the eight subtests that compose the VCI
and POI (i.e. those with the highest loadings on the general ability factor g).
Tulsky et al. used the WAIS-III standardization sample data to derive the
norms for a general ability index (GAI) and proposed that the GAT is a
better measure of general cognitive ability than the FSIQ because it is com-
posed of subtests that are more interrelated and have the highest loadings
on g, and the GAI subtests are those most likely to be spared the effects of
brain insult and thus to provide a more stable estimate of general ability.
However, the interpretation of the GAI as opposed to the FSIQ should be
informed by the knowledge that GAI intentionally excludes measures of
working memory and processing speed, which may be differentially sensi-
tive to brain injury and pathology (see the later discussion).

Wechsler Abbreviated Scale of Intelligence

The WASI (1999) was developed by The Psychological Corporation to provide
es
a fast, reliable method of estimating intellectual functioning in examine
is compos ed of four subtests and yields estimate d
aged 6-89 years. The WASI
Evidence Base for Validity of the WAIS-III and the WMS-III

FSIQ scores based on either two or four subtests. The VIQ and PIQ can also
be estimated if all four subtests are administered. The subtests measure sim-
ilar constructs as those measured by the WAIS-III, WISC-III, and WISC-IV
(in development) but do not contain the same content. The WASI subtests
are alternate forms of the corresponding subtests of the WAIS—III, WISC-III,
and WISC-IV. The WASI was normed on a large, nationally representative
standardization sample and is directly linked to the WAIS-III and WISC-III.
Predicted intellectual functioning and prediction intervals are provided.
The WASI may be completed in 15-30 minutes based on the number of
subtests administered.
Axelrod (2002) compared estimated VIQ, PIQ and FSIQ scores based on
the WASI and prorated WAIS-III short forms. The WASI scores exhibited
higher reliability and similar level of association with WAIS—III measures
(controlling for part-whole associations) as the prorated short-form scores.
Axelrod reported better prediction of IQ scores with the short forms based
on actual versus predicted discrepancies. This conclusion does not account
for the presence of part—whole associations, which cannot be statistically
controlled in a difference-score methodology. The strength of the association
between two variables affects the distribution of difference scores; specifi-
cally, when the inflated association due to part—whole relationships is not
controlled, the difference scores will overestimate the predictive accuracy of
the short forms.

WMS-Ill Factor Analytic Studies

The construction of WMS-III composites and Indexes was based on theoret-
ical principles and on the results of research on the test’s predecessors. An
initial confirmatory factor analysis of the data from the WMS-R standardiza-
tion sample and a mixed clinical sample indicated that a two-factor solution
was the best fit. These two factors consisted of an immediate memory
dimension and an attention/concentration dimension (Roid et al., 1988).
Measures of delayed recall were not included in these analyses. Roth,
Conboy, Reeder, and Boll (1990) reported that when delayed recall measures
were added to the analyses and the method variance shared by the immedi-
ate and delayed measures was controlled for, a three-factor solution pro-
duced the best fit of the data from a sample of individuals with closed head
injury. This three-factor solution consisted of attention/concentration,
immediate memory, and delayed memory. Using a similar methodology that
controlled for method variance, Burton, Mittenberg, and Burton (1993)
found that the three-factor solution found by Roth et al. (1990) also best fit
the observed relationships in the WMS-R standardization data. With findings
consistent with this three-factor solution, Woodard (1993) reported that the
attention/concentration, immediate memory, and delayed memory dimen-
sions were the best fit of several competing models of the data from a mixed
clinical sample of individuals who had sustained mild, diffuse brain injury.
 Evidence Based on Internal Structure

The results from these WMS-R studies suggest at least three general conclu-
sions. First, in all of these studies, those measures that compose the con-
struct of attention/concentration and the memory measures were identified
as separate dimensions. Second, when both the immediate and delayed
memory measures were used in the model, a three-factor solution appeared
to best fit the data based on both normally functioning and clinical samples.
These studies of two- and three-factor dimensions of the WMS-R provide
evidence of construct validity for two useful dimensions of clinical memory
assessment (i.e., attention/concentration versus memory, and immediate
memory versus delayed memory). Third, the separation of the memory mea-
sures (whether immediate or delayed) into modality-specific dimensions
does not provide a more parsimonious explanation than does conceptualiz-
ing memory as a unidimensional construct.
In the original publication of the technical manual, the results of confirma-
tory factor analytic studies were reported as best supporting a five-factor
model. The five factor-model specified working memory, auditory immedi-
ate memory, auditory delayed memory, visual immediate memory, and visu-
al delayed memory factors. Millis, Malina, Bowers, and Ricker (1999) used
the correlation matrices provided in the technical manual but failed to repli-
cate the results reported there. The authors noted the presence of inadmissi-
ble parameter estimates (e.g., correlations exceeding 1.0) that were
attributable to the very high correlations between immediate and delayed
memory factors. The authors concluded that model specification errors
(immediate versus delayed) produced results that indicated higher correla-
tions between subtests across factors compared to the correlations among
measures within the factor. The authors also concluded that the low correla-
tion between Faces and Family Pictures and the overall low communality
estimates for the Faces subtests contributed to the model-specification
problems. Despite not having empirical support for the differentiation
between immediate and delayed memory in the factor analytic study, the
authors advised the continued use of immediate and delayed index scores
on clinical and theoretical grounds and also suggested that further research
is needed.
Subsequent to the publication of Millis et al.’s (1999) findings, The
Psychological Corporation, in conjunction with Millis, completed further
study of the WMS-III factor structure. Price, Tulsky, Millis, and Weiss
using
(in press), analyzed the original WMS-III standardization data
and confir med the results in a
confirmatory factor analytic procedures
cross-validation study. Five models were tested:

° Model 1 (One Factor): General Memory

© Model 2 (Two Factors): Working Memory and Memory

97
Evidence Base for Validity of the WAIS-III and the WMS-III

* Model 3 (Three Factors): Working Memory, Immediate Memory,

and Delayed Memory
* Model 4 (Three Factors): Working Memory, Visual Memory, and
Auditory Memory
° Model 5 (Five Factors): Working Memory, Auditory Immediate Memory,
Auditory Delayed Memory, Visual Immediate Memory, and Visual
Delayed Memory.

In the models specifying more than one factor, the factor scores were
allowed to co-vary. The reassessment of the factor structure originally
reported in the WAIS-IJI—WMS-III Technical Manual was conducted to pro-
vide clinicians with the most up-to-date modeling of the WMS~III latent
structure. This study included an assessment of the data for nonnormality,
an analysis of the model for all three age groups combined and for each of
the three age groups separately, and a cross-validation of the results with
data from an independent sample.
Parameter estimates were derived with maximum likelihood procedures.
Models were compared through the computation and analysis of multiple fit
indices including the likelihood ratio chi-square (x?) statistics; the rescaled
(robust) chi-square statistics (corrects for multivariate nonnormality), the
adjusted goodness-of-fit index (AGFI), the root mean error square of approx-
imation (RMSEA), the nonnormed fit index (NNFI), the comparative fit index
(CFD, the relative chi-square, the Akaike information criterion (AIC; Akaike,
1987), and the expected cross-validation index (ECVI). The models were
screened for model-specification errors, that is, inappropriately estimated
parameter values and positive definiteness of the covariance matrices. Table
4.12 presents the results of these analyses for the calibration sample (i.e., the
WMS-III standardization sample), and Table 4.13 presents the results for the
cross-validation sample. Results are presented for each of the three age
groups and for the total group.
The results confirm the work of Millis et al. (1999); that is, Models 3 and 5
were identified as having nonpositive covariance matrices and boundary
solution violations due to an estimated correlation of greater than .99 for
immediate and delayed memory. These errors indicate the high degree of
multicolinearity between immediate and delayed memory measures, which
results in inaccurate model estimations and precludes the use of factor
analysis for validating this structure.
 “suolnjos Auepunog jo UOHRIOIA pue SayeWISe jePOW a|qissiWPeu| ,
Evidence Based on Internal Structure

‘(Auowew Hurjiom pue ‘padejap jensia ‘paAejep Auoypne ‘ayeipewu! jensia ‘eyeipewuui Auoyipne)
anbijqo 10}9e}-aAl} = s japoy ‘(s}sayqns peAejep puke syeIpeW! paulquiod YIM ‘uowaW Buryom pue ‘Auowealw jensia ‘iowa Aioyipne) enbijqgo J0}9e}-8e1U} = 7 |ePOy ‘(SUONIPUDD peAejap
pue ejelpeww ejeredes UM AIoweW HuDom pue ‘Auowew patejap ‘Aiowew ayeipattu!) anbijqo 10jOe}-de1y} = ¢ Japoyy ‘(AsowWetwW Bunom pue AyoWew Auoypne) anbijgo 10}0e}-OMm} = Z [EPO
‘Jo}0e} BUO = | JaPOyy “sisAjeue siyj} JO} pasn sem (Qgz‘L =) a|dwes |I|I-SWM pelyBiem eu] ‘uoIssiwued YM paydepy ABojoyoAsdounay jejuewledxy pue /ediul/D Jo jeusnor ‘sseid ul
‘ssiaM 7] PUR ‘SIN S ‘ANSINL “GQ ‘80d *Y 77 Aq ,‘uolepiea-ssoso YIM siskyeue s0jOR} MIOEWUIUOD *|||-a/eoS AiOWEW Ja|SYOa/A OY} JO SuNjONuJs 10}Oe} Oy) BulUyepay,, WO ‘aJ0N
er 30) OL’ €S'6cl 66° 66° €0° Le" 00° €6'c lL? eg 19 vS' 79 2G
vl -60° rc OL rl 66° 66° bO° Le 00° ocos 82 91°06 08'°€6 v
pA & vl BL SZL 66° 86 SO’ 96° 00° cer 82 SL LSL vS ES Be
ZE=ck vl B89 PLL 66° 86° SO’ 96° 00° Sl Y Oe 89° Pel €9°9cL @
lé-6y ec 86° 98¢ LO 96° LO" ee" 00° OLZ Le 86°8E¢ plLerve L
69-91
(oSz‘
seby
=u)
Seg ee 92° cB LoL 66 86° bO° cioe $0" LOL Le cs ce seZe eG
4 eet Bg vo cZ 601 66° 86° vO" Se 00° 661 82 cl SS €L 19 v
th —ca Lie eZ 6LL 86° L6 SO’ v6 00° ve? 82 €L°99 cl 69 Be
ve ec Lo OS’ eel 86° LO 90° v6" 00° Ly? oe os'cl SLOL CoA
Sr -0E° LE Ov r9l L6 S6 80° le 00° SLE Le Ov OLL L8’6LL }
(0Sp =u) 68-S9 seby
92 -co co 12°68 O0O'L 00° 00° L6 ov €0'l Le LLL? Sc2cc eG
LG—-1é ra 16°68 00'L 66° €0° L6 vl’ 82 b 82 L6°SE S09 v
Sc—le. Eo OO'OLL 00'L 66° €0° 96° 60° Leak 82 6v'°8E 96°8E Be
82-02 Ge Le°'06 O0O'L 66° €0° 96° 20° pel O€ Le Or ce Ov (4
cece fe pr 90L 66° 86° v0’ sé 00° 8e8'L Le vy ss LO°8S L
=p9-0¢
(00r
seby
4)
Ov -82° 9 Oa PL LEL 86 96° LO c6 00° £0'e Le bleg €9°S9 2G
6€-90° ce gs 9c 86° L6 90° £6 00° 6S°¢ 82 SS'cl os'el v
er —-62 se SS 6EL L6 96° 20° ce 00° SO’e 82 so’S8 LS°06 ue
roa ve" cL 9EL L6 96° LO ce 00° 68°c oe e298 19°06 A
8S -—6o° Sr €Z 061 sé 6 60° 88° 00° 09'r Le el evk 8Z'0SL L
(00b =U) 62-91 seby
IA0d%06 IADH «=[Link]—Cié‘*STNNN~=Ss«CSSIN—CdYtC‘“‘é JP /2X Jp (asnqoy) 7X 2X lapoy/dnoiy aby
sasAjeuy 10398]
AIOVULIJ 10J
SONSHLIS
1,J-JO-ssaupooy a[dureg
(uoHezIprepueys) UONeAqIeD II-SWM age ‘ZL"h
 OIGe]L “EL TII-SIWM. UONePTLA-
a|duresg
SSOID }L,J-JO-ssa
SINSTILIS
IOJ
upooy AIOJCULILJUOD a2)
5

100
10198] sasAjeuy a>)

5
aby lepow/dnosy5X 7X (asnqoy)
Jp 2X{P| d IdDV VaASINY IUNN IdO ~=OOIV IADR %06 IAOd E
Q
saby 62-9)=U) (Shp Z
L v9V8 €6°08 KS LO? 00° v6 90° 86° 86° €6'8ck LG Sacer a
Cc 67 c0 Lr OO’ O€ 1 ZG €0° 96° SOl 66° 66° 00°26 0d -Ll°Go 5
ae ry ce Vara 8¢ LoL vO 96° €0° 66° 66° Z€°96 0c BeAl Aro
if >
v VV cL er
OL 8c tL vS €0° 96° €0° 66° 66° OL'Z6 0c Sie =
2G 0c 6Z' EC 9E ke Let (Ao Z6 LO’ O0O0'L 00'L 9E°L6 6L° 8k co =
seby p9-0¢=u) (662 >
fd S66E 8E LZ Le Gol OL" v6" €0° 66° 66° L298 ve ShesER a
G L9°S¢ LEE? 0¢ 620 BL 96° 00° 00'L 00'L LeveL EG: -le’pe a
ae GOES L66L 8c EZ0 L8 LO 00° LOL O0'L LEEL (ey —-eo°
Se =
v L9°%?o le €0' 82 SZ'0 con LO 00° LOL O0O'L €0°SZ ce —cevs <
2G S8'sl OZ'9L Lo 080 el 96° 00° OO'L 00'L v8 OZ’ ve’ SE=7Si A
saby 68-s9=u) (151 =
L rz S6 88°02 Le 8c°° 00° v3" 60° £6" clos 88'8LL 62° 86'—-S9° es
A, 0S 8c Ly OL 0€ I ZG (Aoy 88° 90° 96° LE 01°26 SQ" 08—-SS° =|
as 8Z°0S OSE 82 L SG €0° 88° 90° 96° 86° 0S'Z6 SQ" 08-95" aa
v 8¢°6E SE 66 82 8b vl 06° v0’ 86° 66° 66°68 09° €Z-SS° =
eG OF oc le 6E° Le lL Oe Of; 06° vO" 86° 66° 6€°S6 bQ° Sl-6S°
seby 68-9)=U) (ggg S
L col c0 Oct EL Le LEE 00° S6° 90° 86° 86° BOL
El 0g -9l°ve as
Gg €9'901 89'rel 0€ Sly 00° 96° SO’ 86° 66° 89'PZL vl 4s Vl =
ls VY €9 ILD vv 8¢ L Br SO 86" c0° 00'L O0O'L vr's6 1 vl—-60° =
v €¢°8E 9 69 82 EGEE Sif 86° c0° 00°L 00'L 69°06 LE Us SiO
2G OZ'6L Syst Lo 68'0 co" 66° 00° 00'L 00°L 98 Sv OL’ Be 0] Al
SSS
‘aJO/y WOJ4 HulUJapay,,
eu; JOJO) aiNjoNAs
JO aU} Ja[syoeM\ Kiowa :|\|-e1e05 KiopeWUyUOD
1ojoe) sishjeue
YIM ,‘UONepIeA-ssoso
Aq “7 “H ‘20d
“G “AYSINL
*S ‘SIIIWAPUR“7 ‘SSIOM
ul ‘sseud jeusnor
Jo /PdIUI|D
pue jejuawiedxy ‘ABojoyoAsdoinay
pe\depy
YIM “UOISS!WWEd
japO|)
| = BU J0}9e} jepoyy
z = s0}9e}-om)
enbijqo Aoyipne) Arowaw
pue Bupyom ‘(Mowat
|EDO|\
€= 10}0eJ-ee1y)
enbijqo eyeipawwi)‘uowew pekejap ‘Auowaw
pue Bupyom owe yim ayesedas a}eIpewUWU!
puk peAelep !(SUONIPUCD
|apoyy
p= 10)0e|-9e1U)
enbijqo Aioypne)
‘Aowewjensia ‘Auowew
pue HuryJom ‘AWoWWeW
YIM PEUIQWOD @JPIPOWLU!
pue peAejep ‘(s}se}gns
japoyy
g= 10}9eJ-anl)anbi\qo A1oyIpne) ‘ayeipawitu!
jensia ‘ayeipewiulKoypne ‘pakejaplensia
‘peAejap
pue Bulyiom ‘(Aiowaw

, a[gissiwpeu|
japow! sa}ew|se
pue UO}eIOIA
Jo Arepunog ‘suolnjos
 Evidence Based on Internal Structure

On the basis of the fit measures in the calibration sample and ECVI values
from the cross-validation study, Price et al. (in press) concluded that Model 4
has the best fit to the data. In this model, the subtests load onto three fac-
tors: Auditory Memory, Visual Memory, and Working Memory. Table 4.14
provides the factor loadings of Model 4 for the calibration and cross-valida-
tion samples. Table 4.15 provides the interfactor correlations for Model 4,
and Table 4.16 presents the invariance statistics for Model 4 for the calibra-
tion and cross-validation samples.

Table 4.14. Standardized Parameter Estimates for Model 4

WMS-—lIl Calibration Sample Cross-Validation Sample
(N = 1,250) (N = 858)
Auditory Visual Working Auditory Visual Working
WMS-lIl Subtest Memory Memory Memory R? Memory Memory Memory
Letter-Number Sequencing 84 64 83 .68
Logical Memory Delayed 78 .60 16 58
Logical Memory Immediate tg 54 SiMe oh)
Spatial Span 55a lees?) Yh eye
Verbal Paired Associates Delayed .63 ou 64 A2
Verbal Paired Associates Immediate .66 42 .69 48
Family Pictures Delayed 79 .60 .69 48
Faces Immediate 37 14 38 ats}
Faces Delayed 39 14 43 18
Family Pictures Immediate es) 56 .66 43

Note. From “Redefining the factor structure of the Wechsler Memory Scale-Il!: Confirmatory factor analysis with cross-
validation,” by L. R. Price, D. Tulsky, S. Millis, and L. Weiss, in press, Journal of Clinical and Experimental Neuro-
psychology. Adapted with permission.

Table 4.15. Interfactor Correlations for the Calibration and

Cross-Validation Samples for Model 4
Factor Calibration Sample Cross-Validation Sample
(N = 1,250) (N = 858)
1 2 3 1 2 3

Auditory Memory
Visual Memory 74 82
Working Memory 65 AQ Le! 57
i TT EEENUIE ESSE SEES

Note. From “Redefining the factor structure of the Wechsler Memory Scale-Ill: Confirmatory factor analysis with cross-
validation,” by L. R. Price, D. Tulsky, S. Millis, and L. Weiss, in press, Journal of Clinical and Experimental Neuro-
psychology. Adapted with permission.

101
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.16. WMS-III Invariance Statistics for Model 4 for the

Calibration and Cross-Validation Samples
Method x2 df F* AGFI AIC CFI NNFI RMSEA ECVI 90% CI

M, 90.16 28 .07 197 1P* 144169".99 BE, 04 A .09-.14

M, 113.86 35 .33 96 153.86 .98 98 05 18 14-.22

M, 141.35 38 34 96 175.35 98 99 .06 .20 SEZ)

M4 171.52 48 33 96. 185.52" .97 .98 05 21 17=.27

Note. From “Redefining the factor structure of the Wechsler Memory Scale-Ill: Confirmatory factor analysis with cross-vali-
dation,” by L. R. Price, D. Tulsky, S. Millis, and L. Weiss, in press, Journal of Clinical and Experimental Neuropsychology.
Adapted with permission. Cross-validation N = 1,250; calibration N = 858. M, = Free estimated model; Mz = Equality of
factor loadings; M3 = Equity of factor loadings, factor variance/covariance; M, = Equity of factor loadings, factor
variance/covariance, and error variances.

* Maximum likelihood fitting function.

Tulsky and Price (in press) developed norms based on the collapsing of the
immediate and delayed recall indexes into new visual and auditory memory
indexes. The clinical utility of these new factors will need to be investigated
in future studies. Despite their findings, Price et al. (in press) did not recom-
mend that clinicians discontinue the use of the immediate versus delayed
memory composites solely on the basis of the inability of factor analysis to
provide statistical support for that original model. The theoretical rationale
for the development of the configuration and procedures for the WMS-III
are detailed in Chapter 1 of this Manual. A review of clinical research in-
vestigations into the nature and severity of memory impairments associated
with specific neurological disorders is presented later in this chapter. The
performance of these clinical groups on the WMS-III indexes is also report-
ed there. The Psychological Corporation recommends the continued use of
the immediate and delayed indexes on the basis of both clinical and theoret-
ical considerations.

Correlation Between the WAIS-III

and the WMS-Ill
Although the WAIS-III and WMS-III index scores were derived independent-
ly of one another, the co-norming of the tests creates the opportunity for
evaluating the relationship between measures of intelligence and memory.
Understanding this relationship enables the clinician to better evaluate the
patterns of strengths and weaknesses. Moreover, the studies described here
establish the degree of relationship between intelligence and memory and
provide support for the use of significant discrepancy models as a means of

102
 Evidence Based on Internal Structure

determining relative impairment in memory compared to general intellec-

tual ability. If intelligence and memory are unrelated, then performance dis-
crepancies may be common and potentially lack clinical meaningfulness.
These studies also provide evidence of divergent validity and establish that
the tests, although related, measure different constructs.
The WAIS-III IQ and Index scores and the WMS-III Index scores were inter-
correlated for the purpose of evaluating the relationships between the two
scales. The sample for this analysis was the WMS-III standardization sample
because each WMS-III standardization participant was also administered
the WAIS-III. The WAIS-III and the WMS-III were administered in counter-
balanced order during the same testing session. The demographic character-
istics for this sample are described in Chapter 2 of this Manual. It was
anticipated that the WAIS-III IQ and Index scores would be moderately cor-
related with the WMS-III Index scores, that is, that most correlations would
range from the .30s to the .60s. Furthermore, the WMS-III visual indexes
were expected to correlate more highly with the WAIS-III PIQ scale and POI
than with the VIQ scale and VCI. The opposite pattern for the WMS-III audi-
tory indexes was also anticipated; that is, these indexes would correlate
more highly with the WAIS-III verbal measures than with the WAIS-III per-
formance measures. Finally, the Working Memory indexes of the two scales
were expected to correlate highly.
Table 4.17 presents the correlation coefficients, means, and standard devia-
tions of the sample's performance on the two tests. The correlation coeffi-
cients between the WMS-III auditory and visual indexes and the WAIS-III
VIQ and PIQ scales are in the expected direction. This pattern of correlations
between the WMS-III auditory indexes and the WAIS-III VCI and between
the WMS-III visual indexes and the WAIS-III POI is relatively higher in mag-
nitude than the other correlations, with one exception. The WMS-III Visual
Delayed Index correlates equally with the WAIS-III VCI and POI. All WMS-III
indexes are moderately correlated with the WAIS-III PSI, with correlations
ranging from .35 (Visual Immediate Index) to .55 (Working Memory Index).
The high correlation between the WMS-III Working Memory Index and the
WAIS-III Working Memory Index (.82) is expected because they measure a
similar construct in addition to sharing one subtest. The correlations
between this WMS-III index and the other WAIS-III measures are in the
moderate range, from .51 (VCI) to .68 (FSIQ).
The pattern of correlations provides evidence of divergent validity, that is,
that the two scales measure different constructs. Some modality-specific
effects can be observed: WAIS-III Performance measures display a higher
correlation with visual memory than do the WAIS-III Verbal measures;
WAIS-IH Verbal measures correlate higher with auditory memory measures

103
 3jge]
|"*/ SUOTIeIALIOD
UseMjog
24) TTI-SIVM
pue 9) ITI-SIWM =
cy

104
III-SIVM IlI-SWM 8
OIA Old oIss IOA 10d IWM ISd uray as 9
I-SWM &
Aoypny eyelpeww| 8S PAC 6S" 1¢° or’ Ly Aa 6°66 v16 5
/eNsiA aye|PewU] og" 6" ge" 6 Ze 92" Ge" 8°66 O'Sl B
8ye/pewiw|Kiowey es" rs 1g’ Zs" Ly vr’ or 666 ZSt S
Auoupny pefejeg Ss os’ Lg vs" vr’ Sy rae 0001 6k a)
sensi, peAejag Se" vr rae 9c" 9" oe" 6e" L‘OOL Sh >
Aoypny uollubooey
pahelaq ly’ vy 6Y" ly Be" ov ge" LO0L Osh oat
(248ue5) iowa 9g" 9g" 09° 9s" Sr Ly Sy" 0'001 ISL s
Bupyom Nowe, rch GQ’ 89° LS" rach Z8° cs’ 0001 Srl =
II-SIVM a
ues S‘0Cl 0001 €'001 S001 8°66 0001 2001 a
as Syl Srl Sv Zvi vv OSI ork G
eee eee
ener eee Ssesss SS SSS Ss— ae!
‘@]0N=N ‘0S2'l =
o
:
=
 Evidence Based on Internal Structure

than with visual memory measures. Overall, the WAIS-III measures correlate
highest with auditory memory than with visual memory. Interestingly, pro-
cessing speed did not display a more significant correlation with visual
memory measures than with auditory measures. All of the WMS-III visual
memory measures have a brief stimulus exposure time, and impairments in
visual processing speed might have a more adverse effect on these tasks than
on auditory memory tasks. The results indicate similar levels of association
between the WAIS-—III PSI and visual and auditory memory performance,
with the correlations with auditory measures slightly higher than those with
visual measures. Clinically, impairment in general intellectual functioning
would be expected to have an effect on memory measures, particularly
working memory, general memory, immediate memory, and auditory imme-
diate memory measures. If intellectual impairment affects performance on
memory, the effects on visual memory tasks may be less observable.
The results of this study support the association between intelligence and
memory and the potential utility of discrepancy models for determining
weaknesses and impairment in memory relative to general ability level. In
subsequent research, Hawkins and Tulsky (2001) reported base rates for
FSIQ versus GMI stratified by FSIQ level. The results of this study indicated
that at higher IQ levels, general memory is more frequently lower than IQ.
The opposite trend was observed for individuals with lower IQs; that is, gen-
eral memory scores were often greater than IQ scores. These findings will
enable the clinician to determine the base rates of discrepancies, which can
aid in the interpretation of an individual’s performance.

Joint WAIS—III/WMS-III Factor Analysis

The co-norming of the WAIS-III and WMS-III has enabled researchers to
perform more sophisticated studies of the relationship between intelligence
and memory, specifically, studies to define and differentiate the factors that
underlie intellectual and memory assessment. In one such study, Tulsky and
Price (in press) examined the combined WAIS-III/WMS-III factor structure,
applying confirmatory factor analytic procedures. The authors used all the
WAIS-III subtests except Object Assembly and included all of the WMS-III
subtests (including Visual Reproduction and Word Lists). The authors devel-
fit
oped several measurement models, which they assessed using multiple
in the Price et al. (in
indicators. Their results were similar to those reported
press) study.
a six-factor model
Tulsky, Ivnik, Price, and Wilkins (in press) concluded that
following fac-
demonstrated the best fit to the data. This model specified the
Visual— Percept ual Organiz ation, Workin g
tors: Verbal Comprehension,
Auditor y Memory , and Visual memory . As in the
Memory, Processing Speed,

105
Evidence Base for Validity of the WAIS-III and the WMS-III

WMS-III factor analysis reported previously, the results of the joint

WAIS-III/WMS-III factor analysis suggest that memory functions cluster
by the modality of sensory input. The authors prorated norms to parallel
the observed factor structure. Further, the authors developed an alternate
visual memory index, substituting Visual Reproduction for Faces, citing the
relatively lower loading of Faces on the Visual Memory factor (Tulsky et al.,
in press).
The work by Tulsky and Price (in press) and Tulsky et al. (in press) provides
further evidence of the structure of the WAIS-III and the WMS-III. However,
it is still not clear whether these findings reflect the neurological structure of
memory or an artifact related to the statistical interdependence of the im-
mediate and delayed tasks within each modality. More clinical research and
brain-imaging studies of the underlying structure of memory are needed
before firm conclusions regarding the composition of the memory compos-
ites can be made. The alternate memory indexes developed by Tulsky et al.
will provide researchers with flexibility in determining the most diagnostic
composites for their specific populations.
As with any composite, if the scores on the subtests that make up the com-
posite differ significantly, the composite score should not be reported as rep-
resentative of the individual’s functioning in that domain. Ryan, Ament, and
Arb (2000) calculated values for determining significant differences between
WMS-II subtest scores.

Convergent Evidence for the WAIS-III

as a Measure of Intellectual Functioning
The WAIS-III is the most recent revision of the WAIS (Wechsler, 1955). The
constellation of tasks and composition of index scores has been changed in
the current version; however, the primary function of the WAIS-III, as with
its predecessors, is the assessment of intellectual functioning. The prelimi-
nary evidence that the WAIS-III measures intellectual ability was gathered
from a series of comparative and concurrent validity studies. In these stud-
ies, the WAIS-II was correlated with established tests of intelligence, with
the expectation that WAIS-III 1Q and Index scores would correlate highly
with these other measures of intelligence.

Correlations With the WAIS-R

A sample of 192 adults aged 16-74 (M = 43.5 years, SD = 20.2) were adminis-
tered the WAIS-R and the WAIS-III in counterbalanced order. The interval
between testings ranged from 2 to 12 weeks, with a median of 4.7 weeks.

106
 Convergent Evidence for the WAIS-III

Participants were recruited according to the same methods used to select

the standardization sample. The sample had the following composition:
51.6% female and 48.4% male; 79.2% White, 11.5% African American, 6.8%
Hispanic, and 2.5% of other racial/ethnic origin.
Table 4.18 presents the correlation coefficients, means, and standard devia-
tions of the sample's performance on the two tests. The correlation coeffi-
cients were calculated in a two-step process to account for differential
practice effects. In the first step, the coefficients were calculated separately

Table 4.18. Correlations Between the WAIS-R and the WAIS-III

WAIS-R WAIS-III
Subtest/Scale/Index Mean? SD Mean? SD 2"
Vocabulary 10.8 2.8 10.2 2.8 .90
Similarities 11.3 2.7 10.4 3.0 19
Arithmetic 10.1 Aa 10.4 3.0 .80
Digit Span 10.4 3.1 10.3 3.3 82
Information 10.5 2.8 10.5 3.0 83
Comprehension 11.0 2.9 10.5 2.9 76
Letter-Number Sequencing — — — = —
Picture Completion 11.1 2.6 10.7 3.0 50
Digit Symbol—Coding 11.8 3.0 10.6 3,1 TT.
Block Design 11.4 2.9 10.7 3.0 J7
Matrix Reasoning — — 10.3 2.8 —
Picture Arrangement 11.1 2.8 10.5 3.2 .63
Symbol Search _ — 10.1 3.0 —
Object Assembly 11.3 3.1 10.4 3.0 .69
VIO 103.4 14.5 102.2 15.1 94
PIQ 108.3 14.4 103.5 15.4 86
FSIO 105.8 14.3 102.9 15.2 .93
VCl — == 101.9 14.4 —
POI — a 102.9 14.8 _
WMI — = = = ms
PSI — — 101.7 15.0 =
r
err

Note. N = 192. Correlations were computed separately for each order of administration in a counter-
balanced design and corrected for the variability of the WAIS-III standardization sample (Guilford &
Fruchter, 1978).
orders.
4 The values in the Mean columns are the average of the means of the two administration
>The weighted average was obtained with Fisher's z transformati on.

107
Evidence Base for Validity of the WAIS-III and the WMS-III

for each order of administration in the counterbalanced design and correct-

ed for the variability of the WAIS-III standardization sample (Guilford &
Fruchter, 1978). In the second step, the average coefficients of the correla-
tions (of the two administration orders) were calculated with Fisher's z trans-
formation procedure.
The correlation coefficients for the sample are .94, .86, and .93 for the VIQ,
PIQ, and FSIQ scores, respectively. The magnitude of these correlations
suggests that the WAIS-III measures essentially the same constructs as does
the WAIS-R.
A comparison of the mean IQ scores shows that the WAIS-III FSIQ score is
2.9 points less than the WAIS-R FSIQ score and that the WAIS-III VIQ and
PIQ scores are 1.2 points and 4.8 points less than the corresponding WAIS-R
scores. These differences between the WAIS-III and WAIS-R scores are
expected and similar to those found between previous revisions of the
Wechsler scales. Such differences are expected to occur according to the
work by Matarazzo (1972) and Flynn (1984, 1987), which suggests that when
an examinee’s performance is referenced to outdated norms rather than to
current ones, the IQ score will be inflated.
Table 4.19 provides the ranges of expected WAIS-III IQ scores for selected
WAIS-R IQ scores. The ranges of the expected WAIS-III IQ scores associated
with particular WAIS-R scores are relatively narrow near the middle of the IQ
score distribution (i.e., 100) and wider at the upper and lower score levels. In
addition, WAIS-III and WAIS-R PIQ scores can be expected to differ more
than VIQ scores do. The expected score ranges reported in Table 4.19 reflect
95% confidence intervals. (Note that some of the WAIS-III IQ scores did not
cover the selected WAIS-R score, because the WAIS-III is relatively harder
than its predecessor.)

Table 4.19. Expected WAIS-III IQ Scores for Selected WAIS-R IQ Scores

WAIS-Ill 1Q Score Range
WAIS-R IQ Score VIQ PIQ FSIQ
55 50-55 47-54 49-54
70 66-70 62-68 65-69
85 82-85 78-82 81-83
100 98-99 94-96 96-98
115 113-115 109-112 ae
130 128-131 123-127 126-129
145 142-147 137-143 140-145
Ss
e

Note. Ranges are 95% confidence intervals based on linear equating (Angoff, 1984, Design
II.B) of
data for 192 adults administered both tests in counterbalanced order.

108
 Convergent Evidence for the WAIS-III

Correlations With the WISC-III

The WISC-III and WAIS-III were administered in counterbalanced order to a
sample of 184 16-year-olds, recruited according to the same methods used
to select the standardization sample. The interval between testings ranged
from 2 to 12 weeks, with a median of4.6 weeks. The sample had the follow-
ing composition: 47.8% female and 52.2% male; 77.7% White, 12.5% African
American, 6.0% Hispanic, and 3.8% of other racial/ethnic origin.

Table 4.20. Correlations Between the WISC-III and the WAIS-III

WISC-III WAIS-ill
Subtest/Scale/Index Mean* SD Mean® SD 144°
Vocabulary 10.0 2.6 10.3 3:07.83
Similarities ea 3.4 10.9 SS} {ahs}
Arithmetic 10.4 33 11.0 3.9'/ 1:76
Digit Span 10.4 3.4 HOES 2.8 eso
Information 10.3 2.9 10.6 ey £810)
Comprehension 10.5 3 10.3 3:0 OO
Letter-Number Sequencing — — 10.0 34 —
Picture Completion ats 2.9 10.6 3.0 .45
Digit Symbol—Coding 10.9 3.6 10.8 eo elt
Block Design 10.4 3.4 11.0 Shi A330)
Matrix Reasoning — _ 10.7 2.6. =
Picture Arrangement 10.2 2.9 10.7 eu) eh
Symbol Search 11.3 3.2 10.6 Aare
Object Assembly 10.4 3:6 10.7 2 Taare oll
Mazes 10.5 3.3 = =
VIQ 103.0 15.2 103.5 15.6 .88
PIQ 104.5 pare 104.9 14:27 .78
FSIQ 103.9 tore 104.6 cil yl Nemes 33}
VCl 103.0 14.8 103.6 16.2 .87
POI 104.0 14.7 104.4 14.7 .74
WMI° 102.8 16.2 101.1 16.2 .80
106.4 15.4 103.7 14.4 .79
PSI
ED
ion in a counter-
Note. N = 184. Correlations were computed separately for each order of administrat
for the variability of the WAIS-III standardiz ation sample (Guilford &
balanced design and corrected
Fruchter, 1978).
the two administration orders.
@ The values in the Mean columns are the average of the means of
ation.
>The weighted average was obtained with Fisher's z transform
° For this variable, N = 44

109
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.20 presents the correlation coefficients, means, and standard devia-
tions of the subtest scaled scores, IQ scores, and Index scores on the two
tests. The correlation coefficients were calculated in a two-step process to
account for differential practice effects. In the first step, the coefficients were
calculated separately for each order of administration in the counterbal-
anced design and corrected for the variability of the WAIS—III standardiza-
tion sample (Guilford & Fruchter, 1978). In the second step, the average
coefficients of the correlations (of the two administration orders) were calcu-
lated with Fisher’s z transformation procedure.
The correlation coefficients between the WAIS-III and WISC-III IQ scores are
very high and statistically significant: .88, .78, and .88 for the VIQ, PIQ, and
FSIQ scores, respectively. Moreover, the magnitude of these correlations is
high enough to indicate that the two instruments are measuring the same,
or very similar, constructs. These results are relatively higher than those
found between the WISC-R and the WAIS-R (Wechsler, 1981). Correlations
between the WISC-III and WAIS-III Index scores are similar, with coefficients
of .87, .74, .80, and .79 for the VCI, POI, WMI, and PSI scores, respectively. As
shown in Table 4.20, the mean IQ and Index scores of the WAIS-III are nearly
equivalent to the corresponding mean WISC-III IQ and Index scores.
Table 4.21 presents the ranges of expected WAIS-III IQ scores for selected
WISC-III IQ scores. The ranges of expected WAIS-III scores associated with
particular WISC-III scores are relatively narrow near the middle of the IQ
distribution (i.e., 100) and wider at the upper and lower score levels. This
pattern occurs because the error variance of equating increases as the scores
deviate from the mean. The expected score ranges reported in Table 4.21
reflect 95% confidence intervals.

Table 4.21. Expected WAIS-III IQ Scores for Selected WISC-III IQ Scores

WAIS-lIl 1Q Score Range
WISC-III 1Q Score vIQ PIQ FSIQ
55 50-58 56-64 53-60
70 67-72 TATE 69-74
85 83-87 85-89 85-88
100 99-102 100-102. 100-102
115 115-117 Gh 113116 eed 4—117.
130 129-134 126-131: 128-132
145 144-150 188-145 142-148
—
Ss
SSSSSSSSSssssuvSSSssSSSsSsn

Note. Ranges are 95% confidence intervals based on linear equating (Angoff, 1984, Design
II.B) of
data for 184 16-year-olds administered both tests in counterbalanced order.

110
 Convergent Evidence for the WAIS-III

Correlations With the

Standard Progressive Matrices
The Standard Progressive Matrices (SPM; Raven, 1976) and the WAIS-III were
administered to a sample of 26 adults aged 16-45 (M = 28.6, SD = 10.8). The
sample had the following composition: 53.8% female and 46.2% male; 96.2%
White and 3.8% of other racial/ethnic origin.
Table 4.22 reports the correlation coefficients, means, and standard devia-
tions between the WAIS-III IQ and Index scores and the total score on the
SPM. The correlation coefficients were corrected for the variability of the
standardization sample. Because the Letter-Number Sequencing subtest
was not administered, the WMI score could not be calculated; therefore, data
are reported for the Arithmetic and Digit Span subtests. Also, because the
WAIS-III Matrix Reasoning subtest and the SPM are similar tasks, the corre-
lation between the subtest and the SPM is reported separately.

Table 4.22. Correlations Between the SPM and the WAIS-III

SPM WAIS-III
Raw Total Score Mean SD

WAIS-III
VIQ .49 112.8 13.8
PIQ 79 ib er 9.2
FSIQ .64 liso Pe

VCl 55 111.4 12.5

POI .65 114.1 12.4
WMI — — —
PSI 25 103.4 11.0

Arithmetic 32 11.9 4
Digit Span 13 14-2 3.0
Matrix Reasoning 81 12.6 2:5

SPM
50.6 Mean
SD 6.1
i
standardization sample
Note. N = 26. All correlations were corrected for the variability of the WAIS-III
(Guilford & Fruchter, 1978).

111
Evidence Base for Validity of the WAIS-III and the WMS-III

The correlation coefficients between the WAIS-III VIQ, PIQ, FSIQ, VCI, and
POI scores and the SPM are statistically significant and range from .49 to .79.
As expected, the correlations with the PIQ score (.79) and the POI score (.65)
are the highest. The correlation with the PSI score is low (.25) and is also
expected because the SPM is an untimed, nonverbal reasoning task. The
SPM apparently is not highly related to working memory, as indicated by its
low correlations with the Arithmetic (.32) and Digit Span (.13) subtests.
The results are very consistent with the previous findings of significant cor-
relations (from the .50s to .70s) between the SPM and the predecessors of
the WAIS-III (Burke, 1985; Burke & Bingham, 1969; Desai, 1955; Hall, 1957; B.
Levine & Iscoe, 1954; McLaurin & Farrar, 1973; C. G. Watson & Klett, 1974). In
addition, Matrix Reasoning has the highest correlations with the SPM (.81)
than do the other WAIS-III subtests. This high correlation provides support
for the validity of this new subtest of the WAIS-III.

Correlations With the Stanford-Binet

Intelligence Scale—Fourth Edition
The Stanford-Binet Intelligence Scale—Fourth Edition (SB-IV; R. L. Thorndike
et al., 1986) and the WAIS-III were administered to a sample of 26 adults. The
sample was the same one used for the correlation studies with the SPM
(Raven, 1976) previously described.
Table 4.23 reports the corrected correlation coefficients, means, and standard
deviations between the WAIS-III IQ and Index scores and the SB-IV Standard
Area Scores (SAS) and overall composite score. The correlation coefficients
were corrected for the variability of the WAIS-III standardization sample.
Additionally, because the Letter-Number Sequencing subtest was not admin-
istered to this sample, the WMI score could not be calculated. Data for the
individual Arithmetic and Digit Span subtests are reported separately.
The correlation between the WAIS-IIi FSIQ score and the global SB-IV com-
posite score is .88, a result consistent with studies testing the relationship
between the Wechsler scales and the SB-IV (R. L. Thorndike et al., 1986). The
result also suggests that the WAIS-III has a strong relationship with the
SB-IV, as did the WAIS-R.

The highest correlations occur between the WAIS-III PIQ score and the
SB-IV Standard Area Scores, generally ranging in the .80s. The exception is
the Short-Term Memory SAS, which has much lower correlations with the
WAIS-III IQ scores (range .44—.50). The correlations with the Arithmetic and
Digit Span subtests (.34 and .52, respectively) are in a similar range. These
data indicate that the WMI of the WAIS-III and the Short-Term Memory Area

112
 Convergent Evidence for the WAIS-III

of the SB-IV do not share an appreciable amount of variance. Another differ-

ence between the WAIS-III and the SB-IV occurs with the measures of pro-
cessing speed. The correlations between the WAIS-III PSI score and the SB-IV
scores are very low, ranging from -.01 to .32. These results are expected
because the SB-IV does not include a separate measure of processing speed.

Table 4.23. Correlations Between the SB-IV and the WAIS-III

SB-IV
Verbal Visual Quantitative Short-Term sBiv WAIS-Ill
Reason. Reason. Reason. Memory Composite Mean SD

WAIS-—Il
VIQ re 53 69 44 78 112.8 13.8
PIQ 81 82 83 48 89 dite Ore
FSIQ .79 .69 .80 50 .88 diese ee

VCl .87 Oy 78 41 85 Tt 225

POI 76 78 .80 .36 86 114.1 12.4
WMI =. _ — _ — _ _
PSI ~.19 32 =) .06 .07 103.4 . 11.0

Arithmetic 18 48 54 .34 51 A129 2a

Digit Span .26 .30 .38 sy 48 A250 5310

SB-IV
Mean 116.6 110.1 115.5 110.7 114.8
SD 13.6 12.0 16.7 14.3 12.1
EEE
standardization sample
Note. N = 26. All correlations were corrected for the variability of the WAIS-III
(Guilford & Fruchter, 1978).

Summary
WAIS-III
As expected, there was a high degree of correlation between the
pondin g scores on the WISC-I II and WAIS-R .
IQ scores and the corres
trated a high degree of correla tion with the
The WAIS-III also demons
II demonstrated a
SB-IV (R. L. Thorndike et al., 1986). The PIQ of the WAIS-I
evel scores of the
strong association with the SPM (Raven, 1976). The index-l
and had a lower
WAIS-III correlated highly with the WISC-III Index scores
This finding suggests that
correlation with scores on other intelligence tests.
functioning not found in
these indexes add unique measures of intellectual
between the WAIS-III
the SB-IV or SPM. Significant mean score differences
II and the WISC-III
and WAIS-R were found, but mean scores on the WAIS-I
two tests yield similar esti-
were very similar, a result suggesting that these
mates of intellectual ability.
113
Evidence Base for Validity of the WAIS-III and the WMS-III

Convergent Evidence for the WMs-Iil

as a Measure of Memory Functioning
The WMS-III was designed for the assessment of auditory and visual declar-
ative memory abilities and auditory and visual working memory abilities in
adults and older adolescents. In part, the development of the WMS-III was
based on its predecessors, the WMS and WMS-R. Many of the procedures
from these previous versions were retained but updated and expanded to
reflect the increased knowledge regarding human memory functioning.
Additional procedures were developed in response to criticisms of tasks on
the WMS and WMS-R and were based on results of clinical studies of novel
memory tasks. The WMS-III was developed to measure memory functioning
and, therefore, it was expected to have a moderate to high degree of correla-
tion with other memory tests.

Correlations With the WMS-R

A sample of 207 adults were administered the WMS-R and the WMS~III in
counterbalanced order. The mean age was 44.7 years (SD = 20.6). The inter-
val between the two testing sessions ranged from 2 to 12 weeks, with a medi-
an interval of 32 days. Participants were recruited according to the same
methods used to select the standardization sample. The sample had the
following composition: 52.2% female and 47.8% male; 76.9% White, 10.1%
African American, 10.1% Hispanic, and 2.9% of other racial/ethnic origin.
By level of education, the sample had the following representation: 3.4%,
<8 years; 4.3%, 9-11 years; 31.9%, 12 years; 25.6%, 13-15 years; and 34.8%,
216 years.

Table 4.24 presents the correlation coefficients, means, and standard devia-
tions of the sample’s performance on the two tests. The correlation coeffi-
cients were calculated in a two-step process identical to that described
previously in this chapter for the WAIS-III—WAIS-R correlation study. As
reviewed in the WMS-III Administration and Scoring Manual, a number of
substantial changes were implemented in the WMS-III. All of the WMS-III
subtests, for example, that contribute to the visual memory indexes are com-
pletely different from those in the WMS-R. Because of the substantial num-
ber of changes from the WMS-R to the WMS-III, especially in the visually
presented memory subtests, it was anticipated that the WMS-III auditorily
presented indexes would exhibit higher correlations with the corresponding
WMS-R indexes than those between the WMS-R and WMS-III visual indexes.

114
 Convergent Evidence for the WMS-III

The correlation coefficients between the WMS-III Auditory Immediate Index

and the WMS-R Verbal Memory Index, the WMS-III Visual Immediate Index
and the WMS-R Visual Memory Index, and the WMS-III Immediate Memory
Index and the WMS-R General Memory Index are .72, .36, and .62, respec-
tively. As expected, summary measures based on visually presented material
show lower correlations than those based on auditorily presented material.
The WMS-III Working Memory Index correlates highest with the WMS-R
Attention/Concentration Index (.64) and relatively lower with the other
WMS-R memory indexes, with correlations ranging from .34 (Verbal
Memory and Visual Memory indexes) to .38 (Delayed Recall Index).

Table 4.24. Correlations Between the WMS-R and the WMS-III

WMS-R Indexes
Verbal Visual General Attention/ Delayed WMS-IIl
Memory Memory Memory Concentration Recall Mean* SD

WMS-lil
Auditory Immediate v2 53 73 .40 .68 OS Samson
Visual Immediate 33 .36 .36 AS 40 103.4 15.4
Immediate Memory .60 15 62 31 62 104.1 15.9
Auditory Delayed .68 49 .69 .33 .67 104.3 15.1
Visual Delayed .36 42 .39 ae Al 103.8 15.7
Aud Rec Delayed 56 .38 5) Ah fit 104.6 16.1
General Memory .65 54 .67 .30 65 105.1 15:8
Working Memory .34 .34 36 64 .38 10353" 1497,

WMS-R
Mean®* 102.2 104.3 103.4 100.5 104.8
SD 15.8 ieee 16.4 14.3 15.8

a counter-
Note. N = 207. Correlations were computed separately for each order of administration in
of the WMS-III standardizati on sample (Guilford &
balanced design and corrected for the variability
Fruchter, 1978).

@ The Mean values are the average of the means of the two administration orders.

Correlations With the Children’s Memory Scale

The Children’s Memory Scale (CMS; M. Cohen, 1997) is an individually
ents aged
administered test of memory functioning for children and adolesc
age ranges overlap at age
5-16 years. The CMS and the WMS-III normative
tered the CMS and the
16. A sample of 86 adolescents aged 16 were adminis
investig ate the interrel a-
WMS-III in a counterbalanced design in order to
the two testing
tionships of these two instruments. The interval between
interval of 30 days.
sessions ranged from 2 to 12 weeks, with a median

115
Evidence Base for Validity of the WAIS-III and the WMS-III

Participants were recruited according to the same methods used to select

the standardization sample. The sample had the following composition:
46.5% female and 53.5% male; 75.6% White, 9.3% African American, 11.6%
Hispanic, and 3.5% of other racial/ethnic origin. By parent education level,
the sample had the following representation: 3.5%, <8 years; 9.3%, 9-11
years; 27.9%, 12 years; 27.9% 13-15 years; and 31.4%, 216 years.
It was anticipated that WMS-III auditory indexes would correlate most high-
ly with the CMS auditory (i.e., verbal) indexes. Similarly, the WMS-III visual
indexes were expected to correlate most highly with the CMS visual indexes.
Finally, the WMS-III Working Memory Index was predicted to correlate most
highly with the CMS Attention/Concentration Index. The correlation coeffi-
cients were calculated in a two-step process identical to that described pre-
viously in this chapter for the WAIS-III—WAIS-R correlation study.
Table 4.25 presents the correlation coefficients, means, and standard devia-
tions of the sample's performance on the two tests. The correlation coeffi-
cients between the WMS-III Auditory Immediate and the CMS Verbal
Immediate indexes, the WMS-III Visual Immediate and CMS Visual
Immediate indexes, the WMS-III Auditory Delayed and the CMS Verbal
Delayed indexes, and the WMS-III Visual Delayed and the CMS Visual
Delayed indexes are .74, .55, .65, and .26, respectively. The auditory indexes
of the WMS-III correlate highest with the corresponding CMS indexes. The
WMS-III Visual Immediate Index correlates highest with the CMS Visual
Immediate Index; however, the WMS-III Visual Delayed Index correlates
highest with the CMS Verbal Delayed Index and next highest with the CMS
Visual Delayed Index. This unexpected finding may be due to differences in
the content of the WMS-III and CMS visual subtests. Finally, as expected, the
WMS-III Working Memory Index correlates highest with the CMS Attention/
Concentration Index (.68) and relatively lower with the other CMS indexes,
with correlations ranging from .21 (Verbal Recognition Delayed Index) to .48
(Learning Index). These observed patterns and magnitudes of relationships
between the CMS and the WMS-III generally provide evidence of convergent
and divergent validity and support the notion that the CMS and the WMS-III
are measuring similar constructs.

116
 a1qR] “G2"y sUOTRII1I0D
Usamjag
ey} SD pur 2) ITI-SINM
SIND S@Xxepu|
jequan jensi, Jequa, jensiA JeqJa, jesaues
=‘ Juojuayy 1IF-SWM
~~
9}e|powu| a}eIpewiw| Huluiee7 poaAejag pahejag jeq2eyH Aioway [Link]
uke as
II-SIWMA
Aioypny ayelpewu| pl Se g9° OL Le es cL Lv S62 9°61
jensi/\ 9}e|pewWww| Or coker vr by Ly’ ge ciey og Vecok GRE
eyeipewiuy|Mioweyy OL 0s 99° OL Or’ 9S gL by’ 986 8Zt
Aioyipny pefejeq €9° ce" 8S" g9° vz 8y 79° Ly LY¥6 Get
jensi, peejaq Se ov Ov Lv’ 9c 9c" 9" Le €0lc vl6
pny 98y j8q 9g 97° vy 9S" eb vy cS 8y S66 89l
|e1eues Kowa €9° er’ 6S" 99° 92" Ac 29° ey" 6°86 OLL
Bupoi Mowe 9¢° Ge 8v vy 0 Lo vy 89° v'86 Lvl

SIND
ples 9°86 O'LOL 2°86 8°86 66 2°86 8°66 6°26
gs 912 OSL Zt8 cLZl Let G9l 68l vl 8
y

‘@JON=N "98 ||¥ SUOIE|A1I09

848M payndwod Ajayesedas
40} YORS JaPsO
JO UOIEJ}SIUIWPe
U! e peoue|eqiej]UuNOd
ubisep
pue pejoe09
10} 84}
AyIqeueA
JO OU} ||I-—SWIM uoeZIpsepUeys
ajdwes PUOJJIND)
Y JOIYON‘(BZEL
, OU! UBAY\ SeNjeA
Ge OU) eBEJEAe
JO Oy} SUA
JO BU} OM} UOIE!SIUIWIPE
‘Si@PsO
ey] pejyHiom eHesone
sem Paule}go
YIM SJBYSI4
Z "UONEUUOJSUBI]
CJ onvergent Evidence for the WMS-III

117
Evidence Base for Validity of the WAIS-III and the WMS-III

Evidence Based on
Test-Criterion Relationships

Assessment of Psychoeducational and

Developmental Disorders
The concurrent validity studies that have been presented demonstrate the
validity of the WAIS-III and WMS-III as measures of intelligence and mem-
ory, respectively. In clinical settings, the measurement of these constructs
occurs frequently as part of the assessment of developmental problems,
especially those associated with academic performance, in adolescents and
young adults. When a diagnosis of specific developmental disorders such as
mental retardation and learning disabilities is made, an assessment of intel-
lectual functioning assessment must be obtained (American Psychiatric
Association, 1994). A clinically valid test of intelligence should sensitive to
these developmental issues. Memory assessment has not been determined
to be diagnostic of specific developmental disorders; however, memory test-
ing enables the clinician to determine if specific cognitive limitations are
contributing to the individual’s difficulties in learning or adaptation. More-
over, in some cases, memory may be a domain of cognitive functioning that
is the focus of remediation efforts. Evidence of the validity of the WAIS-III
and WMS-III for such applications was obtained from clinical studies of
adolescents and adults with developmental disorders and through concur-
rent assessment of academic skilis correlated with intelligence and memory.
For the following studies involving individuals with psychoeducational and
developmental disorders, the demographic data and performance data for
the WAIS—III are presented in Tables 4.26 and 4.27, respectively. The demo-
graphic data and performance data for the WMS-III are presented in Tables
4.28 and 4.29, respectively.

Mental Retardation
According to the definitions of the DSM-IV, the American Association on
Mental Retardation (1992), and the Developmentally Disabled Assistance
and Bill of Rights Act of 1975, mental retardation is a developmental disorder
that manifests before age 18. Individuals with mental retardation exhibit sig-
nificantly subaverage intellectual functioning that exists concurrently with
related limitations in 2 or more of the following 10 adaptive skill areas: com-
munication, self-care, home living, social skills, community use, self-direc-
tion, health and safety, functional academics, leisure, and work. According to

118
 Evidence Based on Test-Criterion Relationships

Table 4.26. Demographic Data of Samples With Psychoeducational

and Developmental Disorders for WAIS-III Studies
Mental Mental Deafness/
Retardation Retardation Hearing
(Mild) (Moderate) ADHD LD Math LD Reading Deficiencies
N 46 62 30 22 24 30
Age
Mean 30.8 32.8 19.8 18.0 17.5 36.5
SD tee 12.9 2.4 2.0 nye ist
Sex
Female 54.3 41.0 22.6 31.8 37.5 63.3
Male 45.7 59.0 77.4 68.2 62.5 36.7
Race/Ethnicity
White 76.1 86.9 93.5 Tie 95.8 96.7
African American 3.3 6.5
Hispanic 21.7 9.8 18.2 4.2 aie
Other 2.2 45
Education
<8 50.0 90.2 3.2 8.3 6.7
9-11 17.4 1.6 32.3 63.6 58.4 aie
12 32.6 8.2 29.0 18.2 12.5 43.3
13-15 29.0 9.4 20.8 40.0
216 6.5 9.1 6.7

Note. Except for sample size (N) and age, data are reported as percentages.

Table 4.27. WAIS-III Performance of Samples With

Psychoeducational and Developmental Disorders
Mental Mental
Retardation Retardation
(Mild) (Moderate) ADHD LD Math __LD Reading
Scale/Index Mean SD Mean SD Mean SD Mean SD Mean SD

VIQ 60.1 5.0 B47 47 104.2 12.4 984 99 96.7 11.4

PIQ 64.0 5.8 55.3° 4.4 100.9 12.8 99.9 10.6 102.1 10.6
FSIQ 58.3 48 50.9° 4.1 103.0 11.8 99.2 82 99.0 10.9
VCl 63.4 6.3 56.8 6.0 1054 123 1020 11.6 97.9 14.0
POI 66.8 5.6 58.9 5:4°6100.9 13.7 102-3eA2 5 102:2) 11,7
WMI ey ee ae gor 136° soa wots 11.4
PSI 63.3 57.8° 3.8
4.0 93.4 13.5 G5: 2iaalile7e 95:654 10:1
N 46 62 30 22 24
pee ee ee ee ————EEE————eee ee
® Sample size for these three variables was 61.

Sample size for this variable was 20.

° Sample size for this variable was 9.
% Sample size for this variable was 18.

119
Evidence Base for Validity of the WAIS-III and the WMS-III

Table 4.28. Demographic Data of Samples With Psychoeducational

and Developmental Disorders for WMS-III Studies
ADHD LD Reading
N 21 18
Age
Mean 19.2 18.2
SD 1.9 2.1
Sex
Female 28.6 33.3
Male 71.4 66.7
Race/Ethnicity
White AS 94.4
African American 9.5
Hispanic 5.6
Other
Unknown 19.0
Education
<8 4.8
9-11 38.1 AA
12 23.8 44.4
13-15 3019 16.7
216 27.8

Note. Except for sample size (N) and age, data are reported as percentages.

Table 4.29. WMS-III Performance of Samples with

Psychoeducational and Developmental Disorders
ADHD LD Reading
Mean SD Mean SD

Primary Indexes
Auditory Immediate 94.9 14.8 98.0 14.2
Visual Immediate 92.6 135 97.4 11.6
Immediate Memory 92.5 13.6 Va 13.3
Auditory Delayed 96.7 15.2 92.7 16.8
Visual Delayed 94.6 10.7 98.1 15.4
Auditory Recognition Delayed 97.1 19.9 93.3 17.2
General Memory 95.1 14.3 93.7 16.7
Working Memory 94.1 14.0 91.7 9.6
Auditory Process Composites Median %ile %ile Range Median %ile %ile Range
Single-Trial Learning 36% 1%-87% 40% 8%-92%
Learning Slope 60% 6%-96% 44% 38%-98%
Retention 55% 1%-91% 22% 2%-91%
Retrieval 65% 1%-99% 42% 5%-97%
N 21 18
—_—_—_—_—_——::0—0g
 Evidence Based on Test-Criterion Relationships

the normal distribution of IQ scores, about 2.3% of adults obtain scores 2

SDs below the mean (100). However, the prevalence of mental retardation
varies from study to study, ranging from 2.5% to 3.0% of the general popula-
tion because the diagnosis of mental retardation must take into account
both intellectual ability and adaptive functioning (Harrison, 1990; S. A.
Richardson & Koller, 1985).
Many studies have been conducted to characterize the performance of indi-
viduals with mental retardation on the previous versions of the Wechsler
intelligence scales. One method is based on subtest-score profiles. Indi-
viduals with mental retardation often exhibit relatively flat score profiles,
with the lowest scores obtained on Arithmetic, Vocabulary, and Coding.
When Bannatyne'’s recategorization method was used, individuals with
mental retardation tended to show weaknesses on the subtests related to
acquired knowledge (Kaufman & Van Hagen, 1977; Naglieri, 1980; Rubin,
Goldman, & Rosenfeld, 1985; Rugel, 1974; Silverstein, 1968; Simon &
Clopton, 1984). For the WAIS-III study, it was expected that participants with
mental retardation would have flat subtest-score profiles (on all except the
PSI subtests), with scores from 2 SDs to 3 SDs below average according to
the severity of the disorder. Scores on the PSI subtests were expected to be
slightly higher, because previous research has reported that children with
mental retardation obtained relatively higher scores on the PSI than on the
other indexes (Wechsler, 1991).
The WAIS-III was administered to 108 adults diagnosed as mentally retarded
(see Table 4.26 for demographic data). These examinees did not take the
WMS-III, nor were they administered the WAIS-III Letter-Number Sequenc-
ing subtest. Without this subtest, the WMI score cannot be computed. See
Appendix F for the inclusion criteria for participation in this study.
Table 4.27 presents the mean scores and standard deviations for the
WAIS-III IQ scales and indexes for the participants with mental retardation.
The mean FSIQ score of the participants with mild mental retardation is
58.3, indicating that this group has global impairment. As expected, the
impairment is equally distributed across all of the domains of cognitive
functioning. The mean VIQ score (60.1) and mean PIQ score (64.0) are
more than 2 SDs lower than those obtained by the general population.
Furthermore, as the severity of the mental retardation increases, the deficits
in cognitive functioning increase, as reflected in the WAIS-III scores. The
mean VIQ, PIQ, and FSIQ scores of the participants with moderate mental
retardation are 54.7, 55.3, and 50.9, respectively. As expected, the variability
in the performance of each of these clinical groups is very small. The stan-
dard deviations range from 4.1 to 5.8, which are much smaller than those
with
found in the general population (15). These results are very consistent

121
Evidence Base for Validity of the WAIS-III and the WMS-III

the previous reports by L. Atkinson (1992), Craft and Kronenberger (1979),

and Spruill (1991) for adult participants, and by Wechsler (1991) for children.
For the WAIS-III indexes, impairment again appears to be distributed across
the different domains of functioning. The VCI, POI, and PSI scores of the indi-
viduals with mild retardation are 63.4, 66.8, and 63.3, respectively. The same
pattern is observed in the group with moderate mental retardation, whose
mean VCI, POI, and PSI scores are 56.8, 58.9, and 57.8, respectively. The stan-
dard deviations of the Index scores range from 3.8 to 6.3 points, which are
significantly smaller than those of the general population (15). The generally
equal impairment across these three indexes is roughly consistent with the
results reported in the WISC-III manual (Wechsler, 1991). However, children
with mental retardation performed slightly better on the PSI. Further
research is needed to determine if this pattern remains consistent.
On the WAIS-III subtests, both of the groups had the most difficulty on
the Arithmetic subtest. This pattern is consistent with previous report
that individuals with mental retardation have the most difficulty on the
subtests measuring acquired knowledge (Kaufman & Van Hagen, 1977;
Naglieri, 1980; Rubin et al., 1985; Rugel, 1974; Silverstein, 1968; Simon &
Clopton, 1984). Additionally, both groups performed relatively poorly on
the Symbol Search subtest.

Attention-Deficit/Hyperactivity Disorder
Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental
disorder characterized by a wide range of chronic problems with inattention.
Sometimes hyperactive-impulsive behaviors are also present, but current
DSM-IV criteria do not require any hyperactive-impulsive symptoms for
diagnosis of ADHD. Discussions by several researchers (Denckla, 1993, 1996,
Pennington, Bennetto, McAleer, & Roberts, 1996; Pennington & Ozonoff,
1996) suggest that ADHD inattention symptoms overlap considerably with
the neuropsychological concepts of working memory and executive function.
In earlier conceptualizations, ADHD was considered a childhood behavior
disorder whose symptoms dissipated in adolescence. Mounting evidence
from longitudinal studies and other research summarized by Spencer,
Biederman, Wilens, and Faraone (1994), however, have indicated that at least
30%-50% of children diagnosed with ADHD continue to maintain significant
ADHD symptoms through adolescence and into adulthood. Estimates of the
occurrence of ADHD range from 3% to 5% during childhood (American
Psychiatric Association, 1994).
Results of neuroimaging studies involving individuals with attention-deficit
disorder (ADD) indicate mild abnormalities associated with the frontal

122
 Evidence Based on Test-Criterion Relationships

lobes, corpus callosum, basal ganglia, and cerebellum (Giedd, Blumenthal,

Molloy, & Castellanos, 2001). Asymmetry (right greater than left) of dorsolat-
eral prefrontal cortex activation has been associated with symptom severity
and impaired performance on tests of attention (Spalletta et al., 2001).
Results of other studies have indicated asymmetry in the frontal lobes, with
right less than left, for boys (Baving, Laucht, & Schmidt, 1999; Rubia et al.,
1999). Findings of functional asymmetry may relate to the age and sex of the
child and to specific tasks used during imaging.
Results of morphometric studies of ADD have found asymmetry of the head
of the caudate nucleus (right greater than left) and decreased white matter
in the right frontal lobe (Semrud-Clikeman et al., 2000). Caudate asymmetry
was associated with symptom severity (Mataro, Garcia-Sanchez, Junque,
Estevez-Gonzalez, Pujol, 1997; Semrud-Clikeman et al., 2000) and poorer
performance on measures of inhibitory control, whereas deficits in sus-
tained attention were related to decreased white matter in the right hemi-
sphere (Semrud-Clikeman et al., 2000). The association between abnormal
brain morphology and behavioral regulation was observed irrespective of
diagnosis (Semrud-Clikeman et al., 2000). Results of MRI studies demon-
strate that the caudate nucleus decreases in size with normal maturation;
however, youth with ADHD do not demonstrate any change in the size of the
caudate nucleus and have abnormal right-left symmetry of the caudate
(Castellanos et al., 1996; Castellanos et al., 1994).
Traditional IQ scores have not been found useful in discriminating persons
with ADHD from a nonclinical population. Yet, comparisons of subtest score
patterns on tests of intellectual functioning by individuals with and without
ADHD have been found useful in assessing inattention symptoms of ADHD.
On the Wechsler scales (e.g., the WISC-III and the WAIS-R), some subtests
are more sensitive than others to impairments in attention, working memo-
ry, and processing speed, domains that are central to ADHD. Intraindividual
comparisons of performance have demonstrated that individuals with
ADHD tend to perform more poorly on a cluster of subtests that are concen-
tration sensitive than on verbal or spatial subtests. Prifitera and Dersh (1992)
have discussed how such comparisons can be used to increase the conver-
gent validity in diagnostic assessment.
T. E. Brown (1996) reported on a group of 191 adolescents diagnosed with
ADHD whose mean concentration index score (indexes are based on
Bannatyne'’s revised recategorization [1974]) was 18 points lower than their
mean verbal index score. Of this sample, 66% obtained a concentration
index score at least 15 points (1 SD) lower than their verbal or spatial index
score (whichever was higher); 24.6% had a discrepancy of 30 points (2 SDs)
or more between those measures. Only 21.9% and 2%, respectively, of the

123
Evidence Base for Validity of the WAIS-II and the WMS-III

WISC-III standardization sample obtained similar discrepancies (T. E.

Brown, 1996).

Using the subtests of the WAIS-R, Biederman et al. (1993) found that adults
diagnosed with ADHD obtained significantly lower Freedom From
Distractibility Index and FSIQ scores than did those in the control group
without ADHD. T. E. Brown (1996) reported on 142 adults diagnosed with
ADHD who obtained a mean difference of 22 points between their mean
concentration index score (101.1) and mean verbal index score (123.9) on
the WAIS-R. Of these, 86.6% obtained concentration index scores at least 1
SD lower than their verbal or spatial index scores (whichever was higher),
whereas 31% obtained a WAIS-R concentration index score at least 2 SDs
lower (T. E. Brown, 1996).

The score discrepancies found by T. E. Brown (1996) are much greater than
the discrepancies obtained by the WAIS-R standardization sample (20% and
2%, respectively).
In addition, results of neuropsychological studies have indicated that adults
diagnosed with attention deficit disorder (ADD) perform more poorly than
control participants on list-learning tasks (Holdnack, Moberg, Arnold, Gur, &
Gur, 1995; Mungas, 1983). Results of neuropsychological studies with chil-
dren and adolescents with ADD have shown deficits in vigilance (Barkley,
Anastopoulos, Guevremont, & Fletcher, 1991), immediate visual memory
(Kataria, Hall, Wong, & Keys, 1992), verbal memory (Barkley et al., 1991;
Loge, Staton, & Beatty, 1990), and working memory (Holdnack, Ledbetter, &
Cohen, 1996).
For the present study, the WAIS-II was administered to a sample of 30 older
adolescents and adults diagnosed with ADHD according to clinical inter-
views, DSM-IV diagnostic criteria, and the Brown Attention-Deficit Disorder
Scales (T. E. Brown, 1996). Of these participants, 21 also took the WMS-III.
(See Tables 4.26 and 4.28 for the demographic data for the WAIS-III and
WMS-III samples, respectively, and Appendix F for the inclusion and exclu-
sion criteria for participation in these studies.)

WAIS-IIl Results
The mean scores and standard deviations of the ADHD sample on the
WAIS-III scales and indexes are presented in Table 4.27. Compared to the
performance of the WAIS-III standardization sample, the mean intellectual
functioning of the sample with ADHD is in the average range, and the mean
VIQ-PIQ score difference is not significant.
The sample with ADHD did show significant intraindividual differences in
their WAIS-III Index scores. Their mean WMI score is about 8.3 points lower

124
 Evidence Based on Test-Criterion Relationships

than their mean VCI score, and their mean PSI score is about 7.5 points
lower than their mean POI score. About 30% of the sample with ADHD had
WMI scores at least 1 SD lower than their VCI scores, whereas 13% of the
WAIS-III standardization sample obtained such discrepancies. About 26%
of the sample with ADHD had PSI scores at least 1 SD lower than their POI
scores, whereas 14% of the WAIS-III standardization sample had such
discrepancies.
For differences between the higher of the VCI or POI score and the lower of
the WMI or PSI score, 61.3% of the sample obtained differences of 1 SD, and
16.1% obtained differences of 2 SDs or more; only 30.5% and 3.5% of the
WAIS-III standardization sample, respectively, had such differences. These
results are comparable to the findings by T. E. Brown (1996) in his study of
the performance of individuals with ADHD on the WAIS-R.
At the subtest level, the data suggest that the ADHD sample performed rela-
tively more poorly on Digit Symbol—Coding, Digit Span, Symbol Search, and
Letter-Number Sequencing. These results are consistent with previous find-
ings that individuals with ADHD tend to perform relatively poorly on tasks
related to working memory (Holdnack et al., 1996) and on tasks requiring
sustained attention and processing speed (Arcia & Gualtieri, 1994). These
results are very consistent with those found on the WISC-III (see T. E. Brown,
1996, and Schwean, Saklofske, Yackulic, & Quinn, 1992).
Because the pattern of performance demonstrated on the WAIS-III by indi-
viduals with ADHD is somewhat similar to that by individuals with learning
disorders, additional measures, such as the DSM-IV diagnostic criteria for
ADHD, the Brown Attention-Deficit Disorder Scales (T. E. Brown, 1996), and a
continuous performance measure should be used to determine whether the
subtest score patterns are due to ADHD or a learning disorder or both. In
addition, because this pattern of subtest scores was obtained by only a
majority (not all) of the participants with ADHD, the diagnosis of ADHD
should not be made solely on the basis of this score pattern.

WMS-III Results
The performance data for the WMS-III indexes and composites for the sam-
ple with ADHD are presented in Table 4.29. As shown, mean index perfor-
mance ranges from 92.5 (Immediate Memory Index) to 97.1 (Auditory
Recognition Delayed Index), and all mean Index scores are in the average
range. Follow-up analyses were performed on the data from the 21 partici-
pants with ADHD and an additional 10 participants with ADHD who were
administered only the Logical Memory subtest of the WMS-III. The perfor-
mance on the immediate recall condition of Logical Memory was first trans-
formed from a scaled score metric to an index metric (i.e., M = 100, SD = 15)

125
Evidence Base for Validity of the WAIS-III and the WMS-III

and was then compared with their Verbal Comprehension Index score of
the WAIS-III. For these individuals, the mean difference score (Verbal
Comprehension Index score minus Logical Memory score) was 8.5, indicat-
ing that their immediate memory performance was lower than their Verbal
Comprehension Index; the mean difference score in the WMS-III standard-
ization sample was 0.5. Additionally, the frequencies of the difference scores
were evaluated for both the ADHD group and the WMS-III standardization
sample. Of the ADHD group, 29% obtained a difference of at least 1 SD
between these measures (memory lower than IQ), whereas only 15% of the
WMS-III standardization sample had such a difference; 10% of the ADHD
sample versus 2.1% of the standardization sample obtained a difference of
2 SDs between these measures. These percentages of differences are lower
than those obtained by Quinlan and Brown (1997), but do show a clear trend
of lower verbal memory compared to verbal intellectual functioning.

Learning Disabilities
Learning disability is associated with difficulties in acquiring a specific acad-
emic skill despite normal intellectual functioning (American Psychiatric
Association, 1994). It is estimated that about 4% of school-aged children
have a reading disorder and 1% of school-aged children have a mathematics
disorder, whereas pure writing disability is relatively rare. Cognitive and neu-
roimaging research has increased our understanding of the underlying
processes and brain regions associated with developmental reading disor-
ders. Results of research on brain functioning for orthographic tasks suggest
multiple posterior brain systems, one involving the temporo-parietal and
one involving temporo-occipital brain regions (Pugh et al., 2001). These sys-
tems develop at different times, with the dorsal system (temporo-parietal
region) relating to phonological and lexical mapping and orthographic rep-
resentations whereas the late-emerging ventral (temporo-occipital) system
appears to enable rapid word-form identification and recognition (Pugh et
al., 2001). Individuals with dyslexia may recruit other brain regions, particu-
larly the posterior right hemisphere (Pugh et al., 2000) and left frontal region
(Georgiewa et al., 2002; Richards et al., 1999), to perform phonological tasks
and to compensate for dominant temporal-lobe dysfunction. During lexical-
judgment tasks, individuals with dyslexia displayed activation in orbito-
frontal region whereas normally functioning readers activated the middle
frontal gyrus on this task (Corina et al., 2001). Research has also focused on
the role of the cerebellum in learning disabilities and poor automatization of
cognitive tasks (Nicolson, Fawcett, & Dean, 2001).
Morphological studies have focused on the identification of abnormal sym-
metry of the planum temporale (a region important in the processing and
integrating of auditory input) in dyslexia (Frank & Pavlakis, 2001). The results

126
 Evidence Based on Test-Criterion Relationships

of these studies need to be evaluated carefully in light of research demon-

strating that sex and age effects may account for differences observed in
morphometric studies of dyslexia (Schultz et al., 1994).
The majority of research on learning disabilities has been conducted with
children and adolescents. Some of this work was conducted with the
Bannatyne recategorization method (Bannatyne, 1968, 1974). According to
this method, the subtests of the Wechsler scales are classified into four cate-
gories: conceptual, spatial, sequencing, and acquired knowledge. The perfor-
mance by individuals with learning disabilities was more frequently found to
exhibit a pattern of spatial > conceptual > acquired knowledge > sequencing.
Other researchers have proposed an “ACID” profile to characterize the per-
formance by individuals with learning disabilities on the Wechsler intelli-
gence scales (Ackerman, Dykman, & Peters, 1976; Kaufman, 1979; Sandoval,
Sassenrath, & Penaloza, 1988). That is, individuals with learning disabilities
are more likely to show relative weaknesses on the Arithmetic, Coding,
Information, and Digit Span subtests. Studies involving adults with learning
disabilities have replicated these findings (Cordoni, O’Donnell, Ramaniah,
Kurtz, & Rosenshein, 1981; Katz, Goldstein, Rudisin, & Bailey, 1993; Kender,
Greenwood, & Conard, 1985).
With the accumulation of factor analytic studies and the release of the
WISC-III in 1991, researchers found similarities between Bannatyne’s four-
category structure and the ACID profile and the factor structure of the
WISC-III. When the WISC-III Index scores are compared, the profiles
become striking. Individuals with learning disabilities tend to show relative
weaknesses in the domains related to the third and fourth factors (i.e., the
WML and PSI), with scores approximately 0.5-0.67 SDs below average
(Prifitera & Dersh, 1992; Wechsler, 1991).
Results of memory studies have also indicated decreased story recall in
adults with learning disabilities (Worden, 1986), verbal and visual memory
deficits and poorer performance on Digit Span in adolescents with learning
disabilities (Ormrod & Lewis, 1985), and working memory deficits in chil-
dren with learning disabilities (Swanson, 1993; Swanson, Cochran, &
Ewers, 1990).
For the present study, 46 adults diagnosed with learning disabilities
(24 in reading, 22 in math) were administered the WAIS-III. Of this sample,
18 with learning disabilities in reading were also administered the WMS-IIL.
Participants for this study were recruited from several university centers.
See Tables 4.26 and 4.28 for the demographic data for the WAIS—III and
WMS-III samples, respectively, and Appendix F for the inclusion criteria
for participation.

127
Evidence Base for Validity of the WAIS-III and the WMS-III

WAIS-IlI Results
Table 4.27 provides the mean scores and standard deviations for the
WAIS-III IQ scales and indexes for the two groups with learning disabilities
(reading and math). The mean IQ scores are all in the average range; scores
for the group with reading disabilities range from 96.7 to 102.1, and for the
group with math disabilities, from 98.4 to 99.9.
The pattern of scores becomes pronounced at the index level. For the group
with reading disabilities, the Index scores range from 91.3 to 102.2, and for
the group with math disabilities, from 89.4 to 102.3. A striking finding, how-
ever, is that the differences between the VCI and WMI scores are 7 and 13
points for the reading and math groups, respectively. Moreover, the VCI
scores are at least 15 points higher than the WMI scores for 41.7% of the
individuals with reading disabilities, compared to 13% of the WAIS-III stan-
dardization sample. The pattern is similar for the nonverbal Index scores.
The difference between the POI and PSI scores is about 7 points for both the
reading and math groups, and the POI scores are at least 15 points higher
than the PSI scores for 30.4% of these individuals, compared to 14% of the
WAIS-II standardization sample.
At the subtest level, the performance of both groups on the ACID profile was
clearly depressed, with 24% exhibiting a partial ACID profile and 6.5%
exhibiting a full ACID profile, proportions that are greater than those in the
general population. These results are very consistent with findings from pre-
vious research (e.g., Ackerman et al., 1976; Cordoni et al., 1981; Katz et al.,
1993; Kaufman, 1979; Kender et al., 1985; Prifitera & Dersh, 1992; Sandoval et
al., 1988; Wechsler, 1991) except that the rate of the partial ACID profile
found in the current study is a little higher. The results also suggest that the
discrepancies between the VCI and WMI scores and between the POI and
PSI scores may be more powerful than the ACID profile analysis in charac-
terizing learning disabilities.

WMS-Ill Results
Table 4.29 presents the WMS-—III performance statistics for those participants
with reading disabilities. Similarly to the participants with ADHD previously
described, the participants with reading disabilities exhibit average memory
performance on all WMS-III indexes. The Index scores range from 91.7
(Working Memory Index) to 98.1 (Visual Delayed Index). Follow-up f tests
were performed to compare the performance on the Primary Indexes by the
group with reading disabilities and a control group (matched for age, sex,
race/ethnicity, and education). The results of these tests did not reveal any
significant differences between mean scores of the two groups (p < .05) for
any of the WMS-III indexes. However, the differences on the Retention

128
 Evidence Based on Test-Criterion Relationships

Composite was statistically significant (p < .01) and indicated that the group
with reading disabilities performed more poorly than the matched control
group. These results suggest that although the group level of performance is
in the average range, individually, participants with reading disabilities have
a higher forgetting rate for auditorily presented stimuli than do the partici-
pants in the matched control group.

Hearing Deficiencies
Deafness and hardness of hearing affect language acquisition and the devel-
opment of verbal skills. Many studies have reported that individuals with
hearing deficiencies performed at the average or low average level on the
Performance scales of the Wechsler intelligence scales and on other nonver-
bal instruments, such as Raven's SPM (1976). The VIQ scores of the individu-
als with hearing deficiencies were about 1 SD lower than their PIQ scores
(Braden, 1992; Wechsler, 1991). At the subtest level, they usually performed
relatively poorly on the Coding or Digit Symbol subtest (Braden, 1990;
Pickles, 1966; Sullivan & Schulte, 1992; B. U. Watson, Sullivan, Moeller, &
Jensen, 1982).

The Performance scales of the Wechsler intelligence scales are the most pre-
ferred instruments for assessing the intellectual functioning of individuals
with hearing deficiencies (Braden, 1992; E. S. Levine, 1974; Maller & Braden,
1992; McQuaid & Alovisetti, 1981; Trott, 1984). As a means of evaluating the
clinical utility of the WAIS-III with individuals who are have hearing defi-
ciencies, the Standardization Edition of the WAIS-III was translated into
American Sign Language (ASL) by Kostrubala and Braden (1997). The quality
of translation was ensured by a back-translation procedure (Hambleton,
1994). The WAIS-III was first translated into ASL and then translated back
into English by different ASL experts so that the consistency with the English
version could be checked. The translation was modified to correct any
inconsistencies, and several iterations were made until high consistency
was reached.
The ASL translation of the WAIS-III was administered to a sample of 30 indi-
viduals with hearing deficiencies, with the following composition: 63%
female, 37% male; 97% White, and 3% Hispanic. The participants ranged in
age from 18 to 75 years (M = 36.5 years, SD = 17.1). See Appendix F for the
inclusion criteria for participation in this study.
The reliability of the ASL version of the WAIS-III was estimated on the basis
of this sample. The average split-half reliability coefficient for the VIQ scale
was .93, with a range of .84-.97; for the PIQ scale (excluding Digit Symbol—
Coding), .81, with a range of .73-.91; and for the FSIQ scale, .89. All averages
were calculated with Fisher's z transformation. Because the sample was
relatively small, further reliability studies may be necessary. However, the
129
Evidence Base for Validity of the WAIS-III and the WMS-III

obtained reliability coefficients are consistent with those reported for the
general population.
The means and standard deviations of the WAIS-III subtest scaled scores, IQ
scores, and Index scores are reported in Table 4.30. The mean PIQ score
(103.2) is in the average range, the mean VIQ score (82.7) is in the low aver-
age range and significantly lower (about 1.37 SD) than the mean PIQ score.
The standard deviations of the IQ scores are similar to those of the general
population. At the index level, both the POI and PSI scores are in the average
range. The VCI score is in the low average range and significantly lower than
the POI and PSI scores (about 1.14 SD and 1.47 SD, respectively). Although
the mean subtest scaled scores on all of the Performance subtests were in
the average range (9.7-10.9), the performance on Digit Symbol—Coding
was the lowest (9.7), about 0.67 SD below the average on the rest of the
Performance subtests. These results are very similar to those reported
previously (Braden, 1990, 1992; Pickles, 1966; B. U. Watson et al., 1982;
Wechsler, 1991).

Table 4.30. WAIS-III Performance of Individuals With

Deafness/Hearing Deficiencies
Scale/index Mean SD
VIQ 82.7 16.8
PIQ 103.2 13.1
FSIQ 90.8 15.2
VCl 81.6 15.7
POl| 103.6 12.9
WMI = =
PSI 99.8 16.4

N 30

Summary
The results of these clinical studies reveal that the WAIS-III is differentially
sensitive to developmental disorders, as expected. On average, the perfor-
mance of individuals with moderate mental retardation was below that of
those diagnosed with mild mental retardation. The scores for both groups
were in the extremely low range and dissimilar to scores obtained by individ-
uals with learning disabilities and ADHD, who performed in the average
range on measures of intellectual functioning. These results indicate that the
WAIS-III provides a good estimate of intellectual functioning. The WAIS-III
is also sensitive to mild processing weaknesses observed in individuals with
psychoeducational difficulties.

130
 Evidence Based on Test-Criterion Relationships

The WMS-III is not typically used for diagnosing mental retardation and was
not included in the studies of these groups. Performance on the WMS-III by
the samples with ADHD and learning disabilities displayed mild weaknesses,
particularly, in comparison to their intellectual functioning.

Assessment of Neurological Disorders and

Disorders Associated With Dementia
The sensitivity of the WAIS-III and WMS-III to cognitive weaknesses associ-
ated with various neurological disorders was assessed. These studies are very
important in establishing the sensitivity of the WMS-III to the effects of dis-
orders associated with dementia and amnesia. In general, the results of
these studies demonstrate the clinical applicability of the WAIS-III and
WMS-III for assessing the effects of neurological disorders and disorders
associated with dementia. Zhu, Tulsky, Price, and Chen (2001) derived relia-
bility coefficients and standard errors of measurement for selected clinical
groups with neurological disorders.
The demographic characteristics of most of the samples for these special
group studies are presented in Table 4.31. The data are presented as
mean age and percentages of the samples by sex, race/ethnicity, and
education level.
The results of the following studies involving individuals with neurological
disorders, except temporal lobe epilepsy, are presented in Table 4.32. The
table includes the mean scores and standard deviations for the WAIS-III 1Q
scales and indexes and the WMS-III Primary Indexes and the median per-
centile scores and percentile ranges for the WMS-III Auditory Process
Composites. The results of the sample with temporal lobe epilepsy are pre-
sented later in Table 4.33.

Alzheimer’s Disease
Alzheimer’s disease is a chronic, progressive neurological disorder that
causes a gradual loss of cognitive functions and, as the disease progresses,
impairments in social and occupational functioning and eventually death.
Alzheimer’s disease is the most common form of dementia among older
persons (Zec, 1993). Although the causes of Alzheimer’s disease are un-
known, neuropathological findings include general cortical atrophy of the
temporal-parietal and frontal regions of the brain. Other brain regions, such
as the hippocampus and amygdala, may be affected by the presence of neu-
ritic plaques (Zec, 1993). The presence of neurofibrillary plaques and tangles
detected at autopsy have been associated with abnormalities on mental-
status and memory examinations (Fuld, Katzman, Davies, & Terry, 1982).

131
 en sige] Ley
* s1ydeis0uI[g
beg

=
jo ay} Saydures
YIM ‘esisofo1nay] ‘poyelay-[oyooyy
pure =
~ d11ye1TysAsdoinany
S1ap1osiq 8
pej}e|ay-joyooly aye1ysAsdo
gy
jesibojoinon
siapiosigsdnoi5 sjapsosiq siapiosig S
SAOWIOUZIY jesodwia,
aqo7 Asdajid3 ‘g
~aseesig s,uojbuuny s,uosupseg Newnes, aidyinyy ye7 y61y auoIYD $,JJOYBSIOY -
PII aseasig aseasiq
= ulesg Ainfuj sisolajas Awoj9eqo7 Awojeqo7 joyoojy asnqy awospudg eluaiydoziyss Ss
N Ge St OL (£4, Go Gl All 8¢ Ol ov 3
aby 2
UES col LUV Wh G 696 a4 GCE SOE ees O19 é8e S
ds BL ora’ Gaby 6S 96 eZ LG Obc 8Ol Ee) =
X98S a
g/ewe+ CV6 Z99 0°02 v9 08 ey 0°0S gE O'6L S
s/eW ZS (ote, 0°08 9°€9 091 L9¢ 00S 96v 0'00L O18 a
Ayojuyy3/e9ey =
OUUM 9°88 0001 0°08 0001 O92 €'€6 €8S €68 0'00L 619 8
UPON UBOWOWY 98 O'OL 091 Z9 les BE OLE =
ojuedsiy O01 e8 s
JBY1O 6¢ ve >)
uMOUyxUt) 08 9° 8'y =
uoljeonpy ep)
gs 62% L9 he O'0€ 8'p a
bL-6 98 91 €'8 evk 0°02 LL 6 poe
ch ey! Ries 00L Bae: O'cEe L9¢ Z9L Le 0°02 LB
Silecae 6'°c¢ 0'0¢ 001 0v6 0°02 0°09 0'0S 9'8¢ 0'0€ 9'°8¢
Ql< 9°87 0'0r 0:02 9'EL 0'9€ Ger 0'S2 (yA Z9l
umouyun Ocl
ee eeee
‘ajox }deoxg
40} ajdwes
ezis (jy) pure ‘abe eyep aie peyode
se ‘sepejuedied
 81geL "ZE"y III-SIVM
PUR [II-SIWM aauvUrIOJ10q
Jo Sa[dureg
YIM Jes1so[oAnaN
SIap1Osig
SJOWIOUZ|V
eseesig s,uo}Buljuny S,UOSU|4Jed oyyeuinely aidninw
2S!SO19]9S
PII
eseasig
aseesig
ujesg
Aunfuj
urea as ure as uray as ural as ue as
|II-SIVM SOXBPUI/SB|B9S
OIA 2226 L's 6°06 OL 976 €8 968 y7l
Old fis= cL 7.8L a segs Scio “ore gee
OIS4 9°98 eL 0'r8 28 78s Lol 98 60!
IO O66 OZ 86 668 696 CL eS eG FzC
\Od S'r8=: yzl 68 86 ~ cZre Lzi- 1°26 OSL
IWM Bcze cle L\8 or re Cogai hele 8'68 lel
\Sd 962 yrl 69 woz Bis ttt@ we. rok
|II-SINM Aiewiid sexapul
Aoyipny ayelpewu) Z'89=: Ol eel ew Boe Fe acta Cobo Vie. frl=
jensiA ayeIpewwu QOL 60! Sel tke =yrd or= 6rd 6 SiS Srl
ayeipewty|Aoweyy -6'@9= vlk 6'0L Senr, cece GZS= Boel Like 088. 6cr=
Auoyipny peAejag 199 9°6 L€8 ect ee SiteS 968 81S) 82 eb.
jensi, pefejaq $29 Lg rag 8§=6S0L 8608 lel See 8c “cee Sol.
Aioyipny uoquBooey
pafejaq 9°¢9 98 0°88 86 006 yl 966 991 v7 Srl
jeseuary Aoweyy 7°09 68 y9l ASue 2198, roi= 618 S9k 298 gz4t
Supjom Koweyy 7°08 691 L'e8 LLL soce Cee B “ct Cie re ol-
III-SWM Moypny
$S9901d saysodwos) UeIPEW
211% a1% aBuey UBIPEW
311% a1% ebuey ueIpew
aj% a1% ebuey uejpeyaj1% aji% ebuey ueIpay
ayi% 21% abuey

Hulued7 jeu-ajbuis WE %~S-Yot Rbvy-%

vl l
Wl

Lt
WIE

ado|s Buiwes WE %ES-Y%t %G8-%ol

YEE %BE-Yol
%6EWEE
Sbvv

Kvy-%l
%GE-%lL

uol}Uua}8y lb %WEE-%b
WIEN

WE

%S9
%VE-Ybt
%86-%l

WEE-Yolt

JeASUIaY Wve %BL-—%E

SI
=%OS

%0S
WGE-WE%lE-Yl

%SE-%t

YS %OL-Y%ol
WL?e Ol
a4
WveS¢

N ge
Wee%BB-%S

WEE-%E

RvE-Yob+

“IIIESIVM 8} perejsiulupe jou aJam sisosajos ajdiyjnw YM sjuedioied ay) ‘sjuresjsuCS Hul|dwes Jo esnedeg ,
Evidence Based on Test-Criterion Relationships

133
Evidence Base for Validity of the WAIS-III and the WMS-IIi

Alzheimer’s disease is associated with diffuse (Alexander, Prohovnik, Stern, &

Mayeux, 1994; Almkvist, Backman, Basun, & Wahlund, 1993; Brinkman &
Braun, 1984; Gfeller & Rankin, 1991; Goldman, Axelrod, Giordani, Foster, &
Berent, 1992; McCurry, Fitz, & Teri, 1994; J. J. Ryan, Paolo, Oehlert, & Coker,
1991) and progressive (e.g., American Psychiatric Association, 1994; Lezak,
1995) deterioration of cognitive functions. The primary purpose of memory
and intelligence testing of individuals with Alzheimer’s disease is to gauge the
degree of deterioration against the baseline of premorbid cognitive ability.
Impairment in declarative memory for new information appears early in the
course of this disease, and the decline becomes more rapid as the disease
advances (Storandt & Hill, 1989; Zec, 1993). Moreover, memory dysfunction
in older persons without apparent disease has been found to predict later
development of probable Alzheimer’s disease (Bondi et al., 1994; Jacobs et
al., 1995; Masur, Sliwinski, Lipton, Blau, & Crystal, 1994). Diagnosis of
Alzheimer’s disease necessarily requires validation of neurocognitive deficits,
including memory impairment, through neuropsychological procedures
(McKhann et al., 1984).
Increasingly, researchers have focused on studies of individuals who do not
meet diagnostic criteria for dementia but who exhibit some form of cogni-
tive impairment (Peterson et al., 2001). Diagnostic issues and group hetero-
geneity in outcomes hamper research of mild cognitive impairment (Palmer,
Wang, Backman, Winblad, & Fratiglioni, 2002; Ritchie, Artero, & Touchon,
2001). Results of neuropsychological studies indicate higher rates of progres-
sion to Alzheimer’s-type dementia in individuals with mild cognitive deficits
(Morris et al., 2001), and memory impairment was found to predict long-
term outcomes (Tuooko, Frerichs, & Kristjansson, 2001) and to signal the
presence of an underlying disease process (Portin et al., 2001). Imaging
research demonstrates reduced volume of the hippocampus and entorhinal
cortex in mild cognitive impairment (Du et al., 2001). Early identification of
dementia enables clinicians to recommend appropriate medical interven-
tions and aid families in long-term planning.
A sample of 35 individuals clinically diagnosed with probable Alzheimer’s
disease were administered the WAIS-III and the WMS-III. These individuals
were identified in a variety of settings, including a geriatric medical clinic, a
dementia diagnostic clinic, a memory disorders clinic, and nursing homes.
The diagnosis of probable Alzheimer’s disease was made in accordance with
the guidelines established by the National Institute of Neurological and
Communicative Disorders and Stroke (NINCDS) and the Alzheimer’s Disease
and Related Disorders Association (ADRDA) Work Group (McKhann et al.,
1984). See Table 4.31 for the demographic statistics for this sample and
Appendix F for the inclusion and exclusion criteria for participation in
this study.

134
 Evidence Based on Test-Criterion Relationships

As indicated in Table 4.31, the sample had a significantly higher education

level than that of the general population (e.g., 48.6% of the sample had com-
pleted at least 4 years of college, and an additional 22.9% had completed
some post-high-school education).
For the WAIS-III and WMS-III studies, the participants with probable
Alzheimer's disease were expected to exhibit decrements in cognitive func-
tioning on the WAIS—III measures as well as impaired memory functioning
on the WMS~III indexes. For this sample, it was also anticipated that memo-
ry functioning would be more impaired relative to global intellectual func-
tioning. Because education level and intellectual functioning are related, this
group was expected to have average premorbid WAIS-III IQ and Index scores
higher than those for the general population. Therefore, results must be
interpreted in light of these possible effects. Table 4.32 presents the WAIS-III
and WMS~III results of this study.
WAIS-IIl Results
As the data in Table 4.32 show, all of the mean WAIS-III IQ and Index scores
for the sample with probable Alzheimer’s disease are lower than the mean
scores of the general population. Mean IQ scores are 92.2 (VIQ), 81.7 (PIQ),
and 86.6 (FSIQ). As expected, the PIQ score is lower than the mean VIQ score
because the verbal scores tend to be somewhat more resilient and less sensi-
tive to the effects of this neurologic condition. Again, in view of the high
education level of these examinees, even their current mean VIQ level may
still represent a 10-point decrement from their premorbid level.
The Index scores show more differentiation of abilities, ranging from 79.6 for
the PSI to 93.0 for the VCI. As expected, the POI and PSI scores show the
largest decrements of all of the WAIS-III indexes, with scores of 84.8 and
79.6, respectively.

WMS-11! Results
Mean scores on the WMS-III memory indexes range from 60.4 for the
General Memory Index to 80.4 for the Working Memory Index (see Table
4.32). With the exception of the performance on the Working Memory Index,
mean scores on the indexes are clearly impaired (i.e., scores of 70 or below).
The percentages of the sample scoring 70 or below on the Immediate
Memory Index and General Memory Index are 71% and 89%, respectively.
These rates can be contrasted with the percentages of the sample obtaining
scores below 70 on the WAIS-III IQ scales: 9% (VIQ), 17% (PIQ), and 9%
(FSIQ). As expected, the group performance on the memory indexes is much
lower relative to the WAIS-III performance. The WMS-III Auditory Process
memory
Composites provide information for various aspects of learning and

135
Evidence Base for Validity of the WAIS-III and the WMS-III

for auditorily presented stimuli. These composites show that as a group the
participants with Alzheimer’s disease demonstrate borderline to impaired
recall performance after the first presentation of memory stimuli (Single-
Trial Learning); show little, if any, improvement with repeated exposure to
the stimuli (Learning Slope); and, at delayed recall, have marked difficulty
recalling the limited information that was learned in the immediate condi-
tion. Performance on the Retrieval Index does not indicate that these indi-
viduals have greater access to previously learned information by recognition
than by recall. Thus, as expected, the examinees with Alzheimer’s disease
demonstrated inefficient encoding and impaired storage of new information
without prominent retrieval deficits.

Huntington’s Disease
Huntington's disease is a relatively rare, genetically transmitted, neuro-
degenerative disorder that produces a characteristic form of dementia and
eventually results in death (Brandt & Bylsma, 1993; Brandt & Butters, 1986;
Martin & Gusella, 1986). The onset of the disorder is characterized by in-
voluntary movements and cognitive impairments that typically appear
during the third or fourth decade of life (Brandt & Bylsma, 1993). The
neuropathology of Huntington's disease includes loss of cells in the caudate
nucleus. Eventually other basal ganglia structures become involved, and cor-
tical atrophy may occur (Brandt & Bylsma, 1993; Martin & Gusella, 1986).
Most individuals with Huntington's disease experience significant declines in
cognitive, motor, and personality functioning. Diagnosis of Huntington’s dis-
ease at an early age is more common than in the case of Alzheimer’s disease
(Lezak, 1995). Intellectual performance, however, shows a pattern parallel to
that of persons with Alzheimer’s and Parkinson's diseases, namely, relatively
preserved verbal comprehensive abilities coupled with impaired perceptual
organizational abilities (Randolph, Mohr, & Chase, 1993). Attention span is
thought to diminish (Brandt & Butters, 1986) in proportion to disease pro-
gression (Lezak, 1995). Other neuropsychological symptoms include impair-
ments in learning and memory, spatial reasoning, visual-motor skills, and
attention/concentration (Brandt & Butters, 1986).
Results of memory studies of individuals with Huntington’s disease have
indicated that memory deficits occur early in the course of the disease and
become more generalized and severe as the disease progresses (Butters, Sax,
Montgomery, & Tarlow, 1978) and that retrieval, rather than storage, deficits
are predominant in declarative memory procedures (Butters, Wolfe,
Martone, Granholm, & Cermak, 1985). Rates of forgetting may not be as
rapid as those for persons with Alzheimer’s disease but are faster than those
of healthy individuals (Tréster et al., 1993). The issue of impaired versus
spared recognition compared to persons with Alzheimer’s disease has yet to

136
 Evidence Based on Test-Criterion Relationships

be resolved; studies support both hypotheses (Brandt, 1992). Studies of

memory functioning in persons with Huntington's disease based on the
WMS and the WMS-R have provided the following findings: increased intru-
sion errors on Visual Reproduction compared to healthy individuals (Jacobs,
Troster, Butters, Salmon, & Cermak, 1990), impaired performance on Logical
Memory and Visual Reproduction on immediate and delayed trials com-
pared to age-matched individuals (Tréster et al., 1993), lower memory quo-
tient scores for those with moderate to severe symptoms of Huntington's
disease (Butters et al., 1985), and impaired logical memory in persons in the
early and late stages of Huntington's disease but only impaired Visual
Reproduction in persons having the disease for 3 years or longer (Butters
et al., 1978).

A sample of 15 participants diagnosed with Huntington’s disease and select-

ed from various research centers from across the country completed the
WAIS-III and the WMS-III. See Table 4.31 for the demographic data for this
sample and Appendix F for the inclusion and exclusion criteria for participa-
tion in this study.
It was expected that participants with Huntington's disease would exhibit
decrements in overall cognitive functioning, with memory performance
being lower relative to intellectual ability. The overall education level of the
15 participants in this group was higher than that of the general population,
with 40% having 16 or more years of education and another 20% having
13-15 years. It would be expected that this group’s performance on the
WAIS-III would be higher than that of the general population if not for
the presence of Huntington's disease. Performance on the WAIS-III Verbal
subtests (e.g., Vocabulary and Information) was expected to be the least
impaired, whereas performance on the PSI subtests (Digit Symbol—Coding
and Symbol Search) was expected to be the most impaired. Previous re-
search has indicated that visual-spatial processing skills are particularly sus-
ceptible to the effects of Huntington's disease. Therefore, it was hypothesized
that the participants would show less severe decrements on the WMS-III
auditory indexes and that memory deficits would be characterized by poor
encoding and retrieval, with relatively spared storage processes.

WAIS-III Results
The WAIS-III mean scores and standard deviations for the sample with
Huntington's disease are presented in Table 4.32. As the data show, the mean
1Q scores are significantly lower than average, ranging from 78.2 to 90.9. As
expected, performance on the PIQ scale (78.2) is more impaired than perfor-
mance on the VIQ scale (90.9).

137
Evidence Base for Validity of the WAIS-III and the WMS-III

On the WAIS-III indexes, the lowest scores were obtained on the PSI (69.3).
Additionally, deficits on the POI (84.9) and the WMI (81.7) are pronounced.
Also as expected, the mean score on the VCI (98.4) is relatively intact for this
sample (although the performance level may be worse than their premorbid
functioning).
The mean scores obtained by this sample are nearly identical to those found
by Randolph et al. (1993), who found that individuals with Huntington's dis-
ease have significant impairment, relative to the general population, in most
intellectual functions, excluding verbal comprehensive abilities.
WMS-Ill Results
As the data in Table 4.32 show, participants with Huntington's disease ob-
tained mean WMS-III Index scores ranging from 70.9 (Immediate Memory
Index) to 88.0 (Auditory Recognition Delayed Index). The obtained index
scores mildly support the predictions that performance on tasks with visual-
ly presented stimuli would be lower than that on tasks with auditorily pre-
sented stimuli. The group obtained Single-Trial Learning and Learning Slope
composite median percentile scores of 14% and 17%, respectively. The medi-
an Retention Composite percentile score (39%) is significantly higher than
that obtained by the sample with mild Alzheimer’s disease. This pattern of
scores suggests that individuals with Huntington's disease exhibit less severe
storage problems than does the sample with Alzheimer’s disease. Finally, the
Retrieval Composite score (65%) suggests significantly more retrieval diffi-
culty than was noted in the sample with Alzheimer’s disease. That is, individ-
uals with Huntington's disease are disproportionately aided by a recognition
(versus recall) testing format.

Parkinson’s Disease
Parkinson's disease is a disease of involuntary movement characterized by
resting tremors, reduced initiation of voluntary movements, shuffling gait,
plastic rigidity, and impaired posture (Hoehn & Yahr, 1967; for a complete
review see Mahurin, Feher, Nance, Levy, & Pirozzolo, 1993). The average age
of onset is in the fifth and sixth decades of life (Hoehn & Yahr, 1967).
Parkinson's disease is relatively common among individuals aged 60 and
older, occurring in approximately 1% of the population (Schoenberg, 1987).
No single cause has been identified; rather, multiple pathological processes
have been implicated (Mahurin et al., 1993). Neuropathological processes
involve loss of cells in the regions of the substantia nigra and reduced
dopamine production (Mahurin et al., 1993).
Impairment of intellectual functioning is often seen in individuals diagnosed
with Parkinson's disease, although clinical dementia is found less frequently

138
 Evidence Based on Test-Criterion Relationships

(J. A. Cooper, Sagar, Jordan, Harvey, & Sullivan, 1991). Verbal comprehensive
abilities are generally preserved, whereas perceptual organizational abilities
and speed of information processing tend to be relatively impaired
(e.g., Randolph et al., 1993). Randolph et al. found that individuals with
Parkinson's disease could be differentiated from those with Alzheimer’s or
Huntington's disease in terms of absolute levels of intelligence. However,
those with Parkinson's disease obtained WAIS-R profiles similar to those of
the two other groups, namely, preservation of verbal knowledge coupled
with relative impairment on Performance subtests. R. G. Brown and
Marsden (1986) postulated that the perceptual organizational deficits fre-
quently observed in individuals with Parkinson's disease are secondary to
difficulties in set-shifting. J. A. Cooper et al. (1991) reported executive dys-
function in their sample of respondents with Parkinson's disease as suggest-
ed by relatively poor sequencing on Picture Arrangement and forward and
backward conditions of Digit Span. Motor deficits are frequently observed,
although there is some evidence of dissociation between cognition and
motor control in the early stages of the disease (J. A. Cooper et al., 1991).
Results of memory studies of individuals with Parkinson's disease have
shown impaired memory span, increased intrusions in delayed recall, and
impaired recognition memory, with average rates of forgetting (Massman,
Delis, Butters, Levin, & Salmon, 1990). Other results have indicated reduced
word-list learning in advanced stages of the disease compared to control-
group performance but similar to that of individuals with frontal lobe
impairment (Daum et al., 1995). Normal performance on WMS-R Logical
Memory passages and selective-reminding tasks but impaired list learning
on the California Verbal Learning Test (CVLT; Delis, Kramer, Kaplan, & Ober,
1987) have been reported (A. E. Taylor, Saint—Cyr, & Lang, 1990). Memory
dysfunction has been thought to relate to dysfunction of frontal-striatal
neural networks (A. E. Taylor et al., 1990). Studies have shown that perfor-
mance by individuals with Parkinson's disease on the WMS or the WMS-R is
lower than that of control-group participants (J. A. Cooper et al., 1991;
Pirozzolo, Hansch, Mortimer, Webster, & Kuskowski, 1982).

A sample of 10 participants diagnosed with idiopathic Parkinson's disease

were administered the WAIS-III and the WMS-III. See Table 4.31 for the
demographic data for this sample and Appendix F for the inclusion criteria
for participation in this study.
on the
On the WAIS-IIJ, impairment was expected to be most pronounced
partici-
POI and PSI tasks, with little or no impairment on the VCI tasks. The
nts in
pants with Parkinson's disease were also expected to show decreme
and the
performance on the WMS-III indexes. On both the WAIS-III
that of the
WMS-III measures, performance was predicted to be lower than

139
Evidence Base for Validity of the WAIS-III and the WMS-III

general population but not impaired to the degree observed in individuals

with mild Alzheimer’s disease.
Direct comparison in these studies between the groups with Alzheimer’s and
Huntington's disease is difficult because the latter group had significantly
lower education levels, which may reflect much lower premorbid general
intelligence. Such comparisons, therefore, are beyond the scope of these
studies.
WAIS-ill Results
The mean score and standard deviations for the WAIS-III IQ scales and
indexes of the sample are presented in Table 4.32. The mean WAIS—III IQ
scores are 94.6 (VIQ), 82.3 (PIQ), and 88.2 (FSIQ). As expected, the VIQ score
is relatively intact, whereas the PIQ score is relatively lower.
On the WAIS-III indexes, the differences between various types of function-
ing are more pronounced. The lowest scores were obtained on the POI (84.7)
and PSI (81.7). The mean score on the WM is 89.6, which is average to low
average. In contrast to these scores, the mean VCI score is in the average
range of functioning (96.9).

Taken together, these results suggest that individuals with Parkinson's dis-
ease have relatively spared verbal comprehensive abilities coupled with
compromised abilities in speed of perceptual processing, visual—spatial
organization, and, to some degree, working memory. The findings are con-
sistent with the pattern of IQ scores and with the relative sparing of verbal
comprehensive ability reported by Randolph et al. (1993).

WMS-lil Results
As the data in Table 4.32 show, the participants with Parkinson’s disease
obtained mean WMS-III index scores ranging from 80.9 (Visual Delayed
Index) to 90.0 (Auditory Recognition Delayed Index). The Index scores were
in the same general range as the WAIS-III PIQ and FSIQ scores. Overall, the
WMS-III Index scores are just over 1 SD lower than the general population
mean, but the extent of memory impairment is much less than that
observed in the participants with mild Alzheimer’s disease. The median
Single-Trial Learning and Learning Slope composite percentile scores are
18% and 51%, respectively. These results suggest that although performance
on the initial learning trial is lower than average, these individuals demon-
strate performance gains over subsequent trials similar to that of nonim-
paired individuals. The median Retention Composite percentile score (27%)
indicates slightly higher forgetting rates compared to those of a nonimpaired
control group.

140
 Evidence Based on Test-Criterion Relationships

Traumatic Brain Injury

Traumatic brain injury (TBI) is a neurological condition of relatively high
incidence in adolescence and early ac ilthood (Goldst in & Levin, 1990).
Head injuries most commonly occur iu individuals bewween birth and 24
years of age (Annegers, Grabow, Kurland, & Laws, 1980; Kraus et al., 1984).
Traumatic brain injury is highly variable in its long-term effects but can be
associated with numerous cognitive impairments (Dikmen, Reitan, &
Temkin, 1983). The actual functioning that becomes impaired depends on
the site and the severity of the damage. The frontal lobes and anterior tem-
poral lobes are particularly vulnerable to contusions, hemorrhages, and
hematomas associated with acceleration and deceleration forces of closed
head injury (Mattson & Levin, 1990). Frontal lobe damage is associated pri-
marily with executive dysfunction; in severe injuries, the dysfunction can be
quite debilitating because the individual experiences greatly impaired flexi-
bility in problem solving or in adaptability (Lezak, 1995). Other neuropatho-
logical processes associated with closed head injury include diffuse damage
produced by axonal shearing and stretching, which may especially cause
reduced speed of information processing (Adams, Graham, Murray, &
Scott, 1982).
Mild head injuries are most commonly associated with attentional deficits
(Capruso & Levin, 1992; Lezak, 1995; Shum, McFarland, & Bain, 1994).
Speeded performance is also likely to be adversely affected (Bawden,
Knights, & Winogron, 1985). However, it is memory dysfunction that is the
most common cognitive impairment reported after closed head injury
(Capruso & Levin, 1992). In cases of moderate closed head injury, symptoms
vary widely, and most individuals continue to experience significant impair-
ment 3 months after injury (Lezak, 1995). Dikmen, Machamer, Winn, and
Temkin (1995) found significant impairments one year after injury, but this
finding was largely a function of the severity of the injury. Cases of impaired
memory performance despite normal intellectual functioning have been
documented in individuals with severe closed head injury who were more
than 2 years posttrauma (Levin, Goldstein, High, & Eisenberg, 1988). The
presence of verbal learning deficits 12 months after head injury was best
predicted by occurrence of posttraumatic amnesia, duration of posttrau-
matic amnesia, and subarachnoid hematoma (Haslam et al., 1994). In older
adults, head injury has been associated with impairment in language pro-
duction and visual and verbal memory functioning (Goldstein et al., 1994).
Reid and Kelly (1993) examined WMS-R profiles of individuals who had
closed head injuries and found increased forgetting rates on the Logical
d by
Memory and Visual Reproduction subtests. Coma severity, as measure
Teasdale & Jennett, 1974),
the Glascow Coma Scale (Jennett & Bond, 1975;

141
Evidence Base for Validity of the WAIS-III and the WMS-III

was unrelated to WMS-R scores; however, duration of posttraumatic amne-

sia was negatively correlated with performance on the visual memory index-
es (Reid & Kelly, 1993). Group performance on the WMS-R indexes varied
from impaired performance on the delayed recall indexes to borderline per-
formance on the Verbal Memory and General Memory indexes to low aver-
age performance on visual memory and Attention/Concentration indexes.
All mean scores were lower than those of the noninjured control group (Reid
& Kelly, 1993). Individuals with severe head injury performed more poorly
than did control-group individuals on the Logical Memory and Paired
Associates learning tasks of the WMS, with duration of posttraumatic amne-
sia having predicted the degree of memory impairment (Brooks, 1976).
Hippocampal atrophy after severe closed head injury, as determined by
magnetic resonance imaging volumetric analysis, predicted decreased mem-
ory performance on the WMS-R Logical Memory and Visual Reproduction
subtests (Bigler et al., 1996).
A sample of 22 adults who had experienced a moderate to severe single
closed head injury were administered the WAIS-III and the WMS-III to
ascertain the sensitivity of these tests to memory problems associated
with this disorder. Participants were selected from various research centers
across the country. See Table 4.31 for the demographic data for this sample
and Appendix F for inclusion and exclusion criteria for participation in
this study.
It was predicted that the most affected WAIS-III subtests would be the
speeded Performance subtests (especially those of the PSI) and the subtests
composing the WMI. Verbal functioning was not expected to be as impaired,
and no modality-specific deficits were anticipated. It was also hypothesized
that individuals with closed head injury would show decrements in memory
performance on all WMS-III indexes.

WAIS-Ill Results
Table 4.32 shows the mean scores and standard deviations for the WAIS-II
IQ scales and indexes for the sample with traumatic brain injury. As predict-
ed for individuals with moderately severe head injury, the group exhibited
some overall impairment (FSIQ = 86.5). The mean PIQ score (84.5) is slightly
lower than the mean VIQ score (89.6), but not significantly lower.
On the WAIS-III indexes, the pattern of relative strengths and weaknesses is
clearer, with scores ranging from 73.4 to 92.1. As expected, the lowest score
was obtained on the PSI (73.4), a score that is significantly lower than the
other Index scores. In contrast to this deficit, performance on the POI is rela-
tively intact, with a mean score of 92.1. These results suggest that tasks
requiring rapid processing are significantly impaired relative to spatial tasks

142
 Evidence Based on Test-Criterion Relationships

for which timing is relatively less important. The mean scores on the VCI
(89.6) and WMI (89.8) are from low average to average and are relatively
lower than the scores obtained by the general population.
Taken together, these results suggest that individuals with moderate impair-
ment due to traumatic brain injury have global cognitive deficits but with
processing speed being predominantly affected.

WMS-lll Results
As the data in Table 4.32 show, the participants with closed head injury
exhibit an unexpected pattern of memory performance, with mean WMS-III
Index scores ranging from 74.3 (Visual Delayed Index) to 93.6 (Auditory
Recognition Delayed Index). Although no modality-specific deficits had been
predicted, the participants performed more poorly on the visual memory
indexes than on the auditory memory indexes. For example, whereas 18% of
the sample received a score below 70 on the Auditory Immediate Index, 38%
of the sample received a score in the same range on the Visual Immediate
Index. The median WMS-III Auditory Process Composite percentile scores
range from 29% (Single-Trial Learning) to 50% (Retrieval); thus, these
examinees evidenced mildly reduced encoding and storage but average
retrieval skills.
D. C. Fisher, Ledbetter, Cohen, Marmor, and Tulsky (2000) compared the
WAIS-III and WMS-III score profiles of patients with mild traumatic brain
injury, patients with moderate to severe traumatic brain injury, and a
matched control group. The authors reported statistically significant score
differences for all of the WAIS—III and WMS-III indexes, with the exception
of the Auditory Recognition Delayed Index. The largest effect sizes were
observed for the Processing Speed, Visual Immediate, and Visual Delayed
Memory indexes. On both the WAIS-III and WMS-III, all index scores for the
group with moderate to severe injury were significantly lower than those of
the control group. For the group with mild injury, the WAIS-III scores did not
differ significantly from those of the control group. On the WMS-III, how-
ever, this group performed significantly lower than the control group on the
Auditory Immediate, Immediate, Auditory Delayed, Visual Delayed, and
General Memory indexes. Fisher et al. concluded that the WMS-III measures
are more sensitive to the effects of head injury than are the WAIS-III mea-
sures even though the largest effect size was observed on WAIS-III PSI.
Donders, Tulsky, and Zhu (2001) compared WAIS-III performance of 100
patients with traumatic brain injury and a matched control group drawn
of the
from the WAIS-III standardization sample. The authors found that two
three new WAIS-III subtests are sensitive to the effects of head injury.

143
Evidence Base for Validity of the WAIS-III and the WMS-IIT

Although the patients with moderate to severe traumatic brain injury per-
formed more poorly than the control group on Letter-Number Sequencing
and Symbol Search, Matrix Reasoning did not differentiate the clinical group
from the control group. None of the WAIS-III measures differentiated the
group with mild head injury from the control group. Performance by the
group with moderate injury was more impaired than performance by the
group with mild injury on the same variables. The authors tested logistic
regression models to determine if WAIS-1II performance could differentiate
those with mild injury from those with moderate to severe injury. Symbol
Search had the largest effect size, and additional variables did not improve
the overall model statistics. A cutoff score of 9 was established to differenti-
ate the two clinical groups. The resulting classification table indicated mod-
est classification ability. The WAIS-III tests are sensitive to the effects of
traumatic brain injury. Futher research is needed to enable the determina-
tion of injury severity based on WAIS-III scores alone (Donders et al., 2001).

Multiple Sclerosis
Multiple sclerosis (MS) is a relatively common neurological disorder of
young and middle adulthood (Hauser, 1994). Neuropathological processes
include inflammation, scarring, and demyelination of neurons in the central
nervous system, producing dysfunction of conductance in these cells
(Hauser, 1994). Differentiated according to two patterns of MS symptomatol-
ogy, individuals with relapsing—remitting MS demonstrated limited cognitive
impairment, whereas those with chronic—progressive MS were diffusely
impaired but with memory deficits found only in verbal memory (Heaton,
Nelson, Thompson, Burks, & Franklin, 1985). White-matter lesions detected
by magnetic resonance imaging have been associated with memory dys-
function in individuals with MS (L. Ryan, Clark, Klonoff, Li, & Paty, 1996),
with total lesion area the best predictor of neuropsychological impairment
(Swirsky—Sacchetti et al., 1992).
Examinations of memory functioning in individuals with MS have indicated
impaired performance on prose recall (Goldstein, McKendall, & Haut, 1992;
Grigsby, Ayarbe, Kravcisin, & Busenbark, 1994; Litvan, Grafman, Vendrell, &
Martinez, 1988) of the WMS and WMS-R but normal recall for important
ideas within the stories (Goldstein et al., 1992); lower WMS MQ scores com-
pared to scores of a control group (Litvan et al., 1988; Rao, Hammeke,
McQuillen, Khatri, & Lloyd, 1984); and lower scores on WMS Verbal Paired
Associates (Maurelli et al., 1992; Rao et al., 1984), Digit Span Forward and
Backward (Grigsby et al., 1994; Krupp, Sliwinski, Masur, Friedberg, & Coyle,
1994), Mental Control (Rao et al., 1984), Visual Reproduction (Rao et al.,
1984), and List Learning (Maurelli et al., 1992). Other studies failed to find
lower performance on measures of verbal memory (Krupp et al., 1994) or

144
 Evidence Based on Test-Criterion Relationships

Digit Span (Litvan et al., 1988). J.S. Fisher (1988) reported three patterns of
impaired memory functioning on the WMS-R for individuals with MS. One
group exhibited significant, global impairment; the second group had intact
working memory and mildly impaired memory and learning; and the third
group was relatively unimpaired. Performance by the group with MS on all
WMS-R indexes was in the average range but in the high average range for
the matched control group. These results suggest loss of functioning, espe-
cially in delayed recall, in the group with MS (J. S. Fisher, 1988). No modal-
ity-specific deficits were observed in that study (J. S. Fisher, 1988).
A sample of 25 participants were administered the WMS-III. Because of sam-
pling constraints, data for the WAIS-III were not obtained for this group. See
Table 4.31 for the demographic data for this sample and Appendix F for the
inclusion criteria for participation in this study.
As a group, the individuals with MS were expected to obtain lower scores on
the WMS-III indexes than the general population. Furthermore, it was
expected that although a small percentage of the sample would obtain Index
scores in the impaired range, the overall mean scores would be in the low
average to average range.
As shown in Table 4.32, the mean WMS-III Index scores range from 81.5
(Visual Immediate Index) to 97.7 (Auditory Immediate Index) for the sample
with MS. From 10% to 30% of the participants scored below 70 on most of
the WMS-III indexes except the Working Memory Index and the Auditory
Recognition Delayed Index. Contrary to some studies previously cited, indi-
viduals in this sample were more likely to perform in the impaired range on
visual indexes than on the auditory indexes. As a group, the participants
with MS exhibited forgetting rates similar to those exhibited by individuals
with Huntington's disease, Parkinson's disease, and traumatic brain injury.

Temporal Lobe Epilepsy

Epilepsy is a relatively common neurological disorder, having a lifetime
prevalence rate of approximately 2% (McIntosh, 1992). Surgical resection of
the epileptogenic foci is one method for the treatment of intractable seizure
disorder (Dichter, 1994).

Hippocampal sclerosis was found in 60% of a sample of individuals with

treatment-resistant epilepsy and was related to deficits in recall regardless of
hemisphere (McMillan, Powell, Janota, & Polkey, 1987). Patients receiving
surgical treatment for left temporal lobe epilepsy tend to perform more
poorly after surgery than before surgery and in comparison to individuals
not receiving treatment (Chelune, Naugle, Liiders, Sedlak, & Awad, 1993).

145
Evidence Base for Validity of the WAIS-III and the WMS-III

Attempts to identify material-specific memory deficits in individuals with

right versus left temporal lobe epilepsy or after right or left temporal lobec-
tomy have yielded mixed results. The results of some studies have indicated
impaired facial memory (Beardsworth & Zaidel, 1994) and impaired spatial
memory (M. L. Smith & Milner, 1989) associated with right temporal lobe
epilepsy but not with left temporal lobe epilepsy. Research studies based on
the WMS or the WMS-R have provided the following findings: Performance
on the Logical Memory subtest was more impaired in examinees with left
temporal lobe epilepsy than those with right temporal lobe epilepsy and cor-
related with left hippocampal neuron loss. The WMS-R has been shown to
be sensitive to material-specific deficits after left but not right temporal
lobectomy (Naugle et al., 1993).
The WMS-III and some subtests from the WAIS-III (Vocabulary, Similarities,
Digit Span, Letter-Number Sequencing, Digit Symbol—Coding, Block
Design, and Matrix Reasoning) were administered to a sample of 15 partici-
pants with left temporal lobe epilepsy and 12 with right temporal lobe
epilepsy, all of whom had undergone hippocampectomy for the treatment of
intractable seizure disorder. Participants were selected from various clinical
centers across the country. See Table 4.31 for the demographic data for
this sample and Appendix F for the inclusion criteria for participation in
this study.
Although material-specific deficits after right and left temporal lobectomy
have not been consistently reported, it was expected that participants who
had undergone left temporal lobectomy would obtain lower scores on tasks
with auditorily presented material than on tasks with visually presented
materials; it was less certain but also of interest whether participants who
had undergone right temporal lobectomy would exhibit the opposite pat-
tern. The mean performance of both groups was expected to be lower than
the mean performance of the general population.
WAIS-III Results
As predicted, deficits in intellectual functioning were relevant to the site of
the lobectomy (see Table 4.33). For the group of examinees who had a left
temporal lobectomy, scores on the WAIS-III subtests that measure verbal
comprehension skills (Vocabulary and Similarities) are lower than the gener-
al population mean (7.5 and 8.1, respectively). Additionally, the standard
deviations are also smaller than those for the general population. The
group's scores on tasks requiring fluid reasoning and visual processing show
no impairment. For the group with left temporal lobectomy, the results are
consistent with previous findings. For the individuals with right temporal
lobectomy, the scores on the WAIS-III subtests are close to the mean of the
general population, ranging from 8.8 to 10.7. These results are consistent
with previous research.

146
 Evidence Based on Test-Criterion Relationships

However, because both sample sizes were quite small (i.e., 15 and 12), fur-
ther research is needed to determine if these results are representative.

WMS-IIl Results
As the data in Table 4.33 show, the participants with left lobectomy obtained
mean WMS-III Index scores ranging from 77.3 (General Memory Index) to
95.4 (Working Memory Index). The participants with right lobectomy
obtained mean Index scores ranging from 83.5 (Visual Immediate Index) to
97.8 (Working Memory Index). The mean performance for both groups is
lower than the general population mean, and the group with left lobectomy
generally show a greater degree of memory impairment than does the group
with right lobectomy (e.g., General Memory Index scores are 77.3 and 87.6,
for the left lobectomy and right lobectomy groups, respectively).
The general trend of lower scores on auditory tasks than on visual tasks by
the participants with left lobectomy was exhibited for both the immediate
and the delayed conditions. For the participants with right lobectomy, the
opposite pattern was observed, with poorer performance on the visually
presented tasks than on the auditorily presented tasks. The median per-
centile scores for the WMS-III Auditory Process Composites also indicate
that the group with left lobectomy performed lower than the group with
right lobectomy on the Single-Trial Learning, Learning Slope, and Retention
composites. The especially poor Retention score suggests a worse storage
deficit (i.e., rapid forgetting over delay interval). On the other hand, perfor-
mance on the Retrieval Composite is relatively higher for the group with left
lobectomy than for the group with right lobectomy, so a retrieval deficit
appears to be responsible for some of the problems. This pattern suggests
that recognition aids retrieval more for the group with left lobectomy than
for the group with right lobectomy.
Additional studies of WMS-III performance by patients diagnosed with left
or right temporal lobe epilepsy have been conducted. Doss, Chelune, and
Naugle (2000) compared the performance on the published version of the
WMS-III and performance on the standardization edition. The standardiza-
tion edition contained more memory subtests than the published version
because some subtests were not included in the published version for psy-
chometric reasons. Doss et al. noted that this shortening of the battery
might have affected the quality of the normative data by changing the time
between immediate and delayed recall conditions, changing potential inter-
ference factors due to the reordering of the test sequences and changing
study,
fatigue factors (particularly on Letter-Number Sequencing). For this
respondent s
the patients receiving the standardization protocol were those
Manual. For the
originally documented in the WAIS—IIJ—WMS-III Technical
to an
comparative sample, the authors administered the published battery

147
Evidence Base for Validity of the WAIS-III and the WMS-III

additional 30 patients. The results indicated no statistically significant differ-

ences in performance between the patients receiving the published battery
and those receiving the standardization battery. Also, the authors noted that
respondents with right temporal lobe epilepsy performed better on auditory
measures than on visual memory measures and that respondents with left
temporal lobe epilepsy performed better on visual memory measures than
on auditory measures. These results confirm the results of the original study
(The Psychological Corporation, 1997).

Table 4.33. |WAIS-III and WMS-III Performance of Samples With

Temporal Lobe Epilepsy
Left Right
Lobectomy Lobectomy
Mean SD Mean SD

WAIS-III Subtests
Vocabulary Wes ie 8.8 2.1
Similarities 8.1 1.8 10.3 2.4
Digit Span 8.5 PS 10.5 2.6
Letter-Number Sequencing On 3.4 9.6 3.2
Digit Symbol—Coding 9.4 2.0 9.8 3.0
Block Design 10.4 2.8 10.2 3.2
Matrix Reasoning 10.2 AS 10.7 3.3
WMS-Ill Primary Indexes
Auditory Immediate 77.9 16.3 95.0 11.3
Visual Immediate 86.5 15.4 83.5 9.1
Immediate Memory 78.1 16.2 87.2 10.7
Auditory Delayed 75.4 14.5 93:5 11.9
Visual Delayed 85.3 16.5 84.3 alilé7/
Auditory Recognition Delayed 83.0 18.5 92.1 15.7
General Memory 77.3 15.1 87.6 2a
Working Memory 95.4 15.6 97.8 12.9
Auditory Process Composites Median %ile %ile Range Median %ile %ile Range
Single-Trial Learning 21% 1%—-44% 36% 14%-87%
Learning Slope 11% 1%-95% 39% 1%-74%
Retention 2% 1%-39% 36% 1%-95%
Retrieval 65% 11%-99% 42% 5%-78%
N 15 12

Wilde et al. (2001) further explored the utility of WMS-III profiles in differen-
tiating respondents with right versus left temporal lobe epilepsy. The results
of this study confirmed findings from previous studies, specifically better

148
 Evidence Based on Test-Criterion Relationships

auditory versus visual memory in patients with right temporal lobe epilepsy
and the opposite pattern for patients with left temporal lobe epilepsy. Wilde
et al. also found that these effects were stronger for delayed versus immedi-
ate memory measures. The application of these score differences did not
consistently classify patients accurately as belonging to the right or left tem-
poral lobe epilepsy group. The authors concluded that the WMS-III may not
be useful in identifying location (right versus left) of seizure foci prior to
surgery but may be useful in establishing baseline information for assess-
ment of change after surgery and identification of those at risk for impair-
ment subsequent to surgery (Wilde et al., 2001).

Assessment of Alcohol-Related Disorders

The study of patients with Korsakoff’s syndrome is particularly important in
establishing the divergent cognitive processes tapped by the WAIS-III and
WMS-III. For the following studies involving individuals with alcohol-related
disorders, the performance data for the WAIS-III and the WMS-III are pre-
sented in Table 4.34.

Table 4.34. WAIS-III and WMS-III Performance of Samples With

Alcohol-Related Disorders
Chronic Korsakoff’s
Alcohol Abuse Syndrome
Mean SD Mean SD

WAIS-III Scales/Indexes
VIQ 108.6 12.7 94.5 10.3
PIQ 101.2 14.5 92.2 17.7
FSIQ 106.1 13.5 92.8 13.6
VCl 109.0 11.4 92.7 9.3
POI 102.0 14.0 96.9 15.4
WMI 104.6 12.3 98.4 15.5
PSI 97.7 12.5 88.2 19.3
WMS-III Primary Indexes
Auditory Immediate 108.0 15.4 73.1 78
Visual Immediate 96.0 14.5 67.8 6.8
Immediate Memory 102.5 16.5 64.4 8.2
Auditory Delayed 107.3 15.5 63.5 5.3
Visual Delayed 97.9 13.5 65.4 8.3
Auditory Recognition Delayed 109.6 13.0 64.5 8.0
General Memory 105.2 14.2 57.8 6.7
Working Memory 98.0 9.3 97.8 13.0
Auditory Process Composites Median %ile %ile Range Median %ile {ile Range
Single-Trial Learning 75% 9%-99% 12% 3%-36%
Learning Slope 52% 1%-95% 3% 1%-34%
Retention 59% 4%-98% 1% 1%-4%
Retrieval 50% 11%-97% 28% 5%-78%
28 10
oeN III ———— eee
149
Evidence Base for Validity of the WAIS-III and the WMS-III

Chronic Alcohol Abuse

Several essential and associated features characterize alcoholism (American
Psychiatric Association, 1994). Alcohol abuse consists of a “maladaptive pat-
tern of [alcohol] use manifested by recurrent and significant adverse conse-
quences related to the repeated use of [alcohol]” (American Psychiatric
Association, 1994, p. 182). In addition to maladaptive behaviors, a growing
body of research has documented cognitive dysfunction in those who abuse
alcohol but who do not have Korsakoff’s syndrome.
The neuropsychological effects of chronic alcoholism are unclear. First,
there seem to be etiological differences in the causes of alcoholism, different
features that are associated with alcoholism, and different risk factors in pre-
dicting alcoholism. Some investigators have classified individuals with alco-
holism into different types (e.g., Bohman, Cloninger, Sigvardsson, & von
Knorring, 1987), with each type associated with differential patterns of intel-
lectual and memory deficits. Others have reported the degree of variability
in the neuropsychological effects of chronic alcoholism (Rourke & Loberg,
1996). Still other research has suggested that there may be premorbid cogni-
tive deficits in those who abuse alcohol that may be expressed in increased
rates of learning disabilities in their histories (see Rhodes & Jasinski, 1990) or
in the higher rates of neuropsychological impairment in individuals at high
risk for becoming a substance abuser (see Tarter, Hegedus, Goldstein, Shelly,
& Alterman, 1984). The most important variable contributing to the mixed
results might be the variety of risk factors that must be accounted for in
studies of the relation between cognitive functioning and alcoholism. For
instance, these risk factors may include moderator variables such as the ten-
dency toward antisocial personality (e.g., Malloy, Noel, Rogers, Longabaugh,
& Beattie, 1989; Schuckit, Smith, Anthenelli, & Irwin, 1993), the number of
years of alcohol consumption (C. Ryan & Butters, 1983; Tarter, 1973), and age
(Jones & Parsons, 1971; Malloy et al., 1989; Parsons & Farr, 1981; C. Ryan &
Butters, 1980). The last factor to be controlled in any study is the length of
time since the individual has stopped consuming alcohol. Often, the neuro-
psychological effects tend to “recover” over time (Ellenberg, Rosenbaum,
Goldman, & Whitman, 1980; Ellis & Oscar-Berman, 1989; Glenn & Parsons,
1990) so that the individual tends to return much closer to his or her baseline
functioning. Additionally, the number of times the individual has returned to
a pattern of alcohol abuse and then stopped the abuse may also be a salient
variable in the equation (Glenn, Parsons, Sinha, & Stevens, 1988).
In view of the variables that might play a part in the effects of persistent
chronic alcoholism on cognitive functioning, it is not surprising that the
results from different studies yield inconsistent patterns. Indeed, some stud-
ies of the intellectual functioning of chronic alcohol abusers have indicated

150
 Evidence Based on Test-Criterion Relationships

average or slightly higher levels of ability (Bowden, Whelan, Long, & Clifford,
1995; Eckardt et al., 1996; Oscar—-Berman, Clancy, & Weber, 1993), whereas
other studies showed mildly impaired performance (e.g., Rhodes & Jasinski,
1990). The reported intellectual deficits have tended to fall into three
domains (Malloy et al., 1989): abstract reasoning, learning and memory, and
visual-spatial ability. On the WAIS-R, subtests that may be differentially
affected are Block Design, Digit Symbol, and Object Assembly (Ellis & Oscar-
Berman, 1989; O'Mahony & Doherty, 1993; Parsons & Farr, 1981). Addition-
ally, Verbal subtests measuring crystallized intelligence (e.g., Vocabulary,
Information, and Similarities) are not typically affected by chronic alco-
holism (Ellis & Oscar-Berman, 1989; O’Mahony & Doherty, 1993; Rourke &
Loberg, 1996; Tamkin & Dolenz, 1990).
The results of studies of alcohol abusers after detoxification have shown
impairment on the Logical Memory and Visual Reproduction subtests of the
WMS and the WMS-R (O’Mahony & Doherty, 1993, 1996). Using the WMS-R,
J. J. Ryan and Lewis (1988) studied memory functioning of alcohol abusers
2-6 weeks after detoxification. Compared to the control participants, the
alcohol abusers performed more poorly on all WMS-R indexes and especial-
ly on the Digit Span, Logical Memory, and Visual Reproduction subtests (J. J.
Ryan & Lewis, 1988). The level of performance of the alcohol abusers was in
the low average to average ranges across indexes, with performance only on
the Visual Memory Index in the low average range (J. J. Ryan & Lewis, 1988).
The observed deficits in memory functioning may be a transient phenome-
non. Ellenberg et al. (1980) reported verbal and visual—spatial memory
impairment in alcohol abusers within 2 days of detoxification. However, ver-
bal memory deficits remitted quickly, and only a few older, chronic alcohol
abusers sustained visual-spatial memory deficits longer than 25 days
(Ellenberg et al., 1980). Long-term abstinence has been associated with nor-
mal neurocognitive functioning (Reed, Grant, & Rourke, 1992). Although
mammillary body shrinkage was found in chronic alcohol abusers without
Korsakoff’s syndrome, this group did not display poorer memory perfor-
mance compared to a matched control group (Davila, Shear, Lane, Sullivan,
& Pfefferbaum, 1994). Memory impairment may occur only in a subset of
chronic alcohol abusers who also have cirrhosis of the liver (Arria et al.,
1991) or in those with comorbid antisocial personality disorder (Malloy et
al., 1989).
A sample of 28 participants with alcoholism who had recently undergone
detoxification and who did not meet diagnostic criteria for Korsakoff’s syn-
drome were administered the WAIS-III and the WMS-III to determine the
sensitivity of these tests to cognitive dysfunction and memory in this group.
See Table 4.31 for the demographic data for this sample and Appendix F for
the inclusion criteria for participation in this study.

151
Evidence Base for Validity of the WAIS-III and the WMS-III

The participants with alcoholism were predicted to show variable perfor-

mance, with their mean level of performance in the low average to average
ranges. Further, their performance on the visual memory indexes was
expected to be relatively lower than that on the auditory memory indexes.
For the WAIS-III, the PIQ score was expected to be somewhat lower relative
to the VIQ score (because of lower scores on the Block Design and Digit
Symbol—Coding subtests). However, because the Digit Symbol—Coding and
Block Design subtests might be related to speed of information processing,
the group’s scores on the POI and PSI were predicted to be the lowest of all
the WAIS-III Index scores.

WAIS-III Results
The results obtained in the study (see Table 4.34) indicate that all the scores
on the WAIS-III obtained by the individuals who had been diagnosed with
chronic alcoholism are in the average range of functioning. Relative to their
Verbal scores, the group’s WAIS-III PIQ, POI, and PSI scores are slightly
lower, a pattern that was predicted. The PIQ and POI scores are about 0.5
SDs below the VIQ and VCI scores. However, the PSI score is significantly
lower (about 1 SD) than the Verbal scores. The results of this study suggest
only very subtle, if any, deficits, but this pattern is in the same direction as
those found in previous studies.
WMS-Ill Results
As the data in Table 4.34 show, the mean performance of this sample on the
WMS-II indexes range from 96.0 (Visual Immediate Index) to 109.6
(Auditory Recognition Delayed Index). As expected, the mean performance
on all of the indexes is in the average range. Also as expected, mean perfor-
mance on the auditory indexes is relatively higher than mean performance
on the visual indexes, although scores on the visual indexes are in the aver-
age range. Only 7.1% of the sample (i.e., 2 participants) received an Index
score below 70, and this occurred on the Visual Immediate Index. This
group's performance on the WMS-III is consistent with results reported in
the literature.

Korsakoff’s Syndrome
Korsakoff’s syndrome is a neurological disorder associated with chronic
alcoholism, resulting in transient and long-term cognitive deficits (Salmon,
Butters, & Heindel, 1993). Acute disturbances in ocular movement, gait, and
orientation are reversible with proper treatment. Nutritional deficiencies are
believed to be responsible for the development of Korsakoff’s syndrome.
Long-term cognitive problems may include executive dysfunction, apathy,
severe anterograde amnesia, and retrograde amnesia, with relatively intact
working memory and overall intellectual functioning (Salmon et al., 1993).

152
 Evidence Based on Test-Criterion Relationships

The results of memory studies have indicated that individuals with

Korsakoff’s syndrome exhibit global impairment in memory compared to
individuals with temporal lobe epilepsy and those with non-Korsakoff’s alco-
holism who may exhibit material- or modality-specific deficits despite aver-
age verbal intellectual functioning (Cutting, 1978). Shimamura and Squire
(1984) reported impaired associative learning in individuals with Korsakoff’s
syndrome compared to those with chronic alcoholism in nonpriming condi-
tions. Butters et al. (1988) reported that, on the WMS-R, persons with
Korsakoff’s syndrome exhibited impaired performance on the Verbal,
General, and Delayed Recall indexes, although visual memory was in the
borderline range and the mean Attention/Concentration Index was average.
A sample of 10 participants diagnosed with Korsakoff’s syndrome were
administered the WAIS-III and the WMS-III to assess the sensitivity of these
tests to this amnesic disorder. See Table 4.31 for the demographic data for
this sample and Appendix F for the inclusion criteria for participation in
this study.
in previous research (e.g., Butters & Cermack, 1980), dramatic differences
between scores on intellectual tests and memory tests have been reported
for individuals with Korsakoff’s syndrome. Therefore, such differences were
expected for this study as well. The participants with Korsakoff’s syndrome
were expected to exhibit relatively intact WAIS-III scores but show impaired
performance on all of the WMS-III memory indexes, except the Working
Memory Index. The level of impairment on these memory indexes was
expected to be similar to that of individuals with Alzheimer’s disease.

WAIS-III Results
Table 4.34 presents the WAIS-III mean performance data for individuals with
Korsakoff’s syndrome. As predicted, the WAIS-III scores obtained by these
individuals are relatively intact. The mean IQ scores range from 92.2 (PIQ
to 94.5 (VIQ). Only the mean PSI score is slightly lower (88.2) but not
significantly.

WMS-IIl Results
Table 4.34 presents the WMS-III mean performance data for the participants
with Korsakoff’s syndrome. As expected, mean performance on the WMS-III
Working Memory Index is in the average range (97.8), and mean scores on
the other indexes range from 57.8 (General Memory Index) to 73.1 (Auditory
Immediate Index). The range of scores on the General Memory Index for the
individuals in this sample was 45-66, indicating consistently, severely
impaired memory performance. The WMS-III Auditory Process Composites
also show a pattern of impaired performance for this group. The Single-Trial

153
Evidence Base for Validity of the WAIS-III and the WMS-III

Learning Composite median percentile score (12%) shows an overall

reduced ability to learn from a single-trial exposure to auditory stimuli. The
median Learning Slope Composite percentile score (3%) suggests that these
individuals benefit little from repeated exposure to the material to be
learned. Also, the material that is learned is quickly forgotten from immedi-
ate to delayed conditions, as indicated by the Retention Composite median
percentile score of 1%. Finally, the Retrieval Composite percentile score
(28%) reveals that these participants were not aided by the recognition for-
mat. Their results suggest severely impaired encoding and storage with rela-
tively normal attention, working memory, and retrieval.

Assessment of Neuropsychiatric Disorders—

Schizophrenia
For the study involving individuals with schizophrenia, the performance
data for the WAIS-III and the WMS-II/ are presented in Table 4.35.

Table 4.35. WAIS-III and WMS-III Performance of Samples

With Schizophrenia
Mean SD

WAIS-III Scales/Indexes
VIQ 88.9 15.2
PIQ 86.2 12.9
FSIQ 86.8 14.8
VCl 93.3 16.4
POI 89.6 13.9
WMI 85.0 15.1
PSI 83.4 11.8
WMS-Iil Primary Indexes
Auditory Immediate 83.3 15.6
Visual Immediate 82.3 14.3
Immediate Memory 79.1 15.7
Auditory Delayed 84.4 15.9
Visual Delayed 79.3 14.8
Auditory Recognition Delayed 86.1 14.9
General Memory 79.7 15.8
Working Memory 85.6 13.9
Auditory Process Composites Median %ile %ile Range
Single-Trial Learning 9% 1%-96%
Learning Slope 34% 1%-95%
Retention 19% 1%-95%
Retrieval 50% 2%-88%
N 42
uw EL

154
 Evidence Based on Test-Criterion Relationships

Schizophrenia is a chronic, severe neuropsychiatric disorder characterized

by “positive” symptoms, such as hallucinations, delusions, and thought dis-
orders, and by “negative” symptoms, such as alogia, anhedonia, avolition,
and flattening of affect (American Psychiatric Association, 1994; Andreasen
& Carpenter, 1993). Schizophrenia is a relatively common disorder, having a
lifetime prevalence of 0.5%-1% (Andreasen & Carpenter, 1993). Although the
causes of schizophrenia remain unknown, the illness is reliably associated
with abnormalities of brain structure and physiology as demonstrated by
neuroimaging studies (R. E. Gur et al., 1991; Shapiro, 1993). Dysfunction of
the temporal-hippocampal region (Arnold, Hyman, Van Hoesen, & Damasio,
1991; Conrad, Abebe, Austin, Forsythe, & Scheibel, 1991) and the frontal cor-
tex (Goldberg, Weinberger, Berman, Pliskin, & Podd, 1987; Seidman et al.,
1994; Weinberger, Berman, & Zec, 1986) have been implicated in schizophre-
nia (Gold et al., 1994; Seidman et al., 1994).
The types of cognitive deficits occurring with schizophrenia are wide rang-
ing. Early work on schizophrenia demonstrated disrupted attention and
slowed reaction time (Shakow, 1963). Heaton and Crowley (1981) observed
poor performance on a series of tests measuring general intellectual func-
tioning and concluded that individuals with schizophrenia perform at levels
characteristic of someone with gross neurological damage. Individuals with
chronic schizophrenia have shown deficits in several areas, including
abstract reasoning, word fluency, sequential memory, cognitive set-shifting,
and attention (Dickerson, Ringel, & Boronow, 1991; Gold et al., 1994;
Seidman et al., 1994). Other research has also suggested that these individu-
als perform poorly on tasks requiring working memory (Gold, Carpenter,
Randolph, Goldberg, & Weinberger, 1997).
Although memory dysfunction is not considered a diagnostic criteria of
schizophrenia, results of neuropsychological studies have demonstrated sig-
nificant impairment in memory functioning, especially verbal memory func-
tioning (Saykin et al., 1991). This pattern of deficits has been reported in
individuals with first-episode schizophrenia as well as for individuals with
chronic schizophrenia (Saykin et al., 1994). This finding suggests that the
deficit is not the product of exposure to neuroleptics. Other researchers have
reported deficits in visual-spatial memory (Hoff, Riordan, O’Donnell, Morris,
& DeLisi, 1992) and in memory for spatial context (Rizzo, Danion, Van Der
Linden, Grange, & Rohmer, 1996). Although schizophrenia is a chronic disor-
der, the severity of neuropsychological deficits, including memory dysfunc-
tion, does not appear to be progressive, as it is in dementia (Heaton et al.,
1994). It has been suggested that verbal learning deficits associated with
schizophrenia appear to be related to difficulties in encoding and retrieving
information without significant storage (rapid forgetting) problems (Paulsen
etal;, 1995).

155
Evidence Base for Validity of the WAIS-III and the WMS-III

The results of studies based on the WMS and the WMS-R have indicated
impaired or reduced performance on the Logical Memory, Verbal Paired
Associates, and Visual Reproduction subtests (Hoff et al., 1992; Saykin
etialy 199De

For the present study, a sample of 42 participants with schizophrenia were

administered the WAIS-III and the WMS-III to determine the sensitivity of
these tests to individuals diagnosed with schizophrenia. See Table 4.31 for
the demographic characteristics of this sample and Appendix F for the inclu-
sion criteria for participation in this study.
It was hypothesized that the participants with schizophrenia would perform
in the low average to average range of functioning on the WAIS—III measures
that reflect general intellectual functioning. However, more pronounced
deficits and, thus, relatively lower scores would be found on the measures
related to working memory and processing speed. Furthermore, these
individuals were expected to perform more poorly on all of the WMS-II
indexes compared to the general population, with no modality-specific
deficits expected.

WAIS-III Results
Table 4.35 presents the WAIS-III mean performance data for the individuals
with schizophrenia. As predicted, the WAIS-III mean IQ scores obtained by
these individuals are in the low average range (86.2—88.9), indicating that
these individuals have a relative decrement in functioning compared to the
general population. The mean performance on the Index scores match the
predictions that had been made. The group’s mean scores on the VCI and
POL are relatively (though slightly) higher (93.3 and 89.6, respectively) than
those on the WMI (85.0) and PSI (83.4). These results support the premise
that individuals with schizophrenia will show impairments on tasks requir-
ing attention, processing speed, and working memory.

WMS-lll Results
As the data in Table 4.35 show, the mean performance on the WMS-III
Primary Indexes ranges from 79.1 (Immediate Memory Index) to 86.1
(Auditory Recognition Delayed Index). The participants with schizophrenia
performed in the low average to borderline ranges on every Primary index.
The results do not indicate any trends with regard to auditory or visual
modality-specific deficits. These results do indicate that the WMS-III is sen-
sitive to memory deficits observed in previous studies of individuals with
schizophrenia. Hawkins (1999) reported that the data provided in the
WAIS-III—WMS-III Technical Manual (The Psychological Corporation, 1997)
suggest mild deficits in new learning, with memory performance similar to
intellectual ability, in most cases.

156
 Evidence Based on Relationships With Other Variables

Evidence Based on Relationships

With Other Variables
WAIS-IIl Correlations With
Selected External Measures
Table 4.36 provides the demographic data and the sizes of the samples on
which the WAIS-III comparison studies were based. The data are presented
as mean age and the percentages of each sample by sex, race/ethnicity, and
education level. The samples are ordered by groups according to the type of
external measure (e.g., cognitive ability). The table also includes the percent-
ages of each sample by clinical diagnostic group because the clinical diag-
noses of the participants might have affected the correlational results.

Measures of Academic Achievement

The WIAT-II is the most recent revision of the WIAT. The WIAT-II measures
reading, math, writing, and language skills in youth and adults. The expand-
ed normative data set includes norms for ages 4-89 and facilitates the
assessment of academic skills through adulthood. The WAIS-III linking sam-
ple had the following demographic composition for adolescents: By sex, the
sample was 52.9% female and 47.1% male; by race/ethnicity, 66.3% White,
10.6% African American, 18.3% Hispanic, and 4.8% Asian. By parent educa-
tion level, the sample was 19.2% <12 years, 35.6% high school or equivalent,
28.3% some college, and 16.3% college. By geographic region, the sample
had the following composition: 35.6% South, 26%. West, 21.2% North
Central, and 17.3% Northeast. For college students, the sample was 51.1%
female and 48.9% male; by race/ethnicity, 66% White, 15.7% African
American, 8.6% Hispanic, and 8.2% Asian. The adult sample was 58% female
and 42% male; by race/ethnicity, 72% White, 12% African American, 7%
Hispanic, and 2% Asian; by education level, 10% <12 years, 40% high school
or equivalent, 25% some college, and 15% college; and by geographic region,
54% South, 22% West, 19% North Central, and 5% Northeast.
Table 4.37 presents correlations between WAIS-III FSIQ, VIQ, and PIQ with
WIAT-II subtest and composite scores for three groups: 17-19 year-olds, col-
lege students, and adults. For the 17-19 year-old group, high positive corre-
lations were observed between the VIQ and FSIQ and most of the WIAT-II
composite scores. The PIQ had a moderate correlation with reading and high
~ correlations with math and language skills. For college students, the correla-
tion between academic performance and intellectual ability was not as large,
a finding likely due to the homogeneity of the group and resulting restriction

157
 a1ge] "OE'H sIydeiZ0UIEg
eyeq
Jo ay) saydures
10} 2U) [2U19}Xq/TII-STVM
sainsvayq] uostreduro;)
saIpnjs 2

158
eaniubop
Ayiqy uoUEny Mowe, 8
4-SIVM Y-OSIM~WdS Bogs
=SHd =Y-SWM Buryew-iteuL
say= Boos Y-SINM §=LTAD O-Aey
= Boje 2
N 88 ve rd 9I vl cic} 9S 9k Le 6L ve 9 S
aby 2
ues|\ 98r eet Lge Lev 88S Ler Les L¢y 92S S'6S O'S Livy ee
as G02 oF os 0°02 srl 6Sl Obl 0:02 Sel St wae 0°02 5
xaSs >
eyewe gle Sle Lee eitord 82 yoe vor See 6h g9e Sle loitsr a
slew G29 leis) ELL GOL OL 9'€9 9'eS GOL L'28 Ze9 g29 GOL a)
Ayoiuyya/e0ey =
auM v6 osZ ell 0'001 S06 e298 3°26 0'001 8°96 768 0001 0°00 a
UO UBOLEWY SY Lice, L'8 60l vs Ze es =)
oiuedsiyq bal 8°02 vt cme cham es es
JeyO av z

=86
cl
gs
ge
Sk
ov
L'6
08
SZl
LL-6
Zk
uonjeonpy
601leer

or Nigecé
L6 0°62gse6

Sle6

e'€8ov (om)
=a= —

cl

LE

GL-EL
6S oesSezOL

SlrSrl

=
Ole AV
=)

LS 98LOEOLE bor

dnoid
vly8'Sl896 bie
zr eee0'Sz0'S2 LOL
sez

6S oesSezOL sez
6S oesSezOLt sez

Le yeeeregze 9°62
9€ 9Sv'0e8:92 9°92

Ly
9€
(PII)

rn
eS
oy
AV
~=9
gs
879
a1jsoubeiq
Z0L
uresg
zy
Aunfuy
Gelee,
=6S
(e}e19poW)9°62
oneuNned
gL
‘oedsun—Awojeqo7
1YH1y—Awojoeqo7 or v9l vs a8

uolsseideq
=
v'Sz
OCP
o'l2
OL
WSI|OYOo|y¥
W1U014D
16
eseesiq s,uo}buyuny PLL BLL £02 che 6 Lt 8 Lh Ze oer Zee BLL =

ye]-Awojoeqo7
S,JOHESIOY ZO vt vol 0°62 26c
GHaVv 89 08 OL OL OLL
yyeW-d7 Sy Lev
Bulpeey—q1 ve eee
(PII) HIN es
(a}e19poW) HIN e's
ZOL eseasiq s,uosuMled Sel 6 Lt 9'Le
elueiydoziyos 0'001 9€2 vb 8b 9'€2 Ze 9°82
e401S 1YBIy

YO
ainsodxy ulxojoiney GZ \'6 LOL LOL

A4ONS
 Q = oe = ¥ a = 2 on6 = 2ga = = Z= 72 = = x 5 a=5 % = = =
—
(panu1juod)
-_ Dp

Huluonjouny
1SOM1Z 2sgeel evh£8 LS86 Br 87 ab eesay ee eeeS'6 Lilys6
satpnyig

O-Aey
uostieduioD

le oesaZl Love6S 0001 LLLLe 962veecoe erylLe LLLZe vl vee6'SZ

Bulssa00/q
O10Paz ©
y ‘6h vl9°82 cs6 9E ge 9SE98929'°8¢ zZzege LL 6 Lebl
SS Lt Lt

BogosIW
gt Lvy0'02 sezSoZ 0'00L 6S o'essezOLL sez 6S oLtLL OLL 9'ES?
sainseayy
[eU19}xq/TII-SIVM

Bodossw
91 0°02
Lvov sezSOL 0001 6S oessezOLL sez 6S OltLL OLL 9'€?

Ay4ayxeq
pseogbag
peroosy
1Z LSE‘tt Cas9°82 S06S6 s'6 2s9°82S6 S'6 L8e o'6L ey By Bez

pue
Bujddel-seGui4
@AI}NIaX"
paeds
jeneds
aul4
J0}0W }s0L rd 8 V6L 2°89Sle $°S6oy Sp 98y9eLoe gelV6 6 6 zeL sy Sy BLE
vy
10 24}

ueyoy Le eee
saydures

180 Ze 89LOL 6'Sz7) S18ceasZe LLLa olelL ble 61Sbbb

yse,
Kuobayep
Hujwen
9k VS2st z'9Sser vL8e9 e9 e9 o'er88here 88h v'29 esl
jo ay}

ebenbueq
s1yde1Z0WIEg
ejeq

yseLuonejsossy
POM

Paljoujuoy Sh eleL29 L°066S 07oz eZ 6S eee9°61vee 6'eS86 6S 6S oz 961 6

1S 9°S9

ING £01 L6L

evs S'6rsos £68wa 61OL 8s 89 Ve62792 0'0€e's OVIOLiam
6} 26 aL £8 OL g'Z
—
a1ge] "9f"7

Aunfuy — eseesiq eseesiq ainsodx3

“oedsuj—Awo}99eqo7
wsjoyoo}y
UBOVEWYy dnoiy 1YBIY—Aw0j98q07
urerg yo
(eye19pow) (2ye18pOW)
eBluaiydoziyos
Ayojuyya/ecey y-Awo\oeqo7 suojuyuny Bulpeey-d7 1YBIy
uolsseideq SyOHeSIOy (PI) s,uosuped
ye7ulxojouneN
uoneonpy
ojuedsiy,
(PWN) opewnes
o1jsoubeig wew-d1
a1U0,YD GHOV
ajewe UBLJOY SI-€L e100
ex011S
ueey\ lew SHUM
LL-6 av Gv HW HW
oby as 8S ral 9l<
N |
xeS ‘ajoyy,
(Vy)
payode
eyep
pue
}deoxy
40)
ejdwes
ezis
‘sefej
se
aie
‘ofe

159
Evidence Base for Validity of the WAIS-III and the WMS-IIT

of range. The FSIQ and VIQ had moderate to high positive correlations
whereas the PIQ had low to moderate correlations with WIAT-II composite
scores. For the adult sample, the FSIQ and VIQ had high correlations with
reading and math and moderate correlations with writing and language
skills. The PIQ was highly correlated with reading in the adult group.
Moderate associations between the PIQ and the other WJAT-II composite
scores were observed in the adult group. In general, the correlation between
specific WIAT-II subtest scores and IQ scores were lower than the correla-
tions between the WIAT-II composite scores and WAIS-III IQ scores.
The results provide support for the use of the WAIS-III as a predictor of aca-
demic functioning. The results lend support for the use of a discrepancy
model of IQ compared to achievement based on the WAIS-III and WIAT-IL.
The correlation between the WAIS-III and the WIAT-II is similar in degree to
that reported for the WAIS-III and the WIAT. Clinicians should expect more
variability between IQ level and individual WIAT-II subtest performance
than is observed for the composite scores. College students may display
more variability in IQ-achievement discrepancies compared to adults in
general and high-school students (a discussion of discrepancy analysis
methodology is discussed in Chapter 5).

Measures of Cognitive Ability

Correlations between the WAIS-III and the other measures of cognitive abili-
ty are presented in Table 4.38. The external measures were the IQ scores of
the WAIS-R and the WISC-R, the total score of the SPM (Raven, 1976), the
Information Processing Accuracy Index of the MicroCog: Assessment of
Cognitive Functioning (Powell et al., 1993), and the total score of the
Dementia Rating Scale (DRS; Mattis, 1988). The mean WAIS-III FSIQ scores
for the WAIS-R, WISC-R, SPM, MicroCog, and DRS samples were 93.5
(SD = 13.5), 90.8 (SD = 18.1), 93.8 (SD = 16.4), 94.3 (SD= 13.7), andig0.9
(SD = 15.8), respectively.
As the data in Table 4.38 show, the correlations between the WAIS-III IQ and
Index scores and the external measures (except between the WAIS-III VCI
and the MicroCog) are moderate to high, ranging from .50 to .91. The corre-
lations between the WAIS-III and the WAIS-R range from .50 to .85. The cor-
relations between these two versions are highest when similar scales are
compared. The correlation between the WAIS-III VIQ and WAIS-R VIQ is .85,
between the PIQ scores, .82, and between the FSIQ scores, .84. Significant,
but relatively lower, correlations occur between the WAIS-III VIQ and
WAIS-R PIQ scores (.66) and between the WAIS-III PIQ and WAIS-R VIQ
scores (.54). The pattern of these correlations, though slightly lower in mag-
nitude than those reported in Table 4.18 for the nonclinical population, pro-
vides evidence of convergent and discriminant validity of the WAIS-III.

160
 -
ajqel Ley
* Uy suoneyers109419)
JO 2) II-SIVM
OI sa109¢
pure 24} []—LVIM 3sa}qng
pure a}1sodur0,)prepue}s
$9109
|I-LVIM SISOIQNS ||-LVIM SOUSOUWOD
3 s
a 2. a Fr
S E 5 a
> aH ro) ° © Co)
ry oO = =
ha eit 2s 3a g 2 s- 5a e
=|
= fo) 3
s a =.
° az ° 8.
3 a
3 =oe 2ro}
a 4 ° a x io) © =o 2 =
= 8 >) = ra) m a = D a a
a gae 3a a
=| | Ss Ss a 3 | 5
a 2 2 ay 2.
= § =2S$
a© —m 52. ®- ®= =. =
IIFSIVM = E 3 3 cs = = 3. Py S = = 3 =
z= ) o 0 ‘ ° - ° st
seby 6L-LL YBIH) (0049S
OIS4 89° pov oy OL cS c8 Sg 08 0S0 al C8 CL L8 88°
DIA OL 69° GL G9 eg vl 09° 7c 970. 8L SL Lg BL G8
Old Lv’ v9 Lg 09° Ov gL LQ” 868 0s0 Ly GL 99° LE 6L
ued 66 EFF 19°86 9166 1866 9700L 1986 2986 €000! 0266 Y9E66 OV86 S986 v6és86 001 Ov
as HL ZB ZL9L OVOL ZE6vL ELvL L89L IG9l €0SlL 8y'vl 921 SOLL OL9L IQ€L LO'SL
ebajo9 sjuapnys
DIS4 ae pare BP cOGe Gor69° | Le 399° Ovo €9’ a rn rn)
(4: LL
DIA 7a, ae Goes Gr = "iy 29 88 99 pe0 99’ 99) tr IS gg’
Old (Ase Gg cv 0g 6c 6S 8e° 8 660 Cv £9 6° Lg” ice

ues EBLOL ELZLL S¥90L LLOLL 9Z'ZOL 9EGOL OO'OLL 78°90L Ov'vOL 6y'SOL 9EZLL IGLLL vy'90L 97601
as 196 GOLL O9OL GZOL 8rZL POOL vOTL ILLL EVEL ZOLL 7@ZL L6vL Q9LZL LL LB
siinpy
DIS4 99° C8 oF 99° 69° 8L 69 el 6vO0 98° C8 €9° Lg 98°
DIA ae er See /8.= Zo a Qe O09 89 0 8" OD” @9t—éi= gg gg’
Old ee re WEG LSS LQ G9 8699S 9 v0 oy 8g" IG Lg tL
ues Or'66 92°00! 08°46 6S°00L 8S'70L OS'°00! 91°ZOL S866 vZ86 80°00L 12001 LOL €€ l€86 LOLOL
as Secy
a bo CSO VG0ISy VO
| 90'S) AGLVLE VL SI GES) heal
I LEGS
A 1067 Cly VE VL 6L

161
Evidence Based on Relationships With Other Variables
 ajqgel “8e"7 SUOTRIIIOD
UsemMjog
2) ITI-SIVM
Puke 19YIO SoINSvI||
jO dAnIUsOD
AVWTIGV

162
0J91\NHoy
UOH}EWIOJUY
u-SIVM
WdS Buissed01dg sua
OIA Old OISS ueeN GS DIA SL urea gs Adoeinsoy ueay gS SL ueaN
as

Old
Y-OSIM
DISA
III-SIVM
DIA ae YES) SPIN Oley BL G6 c9Ol 6S 686 E€E€L 6S v96 Zvi

ol
Soma
Old SE COMM nc OG ei 72See SS €8° v06 cSt eee 688 GEL 8S ecg Zeit

06
68
OlS4 OSLise Ghee) IE SMS Moy v8" 8&6 VOL 69° [Link] 4S OMe Ha GiOG 8'Sl

83°
IDA Os OOM eee eno COcr Ola OCigee Bye§— AOL. VAM, 8c tE0L trl 6S G86OVI
lOd LG yk OL este) tab Oy vs 6€6 GSt vL bares} teak fe Arse} 9'El

iy

Sisy
WM LQ Sih ish {sits vas = LQ’ 928 O91 19° chore
SSS| Vi06e
Zh

[6 COMmmCLS
ISd IS Rep 1S ee vs v8 Belt 99° YQ9P8 ¢9lL 9S 618 GL

Le
LZ
uel <c86 9€6 656 0°86 L°8¢ 8°88 Let6
ads ool ESE cet ve 6 VEL BOL Gel
N 88 88 88 ve oo

S86Sve
OL vl

186 Ob~e
ve
=e
Evidence Base for Validity of the WAIS-III and the WMS-III
 Evidence Based on Relationships With Other Variables

A similar pattern of correlations, yet with higher magnitude, is found

between the WAIS-III and the WISC-R (ranging from .71 to .91). The magni-
tude of these correlations might have been influenced by the composition of
the sample (e.g., 16% Mental Retardation and 75% Learning Disability) or by
the smaller sample size.
The correlations between the WAIS-III VIQ, PIQ, and FSIQ scores and the
SPM are .78, .83, and .84, respectively. These results are consistent with those
for a nonclinical sample reported earlier in this chapter and with results of
research reported in the literature. In addition, the correlation between the
WAIS-III Matrix Reasoning subtest and the SPM was .79, which is also con-
sistent with the results from the nonimpaired sample.
For the most part, the correlations between the WAIS-III IQ and Index scores
and the MicroCog are all in the moderate range; the only exception is
between the WAIS-III VCI and the MicroCog (.28). The correlations are in the
expected range and are higher than those obtained between the WAIS-R and
the MicroCog (Powell et al., 1993). Similarly, moderate correlations (ranging
from .55 to .61) were obtained between the DRS Total Score and all of the
WAIS-III variables.

Measures of Attention and Concentration

The correlations between WAIS-III and a variety of attention and concentra-
tion measures were obtained from various clinical groups. The external
measures were the Attention/Concentration Index of the WMS-R, the Trail-
Making Test (Trails A and Trails B) from the Halstead—Reitan Neuropsycho-
logical Battery (HRNB; Reitan & Wolfson, 1993), and the Attention/Mental
Control Index of the MicroCog (Powell et al., 1993). The correlations, means,
and standard deviations for each measure are shown in Table 4.39. (See
Table 4.36 for the demographic makeup of the samples.) The mean WAIS-III
FSIQ scores of the samples who took the WMS-R, the Trail-Making Test, and
the MicroCog were 87.3 (SD = 17.5), 90.0 (SD = 11.8), and 94.3 (SD = 13.7),
respectively.
Because the WAIS-III is not a comprehensive measure of attention and con-
centration, moderate correlations were expected. Of the WAIS-III indexes,
the one most related to measures of attention is the WMI, and, therefore, the
higher correlations were expected with this variable.
As shown in Table 4.39, the WAIS-III IQ and Index scores have low to moder-
ate correlations with the Attention/Concentration Index of the WMS-R
(ranging from .36 to .46). As expected, the exception is the WAIS—III WMI,
which has a much higher correlation with this WMS-R index (.66). The mag-
a
nitude of this result is partly due to the fact that these two indexes share

163
 3j/gel "GE"7 SUOHPLIIIOD
Usemjog
24) I-STVM
[J PUB Soinseay]
JO UOT}U9}V
PUP uOr}eNUIIU

164
Bupyeyy-!e4L
yso) Hod0J911\
jequa;/uolusny
as SielL
=v sie,
G as

u-SIWM
|01]U0DXapu| ueal as

ues

ueay
III-SIVM

Xepu|
/uonuspy
DIA ov Gil 6'86 eel

6ZI
Old ov"

ZOL
19° 6'88 Sel

UO!}e4}Ue9U0D
covl
DISA

LGi-oS
ov"

eG’.G9".
el €'r6 LEh

8
Lh

O'91si
IOA 9g"

9V'-
99'-
Sr’ L€OL vl

LOL
lOd ov” 8S" L'26 cma

-
IWM 99° cio) 628 SN

9°€6\Z8 0°066°S68°06

6 iS
vl GIL
ISd 8e" 09° 98

S68
9kco

6vl
Or’-c9 g9- cis
Oth VclLE}vst

Z286°88€28616L°c68°06¢S8
cl8

cl €Gi-Lee 67'-
ueel\| 0°26
as BEL

LEIS)
c9Si 86
N SS
9289'8L

G'9¢9S
QI
Evidence Base for Validity of the WAIS-III and the WMS-III
 Evidence Based on Relationships With Other Variables

common subtest (Digit Span). The other external measures (Trails A, Trails B,
and MicroCog) generally correlate higher with the WAIS-—III WMI than with
the other WAIS-III IQ scales and indexes. The WMI correlations with the
external measures range from -.37 (Trails A) to .66 (WMS-R Attention/
Concentration Index).
The differences between the WAIS-III WMI correlations with the Trail-
Making Test (Trails A and Trails B) provide additional evidence of construct
validity. For Trails A, the correlation with the WMI is -.37. However, when the
complexity of the task is increased, as in Trails B (for which working memory
is needed to a much greater extent), the correlation between these two vari-
ables is much higher (-.65).
For all three external measures, correlations with the WAIS-III VCI are rela-
tively lower. All of these results provide evidence of concurrent and discrimi-
nant validity.

Measures of Memory
The correlations between the WAIS-III and a number of memory measures
were obtained. The external measures were the WMS-R memory indexes,
the CVLT (Delis et al., 1987), the Rey—Osterrieth Complex Figure Test (Rey—O;
Rey, 1941, 1959), and the Memory Index of the MicroCog (Powell et al., 1993).
The correlations, means, and standard deviations for each measure are
shown in Table 4.40. The mean WAIS-III FSIQ scores for the WMS-R, CVLT,
Rey-O, and MicroCog external samples were 99.2 (SD = 12.8), 86.8 (SD =
10.5), 85.5 (SD = 11.5), and 94.3 (SD = 13.7), respectively.
The WAIS-III IQ scales and indexes were expected to show low to moderate
correlations with most of the memory measures. Higher correlations were
predicted between measures with the same presentation modality. For
instance, the WMS-R Verbal Memory Index was predicted to have higher
correlations with the WAIS-III VIQ and VCI scores than with the other IQ
and Index scores.
Correlations between the WAIS-III IQ and Index scores and the WMS-R
Index scores range from .33 to .77. The VIQ has a relatively high correlation
with the Verbal Memory Index (.71) and the Visual Memory Index (.73),
whereas the PIQ has a higher correlation with the Visual Memory Index (.65)
than with the other indexes. These results are similar to the pattern obtained
between the WAIS-III and WMS-III for the normative sample. Similarly, low
to moderate correlations were obtained between the WAIS-III and the CVLT
scores. However, as expected, relatively higher correlations (.58) were
obtained between the WAIS-III VCI and the short-delay recall and long-delay
were
recall variables of the CVLT. Much lower, almost negligible, correlations
and the memory component s of
obtained between the WAIS-III IQ scores

165
 3142] “O77 SUONRIIII0
UsamMj
) 2U)og TIT-ST
pueVM Soinse
JO ay] AIOUIIIN aa)>

166
u-SWM 19 ne re ceno-Aey el see Bogovo1yen 8
sel
= wous Huo7 aauig
= «=wuy peArjag Aioway 9
jeque, jensi, jesauay petejeaq ues, GS S-| Aejaq Aejeg ueay gs Adop waw wey ues gs xepu; ues) Gs eS
II-SIVM »
I
DIA Le el cL 9) FZ ZO0HE re FSi 9S) Lv sae OL iy 80'- ee SO OF} (ie Gey Sel 5
Old 67 GQ” GS ASR G26 VS 9c JEL 6d. AS OIL ie - OL ills Eis weik ier tk} Soh Ss
Olsd Lg ie EAS Lg 66 8cl ce 8c ev 898 GOL
6 8p [P= ole SSs Gli 67 rver Lele Ss
IDA LZ v9 Oe 9 GILOL S8cl se 8S" 8S S6 SHOES LG SO’ 10 806 O68 80 -eOlL vert ia
lOd ve vs Lv ge" G6 Si Cais Ol 0g G98 16 9v = (0) score Oicem=hE 6cSS LS come SULTS s
NM eS) 9S" OV ve 66 ELC}60rUC« co 60° io) €98 ©8rl 9c Sic fro= 68" ESE Si 628) ZLEk =
ISd or" gs’ oS 67" METAS Sei Sy [Moye oo Lees ci On Cais Cae Zeee EZ vG 9B c9l =
ueds|\ 2886 LO2c S865 062 0'0€ CG ve- 8'8¢ Z8 9'9 v'68 a
gs gist 26 Lié 9'°9¢ LSE bh bh G9 fae: 99 v0¢ a
N ts Bey LE LE 61 6} 6h ve ve ve 9} &

‘ajo104 aU} {TAD

B J a109S
Sem PasnJO} BU} S[eL]G—| ‘a|qeUeA
pukeZ sasoosalam pasnJO} ey} -YOYspue Aejap-BHuo) ‘sejqeuen
Ge —a ® Z
=
 Evidence Based on Relationships With Other Variables

the Rey—O. These results are predictable because of the complex nature of
the Rey—O tasks. The Direct Copy component of the Rey—O is moderately
correlated with the WAIS-III IQ scores and is especially related to the
optional procedure, Digit Symbol—Copy (r= .51). Finally, moderate
correlations were obtained between the WAIS-III variables and the
MicroCog memory scores.

Correlations between the WAIS-III WMI and the other measures are much
lower, ranging from .09 (CVLT short-delay recall) to .40, with the exception of
the correlation (.56) with the WMS-R Visual Index. The magnitudes of these
correlations tend to be lower than the correlations with the WAIS-III VCI
index and are commensurate with the hypothesis that working memory is
different from and significantly independent of learning efficiency, episodic
memory, and delayed recall.

Measures of Language
The correlations between the WAIS-III and measures of language, along with
the means and standard deviations for each measure, are presented in Table
4.41. The external measures were the total score of the Boston Naming Test
(BNT; E. Kaplan, Goodglass, & Weintraub, 1983), the FAS Total Score (FAS TS)
of the Controlled Word Association Test and the Token Test of the Multi-
lingual Aphasia Examination (Benton & Hamsher, 1994), and the Animals
part of the Category Naming Test (Morris et al., 1989). The WAIS—III VIQ and
VCI scores were expected to have the highest correlations with all of these
measures; correlations with all the other WAIS-III IQ and Index scores would
be low to moderate. Also, the WAIS-III verbal indexes were expected to cor-
relate more highly with the external language measures than would the
WAIS-III performance indexes. The mean WAIS-III FSIQ scores for the BNT,
Controlled Word Association Test, Token Test, and Category Naming Test
external samples were 84.5 (SD = 13.9), 88.4 (SD = 13.9), 84.3 (SD = 11.5), and
84.4 (SD = 7.9), respectively. (See Table 4.36 for the demographic information
for these samples.)
As predicted, most of the variables correlate highest with the WAIS—III VCI
and VIQ scores. The BNT has moderate correlations with all of the WAIS-III
IQ and Index scores, but its highest correlation is with the VCI (.48).
Similarly, the Controlled Word Association Test also has moderate correla-
tions with all of the WAIS-III measures, but its highest correlations are with
the FSIQ, VIQ, and VCI scores (.59, .61, and .57, respectively). The Token Test,
which measures an examinee’s ability to follow simple to multistep com-
mands without having to produce a verbal response, has higher correlations
scores.
with the WAIS-III VIQ (.62) and VCI (.59) than with the other WAIS-III
The Category Naming Test, a measure of semantic fluency, has its highest
correlations with the WAIS-III VIQ and VCI (.55 and .62, respectively).

167
 B|qe] Ly"
* SUONEIAII0Usamjag
9 AY) [I-SIVM
pur sainsvay]
Jo asensury

168
P2[04}U0D
PJOM
Ing uOeIN0Sssy
Sa UdyOL}S9] Aiobaje5Bulwey
jsay
SL uealy) as SLSV4 ueaN as SL ueajj) = sjewiuy
= ues) as
II-SIVM
DIA 8c €Z8 BEL 19° v'e6 Laa\ rach 168 ELL SS’ 8°88 98
Old ov 9'€8 Syl 8 L'78 GSI Ce 6°08 IGIl v0'- 918 S'6
DISA vy G'v8 6El 6S" 788 El6 ov €'r8 SLL 9¢° v8 6L
IOA By 706 Gel Lg" S'v6 ral 6S" 6'06 LL 6 rach 9°€6 eye
lOd ev G28 WB) Or 798 LEh ale 0'S8 Vth 20° 0°28 68
INM Ze 128 7Sh vs L'68 GSI ov’ 2S8 Ov Ae 708 L6
ISd ve 708 vv cio 628 8°S| ale OLL vol oil JEXSy) LOL
Ueda 89r 62L L'6€ ore
gs LL | 8g 6¢
N €0l LS Le 9|
Evidence Base for Validity of the WAIS-III and the WMS-III
 Evidence Based on Relationships With Other Variables

Measures of Fine Motor Speed

and Fine Motor Dexterity
The correlations between the WAIS-III and the following measures of fine
motor speed and fine motor dexterity were obtained: the average number of
taps over five trials for Dominant Hand and Nondominant Hand on the
Finger-Tapping Test of the HRNB (Reitan & Wolfson, 1993), time in seconds for
Dominant Hand and Nondominant Hand on the Grooved Pegboard (Lafayette
Instrument Company, 1989), and the Reaction Time Index of the MicroCog
(Powell et al., 1993). The correlations, means, and standard deviations for
each measure are shown in Table 4.42. The mean WAIS-III FSIQ scores for the
Finger-Tapping Test, Grooved Pegboard, and MicroCog external samples were
95.2 (SD = 13.5), 95.0 (SD = 17.2), and 94.3 (SD = 13.7), respectively.
The Finger-Tapping Test and Grooved Pegboard are measures of simple
motor functioning and eye-hand coordination. Dominant-hand and non-
dominant-hand scores are reported for each. These measures were predicted
to have low to moderate correlations with most of the WAIS-III IQ and Index
scores. The Finger-Tapping variables were expected to have the highest cor-
relations with the WAIS-III PSI and to be moderate in magnitude. The corre-
lations with the Grooved Pegboard were expected to be somewhat higher
because of the increased complexity of the task.
As expected, the Dominant-Hand score of the Finger-Tapping Test correlates
moderately with the PSI. Moderate, negative correlations occur between the
WAIS-III measures and the Grooved Pegboard dominant-hand and non-
dominant-hand conditions. The highest correlation is between the WAIS-III
PSI and the Grooved Pegboard dominant-hand condition. Moderate correla-
tions also occur between all of the WAIS-III IQ and Index scores and the
MicroCog Reaction Time Index (range .32-.85). This pattern of correlations is
likely because the MicroCog is more complex than the other external mea-
sures in this study.

Measures of Spatial Processing

The correlations between WAIS-III and a number of external measures of
spatial processing were obtained from several studies. The external mea-
sures were the Spatial Processing Index of the MicroCog (Powell et al., 1993),
the total score on the Judgment of Line Orientation (JOLO; Benton, Hamsher,
Varney, & Spreen, 1983), and the Direct Copy condition of the Rey—O (Rey,
1941, 1959). The correlations, means, and standard deviations for each
measure are shown in Table 4.43. The mean WAIS-III FSIQ scores for the
Micro Cog, JOLO, and Rey-O external samples were 94.3 (SD = 13.7), 85.6
(SD = 13.2), and 87.0 (SD = 11.8), respectively.

169
 Bjqe] “ZH SUOT}II1I0D
Usemjag
24} TII-SIVM
pue Sammsvayy
JO JUL 10)0]] psodspue JUL] 2

170
1010J] A}I19}X9q s
=

Buidde|-iabul4
1s0| paroosy pseoqbad Hoj0J911N\ 2
jueUIWOg }UPUIWOpUON jueulWwog JUBUIWOPUON uoljoeeY g
puey puey uealj as puey puey ueajl gs xXepujewiy
ues) as =
III-SIVM %
DIA LO yes 6'°S6 v6 Or’- QP". 6'€6 Zyvl SQ" 6'86 eel ic
Old 82" 90° 2'v6 GOL 6¢°- 6r'- 296 8'6L G9" 6'88 Gel 3
OISs co 20° 256 Gel v¢'- LG. 0'S6 cL 69" €'V6 Ziel >
IOA 20° Vhs 196 sill Ly’. ae 6'26 8'Sl oe" LeOL vv 5
lOd 60° S0- 0'v6 Oz Or'- 9r'- 9°96 BLL €9° 1°26 cma =
INM — a= as = = = = = Sg" 628 Lek =
ISd Lv 8 162 6h2 €9"- Ly’. L'v6 LL 6S" 9°78 291 2
|
UES 8°8E 9ZE 0'€6 6°c0l 628 =
as Lyk 6'LL ye LSS AX: =|
N 22 22 12 Lz 9} aan
s
o
:
 Evidence Based on Relationships With Other Variables

It was anticipated that WAIS-III IQ and Index scores would be at least moder-
ately correlated with the MicroCog Spatial Processing Index, because spatial
processing is related to overall intellectual functioning and this MicroCog
index contains components of short-term memory, attention, perception,
and spatial processing. A similar pattern was expected with the JOLO, which
is a relatively complex task. The Rey-O Direct Copy task involves predomi-
nantly constructional processes and perception and would be slightly less
related to the verbal scales. Therefore, the WAIS-III PIQ, POI, and PSI scores
were expected to show higher correlations with the Rey—O than would the
WAIS-III verbal scales.
Correlations between the WAIS-III PIQ, POI, and PSI scores and the Rey—O
are in the moderate range but generally higher than the correlations ob-
tained between the WAIS-III verbal scores and the Rey-O. Also, as expected,
the MicroCog Spatial Processing Index and the JOLO have moderate correla-
tions with most of the WAIS-III IQ and Index scores.
These results suggest that the WAIS-III IQ scales and indexes are related to
other measures of spatial processing. However, as the tasks become more
constructional and involve less reasoning, the correlations with the VCI and
WMI become lower.

Table 4.43. Correlations Between the WAIS-III and Measures of

Spatial Processing
MicroCog JOLO Rey-O
Spatial
Processing Direct
Index Mean SD TS Mean SD Copy Mean SD

WAIS-III
VIQ .63 98.9) 13.300 544.891 12.9 43 91.5 9.5
PIQ .65 88.9 13.5 .54 83.7 149 a0) fete) ere
FSIQ .67 O43 613.7 Wy n6ln 856: 13.2 fon) to abs}
VCl 41 NOS 14 Ae en O lek Olde ete. Pah GS Oke
POI 54 Sot 418 6465" 67.6 13.1 48 89.2 13.5
WMI BY 87.9 13.7 .56 873 13.5 pay ihe “i!a.s
PSI .63 846 162 47 804 14.9 4857867 18:3

Mean 87.4 24.9 29.3

SD 16.5 6.4 6.4
N 16 28 27
je EE

171
Evidence Base for Validity of the WAIS-III and the WMS-III

Measures of Executive Functioning

The correlations between WAIS-III and the Wisconsin Card Sorting Test
(WCST; Berg, 1948; Grant & Berg, 1948; Heaton, Chelune, Talley, Kay, &
Curtiss, 1993) were based on a sample of 21 participants. The WCST is a
neuropsychological instrument for assessing problem-solving strategies,
cognitive flexibility, and the ability to use feedback in problem solving. The
scores obtained on the WCST are number of correct responses, number of
categories completed, number of errors, and number of perseverations
(cognitive inflexibility or the inability to switch tasks and apply new rules).
It was predicted that two WCST scores, the number correct and number of
categories completed, would be more related to general intellectual func-
tioning (FSIQ) and the more fluid, processing tasks of the WAIS-III PIQ and
POI than to the other scores on the WAIS-III. These WCST scores were also
predicted to be less related to the WAIS-III tasks that require more crystal-
lized, acquired knowledge (e.g., the VCI). Additionally, because the WCST
also requires intact working memory (i.e., the examinee must keep track of
correct and incorrect responses to complete the task successfully), relatively
higher correlations with the WAIS-II WMI were also anticipated. The
remaining scores on the WCST, which are associated with impaired perfor-
mance, total errors and perseverative errors, were expected to have low, neg-
ative correlations with all of the WAIS-III indexes. The correlations, means,
and standard deviations for each measure are shown in Table 4.44. The
mean WAIS-III FSIQ score for the WCST sample was 85.9 (SD = 11.9).

Table 4.44. Correlations Between the WAIS-III and Measures of

Executive Functioning
WCST
Total Categories Total Perseverative
Correct Completed _—_ Errors Errors Mean SD
WAIS-III
VIQ 34 31 = 04 Te 90.7 We
PIQ 42 .39 -.19 23 82.4 13.2
FSIQ 42 .40 lil 26 85.9 11.9
VCl 33 .26 .00 ~.20 92.8 12.6
POI 45 .40 ees alte) 86.2 12.3
WMI .48 31 .08 silts 85.0 13.2
PSI .30 29 meall7g 23 78.4 11.3
Mean 57.6 1.6 53.5 34.6
SD Pi 2 ist 18.0 15.9
N 21 21 21 21
eS

172
 Evidence Based on Relationships With Other Variables

As predicted, the highest correlations are between the WCST total number
correct score and the WAIS-III FSIQ (.42), PIQ (.42), POI (.45), and WMI (.48)
and between the WCST category score and the WAIS-III FSIQ (.40), PIQ (.39),
and POI (.40). These correlations support the premise that the WCST correct
scores (which reflect an individual’s problem-solving strategies and flexibil-
ity) would be most related to the WAIS-III general intellectual functioning
and perceptual scores. Moreover, the WAIS-III WMI is most related to WCST
total number correct (.48).

WMS-Ill Correlations With

Selected External Measures
The correlations between WMS-III indexes and the external measures are
based on data from mixed samples of individuals who were clinically diag-
nosed with various neurological disorders. Table 4.45 presents demographic
characteristics for individuals who completed each external measure for
the WMS-III comparison studies. The data are presented as mean age and
percentages of the samples by sex, race/ethnicity, and education level.
The samples are ordered according to the type of external measure
(e.g., cognitive ability).
For these comparison studies, if multiple measures from one external instru-
ment were used, then the correlations for those measures are based on the
same individuals. In Table 4.46, for example, correlations between the
WMS-III and the WAIS-R and the MicroCog (Powell et al., 1993) are reported.
For the WAIS-R sample, correlations for the VIQ, PIQ, and FSIQ scores are
based on the same individuals. The sample for the Information Processing
Accuracy Index of the MicroCog is different from the WAIS-R sample. Relative
comparisons between the measures within each test can be made, but
because the samples for the WAIS-R and the MicroCog vary greatly in size
and are composed of individuals who have different clinical diagnoses, rela-
tive comparisons between the WAIS-R and the MicroCog may be misleading.
Therefore, to the degree that clinical group composition affects the observed
relationships, correlations from test to test may not be comparable.

173
 3]ge] "Gy dIydesSoWIag
eyeq
Jo ay} sajdures
10J dU} [PEUI9IXA/TI
sainseay]
-SWM uosteduroD
SaIpnig 2

174
eanjubo5Auuiay uonueny :
Kiowa 8
II-SIVM y-SIVM SBodo;INSHa Y-SIVM Y-SWM Gogoi Buryew-lre4,
ise) =«U-SIWM =ITAD) =BogoiN O-Aey ¥
N vO th 6 S38 SS vk pL LS ve
aby vb By “
g
uray Sly kes 8°09 S'9S o'er beg Z'0S er6 SVS
as Les 00S cet
9°02 961 St elt 6St 961 L'6L ool 89 961 zel
X@S 5
ayewe ely Lge >
L6z Z8e yg9e Se vss eee 62S Lse Zl |
ale Ls e719 739 9°0Z S19 e179 or LY9 Ly 19 88S AS}
umOoUxUN) ome G2 ra ‘1 02 =
Ayoiuyy3/e0ey >
SUM s'88 0001 9°88 188 s's8 0'001 828 z'98 ves 0°00 0001
UBD =)
UBILEWY- Lg OL WZ 60L L's el Lh =
oluedsiy 6 ek v2 ce vb oz 6% a
JEUIO SS
UMOUMUN) 672 SZ v2 81 Dg, (ore 6% =
uoneonpy
=
gs 6 gL 6S ro vb Or a
LL-6 es be: Gz [uve 601 ML ig 86 ee 6S =
ral G's els 622 L'Se 9'€% els LLe 912 Lp eZs oly —_
SI-€L 8°82 aa Lee €'8z G9e Zvl 8°62 zse 9°02 ran Sez 5
glz Srl eZ 9°92 ome} Lek Wo 912 st p92 ey Sez as)
umMoUuyuy) GLb ert EA LZ Lg ev 89 Let 6S ert 6S -
213souBbeig
dnoiy5
a
GV (PII) ev G82 Le 9 88l G82 Gl] Lyk S82 6S
AV (s}e18pOW) ey e9 ze Nee 6S StL =
oewnes
uresg Aunfuy 8s [i Gs Ls 6'e OL 2
uolsseideq (oie |
ye7-Awoyego7 es ev 9'€z ev vs Sez er =
‘gedsuj—Awojoeqo7]
om BL lore a
1YBIy—Awojoeqo7
Z9 6S vOl Ly Ll
91U014YD WSI|OYoo|y hh ev €'Sz 282 se rh eikerd er
s,uo}bujuny
eseesiq 96 ert O6L [EZ cme evh Sel 0% Sez er ral
S,JOHeSIOY Z8 eh v'6 vol OLL Gee
GHaV 8S
yyeW-a7 ge
Hulpeey-d1 62
a\dhinw SISO1a|9S se OLL €'8e
S,UOsUulyJed
aseesiq Z8 Lek cha Gel OLL
elueydoziyos 9°82 el 9°82 vt (Wire 9°82
ulxo}OJNEN einsodx3 9'°0L

YO]
Ge 16 rae 86

O}01S
a a er
 Es =
=| = 3 PAz i = o = 5 2 i)=Fa = ro)= 3) 4 < = —
=
(panuyjuod)
M2]

pieoqbag
perooy
6SEELL o'so'sz 9'6L7st Ze ee Olye'9€ 6 SY e9eezQ'EL ele ez ve
vb
sarpnyg
uos|reduroD

paeds =
Guiddey-sabul4
Ayayxaqg
Jojo so] asp9°02 z's98've £56ev B'veSer Z8 yi?)£8 i:) vLt SOE
ez

pue
aul
sanseayy
[PUI9}XA/TII-SINM

609091
bes961 LSEe'”9 0'001 VL elsZ'vh e'vk ert ev 9°82
vb

Bujuojouny
dAI}NIEXZ
LSOM £99Gel oel028 028Z8 ey ey s‘2s98 ev 28 £8 Ser£8
€z

O1or 6'SSL'02% o'szosZ 9€ £'6€Lee Z0L LL ZeeVL

VeeBLL
82

‘soeBejusoied se poyiodas aie eyep ‘abe pur (jy) azis ajdwes 10} }deoxy ‘ajoyy
Bulssed0/q
Bogoiw
jeneds L's961 LSEe”9 0'001 bL eLs2b er er ev 9°82
105 9}

ph
safdureg

O-Aey 0001 89 ve OleLye ve 69 0 ve 69

srseel ly98S
62
Jo ayy
s1ydeisoulIg

jseL
PJOM
eye

uoeossy €9 ce
abenbue
P2]|osjU04 9°69
a elyLs 22)S6 ot Ze S6 ze Eigarea S'6 6 91 LLL6'St Ze
€9

ING €02 S'6b

emete) S'0S 328GL 60 6L 9S 9S BLE662 gS s‘9 O'vh60ra61 60 v8 60 GL
204
a1qel “GH"y

Aunfuy
“oedsun—Awojoeqo7 eseesiq ainsodx3
WSHOYOO|y
UBOWeWY- dnoiy qyB1y—Awojoeqo7 SISOU@]OS
uresg ye7-Awoyoeqo7
Ayoiuyya/ecey (eye18poW) elueydoziyos
Bulpeey—-a7
s,UOSUDMed
oysoubeig uolsseideg HO1Ulxo}OuNEN
uojeonpy
uMOUUP) UMOUMUN onewnes weW-1
ayewe4 DIUOIY4O e/dhiNW B40NS
UROL
SUM JOUIO SI-€L GHav
uray ae LL-6
aby as 85 ZL GV
N uMOUyUN)
OL ont
oluedsiy
61
OL
ev 91z
Siz
sez
£02
bL
o'sz
0'92
ie
O'eL
ebk
° x)
av
(PII)
Ole
0'9r
ZLt
s'8z
Ze
v0
S82
yLt =aseesiq
ev
er
s,uojbuquny
e6=S,HOHESIOY
ole
e'vh
ez
ere
ey
GL
a1
xeS

175
Evidence Base for Validity of the WAIS-III and the WMS-III

Measures of Cognitive Ability

The correlations between the WMS-III and measures of overall cognitive
ability, as well as the means and standard deviations for each measure, are
presented in Table 4.46. The external measures were the VIQ, PIQ, and
FSIQ scores on the WAIS-R, the Information Processing Accuracy Index of
the MicroCog (Powell et al., 1993), and the total score of the DRS (Mattis,
1988). The mean WAIS-III FSIQ scores for the WAIS-R, MicroCog, and DRS
external samples were 94.7 (SD = 13.2), 91.3 (SD = 13.2), and 90.9 (SD = 15.8),
respectively.
Because of the moderately high correlations between the WMS-III indexes
and WAIS-III IQ scales and indexes for standardization participants, correla-
tions of the WMS-III with other measures of general cognitive ability were
expected to be in the same general range (i.e., in the .50s and .60s between
the WMS-III General Memory Index and overall ability scores) or slightly
higher because of the relatively wider range of scores typically observed in
clinical populations.
The correlations between the WMS-III indexes and the Microcog and the
DRS are in the moderate to high range, as expected. In general, the correla-
tions between the WMS-III indexes and WAIS-R IQ scores are in the moder-
ate range and are somewhat lower than expected. The descriptive statistics
and clinical group composition for the WAIS-R sample provide an explana-
tion for the relatively lower correlations for this sample. First, relative to the
mean IQ scores, the mean scores for WMS-III indexes are generally 10-15
points lower. Second, the WAIS-R sample is composed of about 23% of indi-
viduals with Korsakoff’s syndrome or mild Alzheimer’s disease. These two
groups usually present more memory loss relative to IQ decrements.
Consequently, the relatively lower correlations of the WMS-III with the
WAIS-R compared to those with the MicroCog and the DRS are expected
and provide evidence of divergent validity of the WAIS-III and the WMS-III
for this clinical sample. When the participants with Korsakoff’s syndrome or
Alzheimer’s disease were omitted from the sample in a follow-up analyses,
the correlation between the WMS-III General Memory Index and WAIS-R
FSIQ score was in the expected range (r= .51, p< .001, n = 80).

176
 177
E4 vidence Based on Relat ion ships With Other Variables

62 vA vOl N
Lvl 6 Vk Létk 9k 6 LL gs
6 9d1 98 296 L€6 216 uesl|
6S 8S ce 6c 0c JEASIOY
Be 8e° Z0° €0° cl UO!}US}SY
LY’ Bl Gz’ 12 Gz’ adojs Buweey
99° vs" vy Ly’ clon Huiusee7 jeu -aj6uis
sayisodwoy ssed0ijqg Aioyipny
est 0'S8 LQ” ESie 88s GS 9S} 616 ev vy ee Mowe, Bupon
02? BLL gg 97 vO08 eg £02 208 LE 9e° ce Nowa eseues
8°02 9'r8 €9° voc 128 09° O1Z 998 ee 62° og —spaAejaq uomuBooey Aioylpny
69 SL LS eet tes ee 691 O18 6e° ov Ley pehejeq sensi,
ng

Lk? 0'€8 Ach VoL 18 OV 86h Zv8 82° 92° Go pehejag Aioupny
ee LL 09° Let 608 Ach Gl er le oy Ly’ 8c" Kiowey eyeipew|
LOL 62 1S” 06 928 OL G9} L28 ee ve 60° eyelpewW| jeNsi/,
Lod €€8 Ace OSt S'S8 Lv 681 0°98 ov 6o° 8c" eyelpewutu| Aioyipny
III-SINM
as ube SL as ueeay xepujAoeinsoy gg ueol| diss Old DIA
Hulssad0lq
UOIJEWOJU]
Sud 090,91 u-SIVM
AVTIGy eANTUsOD Jo sainsvay] pure [II-SINM 24} Usamjog SUOTLIALI0D “gy a]qel
Evidence Base for Validity of the WAIS-III and the WMS-III

Measures of Attention and Concentration

Correlations between the WMS-III and several measures of attention and
concentration were obtained for various clinical groups. The external mea-
sures were the age-corrected scaled score on the Digit Span subtest of the
WAIS-R, the Attention/Concentration Index of the WMS-R, the Attention/
Mental Control Index of the MicroCog (Powell et al., 1993), and Trails A and
Trails B (time in seconds) of the HRNB Trail-Making Test (Reitan & Wolfson,
1993). The correlations, means, and standard deviations for each measure
are presented in Table 4.46. The mean WAIS-III FSIQ scores for the WAIS-R,
WMS-R, MicroCog, and Trail-Making Test external samples were 93.3 (SD =
15.0), 97.9 (SD = 14.6), 91.3 (SD = 13.2), and 90.0 (SD = 11.8), respectively.
The WMS-III Working Memory Index was predicted to correlate higher with
the external measures of attention and concentration than would the other
WMS-HI memory indexes.
As the data in Table 4.47 show, the WMS-III Working Memory Index general-
ly correlates much higher with the external measures of attention and con-
centration than do the other WMS-III indexes. The correlations between the
WMS-III Working Memory Index and the external measures range from .48
(WAIS-R Digit Span) to .85 (MicroCog Attention/Mental Control Index).
Additional evidence of validity is provided by the pattern of correlations
between the WMS-III Working Memory Index and Trails A and Trails B of the
Trail-Making Test. The correlation between the WMS-III index and Trails A is
-.31. However, when the complexity of the task is increased (and working
memory is needed to a much greater extent), as in Trails B, the correlation
between these two variables is much higher (-.62). This pattern is similar to
the one found between the WAIS-III WMI and Trails A and Trails B.

178
 179
Evidence Based on Relationships With Other Variables

vl vl SS Sg N
y88 €°S¢ vst 6 EL 6€ as
Lert 8°cS 228 0°46 9'El ure
8r'- 8l’- LE 92° Go JBASUIOY
Lv’- Z0'- 9} '- 60° ek UO!}UB}OY
9¢'- Z0'- 9L’- 9° rol adojs Buluee7
8r'- ro)ie ie) ve Se’ Huiuee7 jeu -aj6uis
seyisodwog sse901q Aiojipny
FS) B68. c9'- Loi ne eee Bae S83 7 Sle ab 26 BZ: St B06 “Sr: Nowe Bupom
G8t S08 7¢G- ve- Obl v08 Le 802 7298 92° vee vis OF Mowe j2s8ue5
Z8i 988 £GY- oC, © P0cm Ze ey Gla 1'd6 gl ez 698 ze pehejaq uonlubooeYy Aioypny
Si VSL © BF- com Gee es ANS y9L €98 yal Fer GOS Se" pefejag jensi,
6L GS8 9F'- St | Mele cs Zt vie +88 ce Sec 998 82 pefejeg Aoypny
SLE s8'6L~ sY- Sh Le earth 08 49° O6+ 98 0g €ec cl8 8c Moweyy ayeipeuwuu|
evt 662 Ly’- to == + OG 9°28 CL 8St 298 39 VLE GI8. 02 sye|powWwW| [ENsi/\
cst 98 €7- 60- OSL GSB 7s O61 606 Ze; Cle. E285 aS eyelpewiu| Aoypny
IIIFSINM
as ueew gst, Vsilel gS ure xepu| as ube xepu| as ueey ueds
jO4]U0D JejUN/\ uoljesjuadUu0D u6iq
/uoKuayy /uojushy
180) Bupyew-esL HogoJ9IW Y-SINM Y-SIVM
UONeIUIIUOD pue UOTJUA}V JO SaANsvay] PUL []I-SINM 94) Usamjog SUOTeIA1IOD */H%"y BIGeL
Evidence Base for Validity of the WAIS-III and the WMS-III

Measures of Memory
Correlations between the WMS-III and a number of other measures of mem-
ory were obtained from several clinical groups. These external measures
were several indexes of the WMS-R, the CVLT (Delis et al., 1987), the
Memory Index of the MicroCog (Powell et al., 1993), and the Rey—O (Rey,
1941, 1959). The correlations, means, and standard deviations for each mea-
sure are presented in Table 4.48. The mean WAIS-III FSIQ scores for the
WMS-R, CVLT, MicroCog, and Rey-O external samples were 99.2 (SD = 12.8),
86.8 (SD = 10.5), 91.3 (SD = 13.2), and 83.1 (SD = 5.6), respectively.

The WMS-III indexes were expected to have moderate to high correlations

with other memory measures. It was also anticipated that the WMS-III
Working Memory Index would have lower correlations with the external
memory measures than would the other WMS-III indexes. Finally, because
the Direct Copy condition of the Rey—O is not a memory task, the correla-
tions with the WMS-III indexes were expected to be very low or negligible.
The correlations between the WMS-III Working Memory Index and the
external measures range from .07 (CVLT Long-Delay Free Recall) to .42
(MicroCog Memory Index) and are generally lower than the correlations
between the other WMS-III indexes and external measures. The WMS-III
auditory memory indexes have relatively high correlations with comparable
WMS-R, CVLI, and MicroCog measures. As expected, the WMS-III visual
indexes generally correlate lower with the CVLT verbal memory measures
than do the WMS-III auditory indexes (the exception is the correlation
between the visual memory indexes and the CVLT Long-Delay Free Recall
score). However, the correlations between the WMS-III auditory indexes and
external measures with visually presented material (e.g., WMS-R and Rey-O)
are nearly as high as those of the WMS-III visual indexes. Unexpectedly, the
WMS-III Visual Delayed Memory Index is unrelated to both the immediate
and delayed measures of the Rey—O. Overall, however, these results are evi-
dence of the convergent validity of the WMS-III. For the indexes with visual-
ly presented material, these results are equivocal with respect to convergent
and divergent validity of the WMS-III. It should be noted, however, that
none of the external measures is generally recognized as representing a
“pure” visual memory construct.

180
 J1GeL "Ohh suONRya1I05
Usamjog
9} [II-SINM
pur 19YIO SBANSvaI]
JO AIOUIa|]
Y-SINM LAD Boj0191/ O-fey
sjeuy yous Buoy Mowe ypaig ww eq
jequa, jensi, jelouseq ueaW gg ¢-| Aejaq Aejeq ueey gs xepuy uray gs Adop wey wey ueay as
III-SINM
pny ww] GL €9° €L «OSL OOL6 6k IL es Or Sic8s BLE LL SSB OSL Z0- aCe: 6S O72" PLES
SI/\ WU] Lg 0s 9g 79 G98 LS Ze ve MEP GSLa Gil os 98 O06 €co- Lacs: ames! Ae Coe
WLU Wea) GL y9° el’ BL’ 26898) «6(0'6L 99 gr 2Gr Sipe Ol el G08 Eke El ecg coh =e ~6:99 Sele
pny l9q OL 19° OL eee. [Link] Lem0 VLi €9° 0s’ V8 cLl 4S 38
} cl £0 G= lo els G8l
SIA 18d (Ac gs 8s” 89° 198 G9 Ob br’ 9F 6bL cok Oe" VER Cel, Gls BYE Becce €6o cOk
pny 98 [8d 6L 79° 7s sz 2£¢6 ke OS by er ¢s8 est vs 128 02 60- ord: 8S 268 G0e
UBS) Wa/\| pl 19° bl 6&8 998 ale 99° 09° ~Es: Gos GiGh= ss 08 QOL 90- mrs: sla EL Ook
BM wal eo 60° 60° ee e926 Pek ee 92° 20 998 cpl ov «8°88 LEL
=O ge Bet ie = go cot

Asoyipny ssas0jq seyisodwog

-aBuls
jeu wee] pL 19° 89° bL’ OL 6r Go 6S 0° =«6r os
Buiwee]adojg €9° 6S" 99° LL or 9¢° Sy Go Lo- 8= ty
UO!US}9Y gS” on S" 99° es Ach ee Lh JAE St es
JEASIOY el G9 GL 18° 99° 1g 0s ve’ €0'- acd: oor
ues O06 €86 1&6 928 LYE be- Ae toa 9°S8 £'6¢ aS oe
pi
gs h Zaz Se SiS iSike veces Syl ol ol 8'8l gS +9 GS
N LS ve vi Zt

‘a]ON104 eu) ‘1TAD

B / a109S SEM pesnJO} SU} S/EU|G—| ‘a|qeUeA
PURZ Sa10Isa1am pasn10} au} -LOYspue Aejap-Buo| “sa|qeuen

181
Evidence Based on Relationships With Other Variables
Evidence Base for Validity of the WAIS-III and the WMS-III

Measures of Language
Correlations between WMS-III and the total score of the BNT (E. Kaplan et
al., 1983) and the FAS Total Score (FAS TS) of the Controlled Word Associa-
tion Test (Benton & Hamsher, 1994) were obtained. The correlations, means,
and standard deviations for each measure are provided in Table 4.49. The
mean WAIS-III FSIQ scores for the BNT and Controlled Word Association
Test external samples were 88.7 (SD = 13.1) and 88.5 (SD = 14.0), respectively.
The WMS-III indexes were predicted to show low to moderate correlations
with the BNT and Controlled Word Association Test. Further, the WMS-III
indexes based on auditorily presented material were expected to correlate
more highly with the external language measures than would the WMS-III
indexes based on visually presented material.
All of the correlations between the WMS-III indexes and the BNT are in the
low to moderate range. These correlations range from .25 (Working Memory
Index) to .39 (Auditory Immediate Memory and Auditory Recognition
Delayed indexes). Correlations of similar magnitude occur between the

Table 4.49. Correlations Between the WMS-III and Measures

of Language
Controlled Word
BNT Association Test
TS Mean SD FAS TS Mean SD

WMS-IIl
Auditory Immediate .39 78.4 17.2 38 78.2 18.2
Visual Immediate ei 78.5 Oa 21 75.4 14.5
Immediate Memory 38 73.8 Lae 33 71.8 18.9
Auditory Delayed 31 77.4 17.6 29 76.6 17.8
Visual Delayed .30 77.0 15.8 .29 ot 14.8
Aud Rec Del .39 79.7 19.7 24 tie 18.4
General Memory 36 TRS 18.7 .30 70.9 18.5
Working Memory 26 88.1 16.4 .50 85.5 15:6
Auditory Process Composites
Single-Trial Learning 37 .50
Learning Slope 43 14
Retention 16 18
Retrieval 38 .27
Mean 46.8 31.0
SD 11.0 13.6
N 107 63

182
 Evidence Based on Relationships With Other Variables

WMS-III indexes and the Controlled Word Association Test, ranging from .21
(Visual Immediate Memory Index) to .50 (Working Memory Index). Although
the trend is not particularly strong, the WMS-III Auditory Immediate Mem-
ory Index correlates more highly with the external language measures than
does the WMS-III Visual Immediate Memory Index. This trend, however, is
not consistent for the corresponding WMS-III delayed measures.

Measures of Spatial Processing

Correlations between the WMS-III and external measures of spatial process-
ing were obtained. The external measures were the Spatial Processing Index
of the MicroCog (Powell et al., 1993), the total score of the JOLO (Benton et
al., 1983), and the Direct Copy condition of the Rey—O (Rey, 1941, 1959). The
correlations, means, and standard deviations for each measure are provided
in Table 4.50. The mean WAIS-III FSIQ scores for the MicroCog, JOLO, and
Rey—O external samples were 91.3 (SD = 13.2), 88.7 (SD = 12.3), and 87.0
(SD = 11.8), respectively.

Table 4.50. Correlations Between the WMS-III and Measures of

Spatial Processing
eee MicroCog se abs _OLO
- Rey-O
Spatial Processing Direct
Index Mean SD TS Mean SD Copy Mean SD

WMS-III
Auditory Immediate 57 8515" 715.0.9 309 584.05) 16:9 1 75.8) 16:3
Visual Immediate 71 826 9:0 35 783° 14.1 707 70.5 7.6
Immediate Memory .69 BO Cmca! MeeOO tiycoe eeticO 06 67.6 13.6
Auditory Delayed Oi B20 ie.4— 298 80.5 17.2 OF Sin "18.0
Visual Delayed 47 SSG) 372 /5:67 4S 01 68.7 9.4
Aud Rec Del 45 8721 -920.4.17850" 82:9 98.4 O05) 1es0.08 819:6
General Memory 47 BO 454. Canate 5246 1753 .05 7102 woe G2
Working Memory 67 88.851 Bin 208 6804g 1657, OS ES27 6.0

Auditory Process Composites

Single-Trial Learning .60 .33 ard
Learning Slope .39 2h ~.04
Retention mailit 21 10
Retrieval 43 43 .06

Mean 87.6 25.9 29.6

SD ded 44 6.3
N 14 28 29
i

183
Evidence Base for Validity of the WAIS-III and the WMS-III

The WMS-III Visual Immediate Index and Working Memory Index were
anticipated to correlate higher with the MicroCog Spatial Processing Index
than would the WMS-III auditory memory indexes, because this MicroCog
index includes components of short-term visual memory and attention. Low
correlations between the WMS-III indexes and the JOLO and Rey—O were
expected.
As expected, the WMS-III Visual Immediate Memory Index and Working
Memory Index correlate more highly with the MicroCog Spatial Processing
Index than do the WMS-III auditory measures. Correlations between the
WMS-III indexes, except the Working Memory Index, and the Rey-O are gen-
erally very low. Unexpectedly higher correlations occur between the WMS-III
indexes and the JOLO. Because of these results, the JOLO total score was
correlated with WAIS-III FSIQ score, and a strong relationship was found
(r=.71, p< .001). Correlations between the WMS-III indexes and the JOLO
were recomputed, with the effects of FSIQ score controlled. All correlations
between the WMS-III indexes and the JOLO were nonsignificant in this
follow-up analysis. These results suggest that the relationship between the
WMS-III and the JOLO can be explained by a general cognitive decline to
which both measures are sensitive. Overall, these data are evidence of the
convergent and divergent validity of the WMS-III indexes.

Measures of Executive Functioning

Correlations between the WMS-III and the WCST (Berg, 1948; Grant & Berg,
1948; Heaton et al., 1993), along with the means and standard deviations, are
reported in Table 4.51. The mean WAIS-III FSIQ score for the WCST sample
was 86.7 (SD = 11.7).
The WCST is a neuropsychological instrument for assessing problem-solving
strategies, cognitive flexibility, and the ability to use feedback in problem
solving. The test requires intact working memory, because the individual
must keep track of correct and incorrect responses to complete the task suc-
cessfully. The WCST score associated with better performance (i.e., the total
number of correct responses) was expected to have low, positive correlations
with the WMS-III indexes, whereas the WCST scores associated with im-
paired performance (i.e., total errors and perseverative errors) were expected
to have low, negative correlations with the WMS-III indexes. Moreover, the
WMS-III Working Memory Index was expected to have the highest correla-
tions with the WCST measures.

184
 Evidence Based on Relationships With Other Variables

Table 4.51. Correlations Between the WMS-ILI and Measures of

Executive Functioning
—eSSSSSSSSSSSSSSFSFSMSFFFss
WCST
Total Total Perseverative
Correct Errors Errors Mean SD
WMS-III
Auditory Immediate is .06 m0 80.5 Wa O
Visual Immediate .09 18 04 alas 15.6
Immediate Memory pe 13 01 74.7 18.4
Auditory Delayed 31 -.09 = ake 78.9 18.1
Visual Delayed 34 .04 =:08 79.3 Wai
Auditory Recognition Delayed .33 -.18 meilly. 83.0 Ales
General Memory of 10 ~.14 74.4 20.0
Working Memory .60 2 -.01 82.0 22

Auditory Process Composites

Single-Trial Learning 41 ~.14 ae ig
Learning Slope .39 .06 ay We
Retention 10 We AO
Retrieval .34 Zils) 219

Mean 60.5 51.4 32.8

SD 22.4 18.8 16.4
N 23

As predicted, the WMS-III Working Memory Index has the highest correla-
tion with the WCST, but only with the total correct score. As the data in Table
4.51 show, there is a general trend of low, positive correlations between the
WMS-III memory indexes and the WCST total correct score and low, nega-
tive correlations between the WMS-III and the WCST total errors and perse-
verative errors.

Measures of Fine Motor Speed and

Fine Motor Dexterity
Correlations between the WMS-III and the following measures of fine motor
speed and fine motor dexterity were obtained: the Reaction Time Index of
the MicroCog (Powell et al., 1993), the average number of taps over five trials
for Dominant Hand and Nondominant Hand on the Finger-Tapping Test of
the HRNB (Reitan & Wolfson, 1993), and time in seconds for Dominant Hand
and Nondominant Hand on the Grooved Pegboard (Lafayette Instrument
Company, 1989). The correlations, means, and standard deviations for each
measure are shown in Table 4.52. The mean WAIS-III FSIQ scores for the
MicroCog, Finger-Tapping Test, and Grooved Pegboard samples were 91.3
(SD = 13.2), 95.7 (SD = 11.8), and 90.4 (SD = 13.5), respectively.
185
 Bjqe] "ZG"y SUONLII1I0D
Usamjeg
oY) []]-SINM
pue sammseay]
JO FULT 10,0] paadspue sul

186
1010] A}I19}x9q
Ho40J91IN Buiddey-sabul4
sa panoosy pieoqbad
uonoeey jUeUIWOG JUBUIWOPUON jUeUIWOG JUBUIWOPUON
owiy xapuy
= ues’ as puey puey ueai\l as puey puey uealyy as
III-SINMA
Auoyipny ayelpewitu| We g¢8 OSL 60'- Ocis WE ool 80° Siz L'88 QS
/eNsiA, a}eIPSWW] by 928 06 WA HOF L28 Gel €¢’- Wee 0s ech)
ayeipewiu|Kowsy| ve 6:08 el €0° 9 ores 8k ve'- Ge’- 828 Cet
Aoylpny pefejeg LES co emt mrel [oom 6- 206 6LZtL €0° Sle 7'S8
= GLZL
jensi, paxejaq ZO" Les Cel 80° 60'- 028 Srl or’. Coe LEB Ort
pny 98y 18g vl V8~ v0e 90'- = LO 296 G02 60'- Ce 288 SLL
jesaued Mowe €0° v08 9vl 60'- vay €98 8st Pee eon sce eSher
Burson Mowe SQ 8°88 ZEL 80° LO 16 Ort oes 8l'- 26 Cvk
sayisodwoy
ssed0ig
Asoyipny
jeu-a|6uisBujweey Gz 80°- OL’- 0" git
ados
Buluee7 fee OL’ 60'- 60° eb
UOUa}OY 6. Gill 00° We 20°
JPASUIOY 20° es Ske 0'- 02'-
0'8E
LSE
6'€8
€'€6
9LOL
ueol\
Evidence Base for Validity of the WAIS-III and the WMS-III

as 6'EL Vel G6 898 0'09

N
vl
€¢
vv
 Demographic Studies

Very low correlations between the WMS-III auditory indexes and all of the
external measures and moderately low correlations between the WMS-III
visual indexes and the external measures were expected.
The WMS-III Working Memory Index shows a strong relationship with the
MicroCog Reaction Time Index but is unrelated to the Finger-Tapping Test
and Grooved Pegboard measures. Additionally, the results show that perfor-
mance for the Dominant Hand on the Grooved Pegboard measure is inverse-
ly related to the WMS-III visual memory indexes. These results indicate that
the WMS-III memory indexes are generally unrelated to simple and choice
reaction time (i.e., MicroCog) and fine motor speed. However, visual memo-
ry is related to fine motor dexterity and perceptual motor speed, with those
who score higher on WMS-III visual indexes demonstrating faster perfor-
mance on the Grooved Pegboard dominant-hand measure. These correla-
tions provide evidence of convergent and divergent validity.

Demographic Studies
Differential age effects on measures of intellectual functioning have been
well documented and have been, in part, the basis for the concepts of “crys-
tallized” and “fluid” abilities. J. J. Ryan, Sattler, and Lopez (2000) evaluated
age trends in WAIS-III performance, using procedures employed by Sattler
(1992) in evaluating WAIS-R age trends. The analysis revealed relatively sta-
ble performance on measures of verbal intellectual abilities across the age
span (Ryan, Sattler, et al., 2000). The authors reported that performance on
Information was the most stable across age groups whereas performance on
Letter-Number Sequencing displayed the most age-related declines. Ryan,
Sattler, et al. also found that measures of processing speed showed the
strongest age-associated effects and that performance measures in general
displayed more age effects than did verbal measures (Ryan, Sattler, et al.,
2000). The authors concluded that the observed age trends are evidence that
the WAIS-III subtests measure aspects of fluid and crystallized intelligence.
Basso, Harrington, Matson, and Lowery (2000) studied sex effects on perfor-
mance on the WMS-III Verbal Paired Associates and Faces subtests. Their
results indicated significant sex effects for Verbal Paired Associates learning
but none for Faces. Basso et al. suggested that male respondents might have
obtained lower scores than expected on Verbal Paired Associates because
norms are not stratified by sex. The study by Basso et al. was based on a rela-
tively small sample of homogeneous participants. An analysis of the sex
effects on Verbal Paired Associates in the standardization sample yielded no

187
Evidence Base for Validity of the WAIS-III and the WMS-III

significant results at the multivariate level, Wilks’ Lambda = .996, F(4, 1245)
= 1.12, p> .05. Individual measures of delayed cued recall, F(1, 1248) = 4.5,
p< .05) and Verbal Paired Associates learning, F(1, 1248) = 4.9, p< .05, hada
small, statistically significant effect. The average performance for women on
the learning trial and the delayed cued recall scaled scores were 10.2 (+3.1),
and for men, 9.8 (+2.8). The mean difference is small and not likely to have
clinical relevance.
M. Taylor and Heaton (2001) studied comprehensive demographically adjust-
ed (age, education, sex, and race/ethnicity) WAIS-III/WMS-III factor scores.
Historically, the IQ and memory scores of the Wechsler scales have been
adjusted for age. The authors proposed that the comprehensive demographi-
cally adjusted norms improves the accuracy of diagnostic classification for
WAIS-III and WMS-III scores. Taylor and Heaton employed sophisticated
fractional polynomial regression procedures with the WAIS-III/WMS-—III stan-
dardization sample and oversample data to create demographically adjusted
norms. They explored multiple cut-off criteria and determined that 1.0 SD
provided optimal sensitivity and specificity statistics. The number of im-
paired factor scores also provides useful neurodiagnostic information.
Demographically corrected norms are available in the upgraded version of
the WAIS-II/WMS-III/WIAT-II scoring assistant software program.
The demographically adjusted norms provide the clinician with an estimate
of the examinee’s current intellectual and memory performance relative to a
homogeneous subgroup similar in age, education level, sex, and race/eth-
nicity. With demographically adjusted norms, the clinician can determine if
the respondent's current cognitive performance is below expectations, given
specific background variables. Below-expected performance may indicate
the presence of a clinical condition that has affected the examinee’s cogni-
tive functioning. Demographically adjusted norms are not intended for the
purpose of making judgments regarding intellectual capacity, expected func-
tional capacity, or predicted academic abilities. These norms are useful in
the context of a neuropsychological evaluation. They are a means of avoid-
ing overestimation of neuropsychological impairments in some groups (e.g.,
low educational level) and increasing sensitivity to impairment in other
groups (e.g., high educational level). The use of demographically adjusted
norms to diagnose a learning disability or mental retardation, to predict an
individual's educational abilities or ability to perform a specific job, or to
determine if the individual has functional disability would be inappropriate.
When using demographically adjusted norms to make judgments regarding
acquired neurocognitive impairment, clinicians are encouraged to also con-
sider any other relevant psychosocial information (e.g., all specific education
or occupational records) that may be available.

188
 Prediction of Premorbid Intellectual and Memory Functioning

Prediction of Premorbid Intellectual

and Memory Functioning
The Wechsler Test ofAdult Reading (WTAR; The Psychological Corporation,
2001) was developed and co-normed along with the WAIS-III and WMS-III.
The methodology for the development of the WTAR is directly associated
with the National Adult Reading Test (NART; Nelson, 1982) and the North
American counterparts of the NART, American Version of the National Adult
Reading Test (AMNART; Grober & Sliwinski, 1991) and North American Adult
Reading Test (NAART; Blair & Spreen, 1989). All of these tests are based on a
reading recognition paradigm, which requires no text comprehension or
word definition, but only recognition of words that have irregular grapheme-
to-phoneme translation. These tests share some item content.
The use of words with irregular pronunciation minimizes the assessment of
the current ability of the examinee to apply standard pronunciation rules
and maximizes the assessment of previous learning of the word (Grober &
Sliwinski, 1991; Nelson, 1982). Unlike many intellectual and memory abili-
ties, reading recognition is relatively stable in the presence of cognitive
declines associated with normal aging or brain insult, although it is not
impervious to the effects of significant intellectual impairments (see
Crawford [1992], Spreen & Strauss [1998], Putnam, Ricker, Ross, & Kurtz
[1999] for reviews).
Numerous methodologies have been developed in an attempt to determine
intellectual loss or decline. Each of the methodologies has strengths and
weaknesses that affect the accuracy and utility of that approach (see Putnam
et al., 1999, for a review). The NART methodology has been studied exten-
sively and shown to be an effective method for predicting verbal, perfor-
mance, and full-scale IQ with enhanced predictive power when
demographic variables are used concurrently in prediction equations
(Crawford, 1992).
The WTAR manual provides tables of predicted intellectual functioning
based on demographic variables and a methodology similar to that devel-
oped by Barona, Reynolds, and Chastain (1984).
The WTAR was developed for use in the United States and United Kingdom
and provides the clinician a test with decided advantages over other reading
tests designed to predict premorbid intellectual functioning.

189
Evidence Base for Validity of the WAIS-III and the WMS-III

¢ The WTAR is normed on a large, nationally representative,

stratified sample.
¢ The WTAR is directly linked to and co-normed with the WAIS-III
and WMS-III. Prediction equations are available for WAIS-III Index
scores and selected WMS-III Index scores.
¢ Norms are provided for a wide age range, from late adolescence
through late adulthood.
¢ Validity data from samples with a variety of clinical disorders indicate
the utility of WTAR prediction of WAIS-III Index scores and WMS-III
memory performance.
e Prediction of WTAR scores based on demographic variables enables the
user to better determine whether demographic variables or WTAR plus
demographic variables is most appropriate for predicting a respondent's
IQ and memory scores.
* Tables provide for easy conversion of WTAR and demographic variables
to predicted IQ and memory performance. Scores predicted from WTAR
performance alone, demographic variables alone, or WIAR performance
plus demographic variables are presented.

The purpose of the WTAR is not for the assessment and diagnosis of devel-
opmental reading disorders, as is the WIAT-I, but rather for the estimation
of premorbid intellectual and memory abilities. The use of WTAR and
demographics predictions of intellectual functioning should not be used as
a direct measure of intellectual ability. These measures have not been vali-
dated for the purposes of identifying intellectual capacity, prediction of
academic potential, identifying mental deficiency or developmental dis-
ability, prediction of employability or job performance, or determination of
disability status. These measures are valid only for the purposes of estimat-
ing an examinee’s intellectual status prior to illness or injury.

The demographically adjusted norms and the WTAR are different methods
for determining whether an examinee’s current performance is consistent
with the expected performance of individuals with similar psychosocial
backgrounds. The WTAR adds another dimension to this assessment by
testing the examinee’s current word-reading ability, which has been shown
to be relatively insensitive to the effects of brain injury and cognitive
decline. The clinician tests the hypothesis: In view of this examinee’s psy-
chosocial background (age, education, sex, race/ethnicity) and reading
skills, what is the examinee's expected intellectual ability? Is the examinee’s
current ability level consistent with what is expected based on his or her

190
 Malingering Studies

psychosocial background and reading ability? The clinician is able to deter-

mine whether the degree of difference is statistically significant and
whether the frequency of the difference is rare or common.

Malingering Studies
Killgore and DellaPietra (2000a) investigated the use of the Logical Memory
delayed recognition to detect response bias and possible malingering. The
recall trial is presented in a yes/no format, so the examinee has a 50%
chance of responding correctly to an item by guessing (Killgore &
DellaPietra, 2000a). The authors concluded that the wording of the items
and the sequential ordering would result in an examinee’s being able to
guess the correct answer without any knowledge of the actual story. The
authors identified three sources of item bias: yes-saying to proper names,
priming of yes responses by previous items of similar content, and nay-
saying to unlikely occurrences. On the basis of these hypotheses, Killgore
and DellaPietra predicted which items would demonstrate a response bias.
They administered the Logical Memory recall trial to a nonclinical control
group without having exposed the respondents to the stories. The results
generally supported their hypotheses. The authors then administered the
items according to standard procedures to a mixed neurological sample to
determine if patients who had received the standard protocol also exhibited
a higher tendency of endorsing the identified items as well as recall items
that did not pull for specific responses. The clinical group obtained higher
scores than did the “naive” control group and a “better-than-chance” level of
performance. The results indicated that patients with memory problems had
recognized some of the story content. Item-level analysis provided further
evidence that presentation of item content resulted in more correct respons-
es than would be expected by chance.
In a second study, Killgore and DellaPietra (2000b) identified six items on the
WMS-III Logical Memory subtest that were correctly endorsed above chance
levels by a nonclinical group. The subtest was administered to a group of
control respondents who were instructed to feign memory impairment and
to a mixed clinical group. The authors’ purpose was to determine if the six
items would discriminate among the three groups: naive, malingerers, and
patients. The six items were entered into a discriminate function analysis,
and the results were highly significant, with 98.9% classification accuracy.
total score
A computational equation was derived for the six items, and this
inate analysis . This analysis yielded the same
was resubmitted to discrim
the items entered individu ally. Killgore and
results as the analysis with

191
Evidence Base for Validity of the WAIS-III and the WMS-III

DellaPietra used a cutoff score of 136 and below to indicate malingering (sub-
test scores range from —22 to 226). The cutoff score resulted in 97% sensitivity
and 100% specificity in identifying malingerers. The authors suggested that
these findings provide initial support for the use of the Logical Memory recall
trial in the assessment of malingering versus actual brain injury.

Sequential Assessment
The clinician may be required to assess the functioning of the same patient
on more than one occasion for a variety of clinical reasons (e.g., assess
change after surgery or cognitive decline), or multiple clinicians might
assess the functioning of the same patient using the same instruments
(medical-legal cases). Iverson (2001) utilized the test-retest correlations to
compute reliable change estimates for the primary WAIS-II and WMS-III
index scores in three clinical groups.
Lineweaver and Chelune (2000d) proposed a methodology for evaluating
significant changes in WAIS-III and WMS-III scores over time. The method-
ology assumes practice effects result in non-zero interval changes, influ-
ences of measurement error, and regression to the mean. The reliable
change index is noted to be limited by assuming a zero change score from
Time 1 to Time 2. The authors proposed the use of a regression-based
approach that controls for non-zero changes over time and the influences of
other variables such as age. The standard error of regression is used as a
measure of significant change. The authors illustrated the application of the
methodology, providing several case examples.

Other Issues
Since the original publication of the WAIS—IIJI—WMS-III Technical Manual,
several issues regarding the use of the WAIS-III and the WMS-III have
emerged.
Tulsky and Zhu (2000) investigated potential fatigue effects on the normative
data for the Letter-Number Sequencing subtest. During standardization,
the Letter-Number Sequencing subtest was administered as part of the
WMS-III, which was administered in counterbalanced order with the
WAIS-IIL. Tulsky and Zhu compared the performance of participants who
were administered Letter-Number Sequencing before completing the
WAIS-III to the performance of those completing the subtest after the
WAIS-III had been administered. The authors found no performance differ-
ences, even for the older age groups.

192
 Other Issues

Zhu and Tulsky (2000) examined all of the WAIS-III and WMS-III subtests
and index scores to determine if order of presentation resulted in differences
in performance. The authors noted very small effects for Digit Span and
Digit Symbol—Coding of the WAIS-III and for Faces II and Logical Memory
II of the WMS-III. The effect sizes were very small, a result suggesting that
the mean performance on the subtests across administration orders was
very similar. This finding indicates that administration order did not invali-
date the norms developed for these tasks. The index scores and most subtest
scores did not exhibit any significant effects of administration order (Zhu &
Tulsky, 2000). The findings from these studies are consistent with those by
Doss et al. (2000), who found no differences in performance level between
the longer standardization versions and the published versions of the
two instruments.
This update of the WAIS—IJ]—WMS-III Technical Manual includes a new
table of scaled scores for Digit Span Backward (see Appendix E). These
norms enable the clinician to determine if the patient has a specific
weakness on Digits Backward relative to his or her overall performance
on that subtest.

193
«
.
 CHAPTER 5

Interpretive Considerations

This chapter focuses on basic interpretive considerations of the WAIS-III

and the WMS-III and addresses general interpretive issues concerning
selected topics and test scores. The chapter also briefly discusses how some
of the new features that have been added and some of the changes from the
previous versions of the Wechsler scales might be used in clinical practice.
Results from the WAIS-III and the WMS-III provide important information
regarding an individual’s neurocognitive functioning, but they should never
be interpreted in isolation. Four broad sources of information are typically
available to the clinician conducting a psychological or neuropsychological
evaluation: medical and psychosocial history, direct behavioral observations,
quantitative test scores, and qualitative aspects of test performance. The
WAIS-III and the WMS-III provide quantitative and qualitative information
that is best interpreted in conjunction with a thorough history and careful
clinical observations of the examinee. Results should always be evaluated
within the context of the reasons for referral and all known collateral
information.

Scores and Descriptive Classifications

Subtest raw scores for the WAIS-III and the WMS-III are transformed to age-
corrected subtest scaled scores with a mean of 10 and a standard deviation
of 3 (the WMS-III also includes some instances in which raw scores are
transformed to percentile ranks). A subtest scaled score of 10 reflects the
average performance of a given age group (or the reference group). Scores
and
of 7 and 13 correspond to 1 SD below and above the mean, respectively,
scaled scores of 4 and 16 deviate 2 SDs from the mean.
various
The WAIS-III and the WMS-III also yield standard scores based on
scores and
composites of subtest scaled scores. The WAIS-III IQ and Index
also share a
the WMS-III Index scores, both computed as standard scores,

195
Interpretive Considerations

common metric for evaluating level of performance. The IQ and Index

scores are scaled to a metric with a mean of 100 and a standard deviation of
15 for each age group. A score of 100 on any of these measures defines the
average performance of individuals similar in age. Scores of 85 and 115
correspond to 1 SD below and above the mean, respectively, whereas scores
of 70 and 130 are 2 SDs below and above the mean. About 68% of all exami-
nees obtain scores between 85 and 115, about 95% score in the 70-130
range, and nearly all examinees obtain scores between 55 and 145 (3 SDs
on either side of the mean). The relation of WAIS-III IQ and Index scores
and WMS-III Index scores to deviations from the mean and the associated
percentile rank equivalents are presented in Table 5.1. Additionally, the
WMS-III yields four composite scores that represent various aspects of audi-
tory learning. Scores for these supplemental composites are presented as
percentile rank equivalents, which represent an individual's standing com-
pared to the normative group. The descriptive classifications corresponding
to the IQ and Index scores for the WAIS-III and the WMS-III are presented
in Table 5.2

Use of Confidence Intervals Around Scores

As described in Chapter 3, reporting a score in terms of confidence intervals
is a means of expressing the reliability of that test score. Confidence intervals
assist the examiner in test interpretation by delineating a range of scores in
which the examinee’s “true” score most likely falls, and reminds the exam-
iner that the observed score contains measurement error.

Normative Reference Groups

The computation of subtest scaled scores for both the WAIS-III and the
WMS~III are based on the performance of examinees in 13 age groups
within the standardization samples. With these age-corrected scaled scores,
an individual’s performance is compared to that of his or her age peers. The
age-corrected scaled scores are used to derive the IQ and Index scores. This
methodology is consistent with the construction of IQ and Index scores for
the WISC-III. The procedure, however, represents a significant departure
from the one used to derive subtest scaled scores of the WAIS-R, which are
based on the performance of a reference group of individuals aged 20-34
years. This change, however, does not affect the IQ scores because the
WAIS-R sums of subtest scaled scores (i.e., based on the reference group's
scaled scores) were age-corrected at the IQ score level. The primary differ-
ence is that the age correction in the WAIS-III occurs at the subtest scaled-
score level rather than at the level of the sums of scaled scores.

196
 Normative Reference Groups

Table 5.1. Relation of IQ and Index Scores to Standard Deviations

From the Mean and Percentile Rank Equivalents
Number of SDs Percentile Rank
IQ/Index Score From the Mean Equivalent*

ede: +373 Stele)

150 +3'/s 309:9

145 +3 39.9
140 +27; 99.6
Res +2'/3 99
130 +2 98
125 +14 U5
120 +1'/ 91
115 +1 84
110 + 13 75
105 +h 63
100 0 (Mean) 50
95 —'h Bid

90 —7; 25

85 -] 16
80 1h 3
75 -lh 5
70 =2 :
65 - 2's 1
60 Big EE 0.4
55 a3 0.1
50 34, <0.1
vis 37; <0.1
ie i OE ee Ee
on.
4 The percentile ranks are theoretical values for a normal distributi

197
Interpretive Considerations

Table 5.2. Qualitative Descriptions of IQ and Index Scores

Percent Included in
Theoretical
Score Classification Normal Curve

130 and above Very Superior 22

120-129 Superior 6.7

110-119 High Average 16.1

90-109 Average 50.0

80-89 Low Average 16.1

70-79 Borderline 6.7

69 and below Extremely Low Zee
2 _______ EEE

The use of age-corrected scaled scores in the WAIS-III is supported by

research and empirical data. In terms of brain functioning, young adulthood
represents the stage when the brain has reached physical maturation but has
not yet been significantly affected by normal, age-related neurodegenerative
processes. Analyses of the WAIS-III and WMS-III standardization data reveal
that for many tasks, young adults (ages 20-34) obtain the highest scores of
any age group across the life span. However, performance on some tasks,
such as Vocabulary and Information subtests, increases until 40-50 years of
age. Reference-group norms do not account for such changes. The age-
corrected subtest scaled scores, on the other hand, reflect an individual's
standing in relation to his or her age-matched peers.
Comparisons to a reference group of 20- to 34-year-olds is still possible at
the subtest level for the WAIS-III, but such comparisons are considered
secondary, or supplementary, for interpretation. For the WMS-III, the addi-
tion of optional, reference-group norms is a change from the WMS-R. When
subtest score interpretations involve questions about performance relative
to that of a younger group, the reference-group norms are appropriate. The
reference-group norms might be compared to the age-adjusted norms as a
way of highlighting the effects of age, if any, on memory and other cognitive
functions.

Level of Performance
In its most straightforward form, level of performance refers to the rank
(usually expressed as a scaled or standard score, percentile rank, and

198
 Level of Performance

descriptive classification) obtained by an individual on a given test in com-

parison to the performance by an appropriate normative group. For clinical
decisions, level of performance is important for estimating the presence and
severity of any impairment or presence of a relative strength. For nonclinical
settings (e.g., industrial and occupational settings), the emphasis of level of
performance is placed more on competency and the patterns of a person's
strengths and weaknesses without necessarily implying any type of impair-
ment. Test results can be described in a manner similar to the following
example:
Relative to individuals of comparable age [or, alternatively, of a reference
group of younger adults], this individual is currently functioning in
the [descriptive classification] (see Table 5.2) range on a standardized
measure of [IQ or Index name}.

Examiners should keep in mind that IQ and Index scores are estimates of
overall functioning in a particular cognitive domain or content area. As such,
composite scores should always be evaluated in the context of those mea-
sures that contribute to the specific IQ scale or index. The interpretation of
an IQ or Index score may be influenced by substantial differences between
the subtest scores on which the IQ or Index score is based. Two component
subtest scores, one unusually high and one unusually low, for example, will
push the Index score toward the arithmetic mean and thus toward the aver-
age range. Such an average score reflects a dramatically different pattern of
abilities than does an average Index score obtained from two subtest scores
that are both in the average range. Evaluation of the component scores of IQ
and Index scores is important, and the failure to evaluate them can lead to
erroneous Clinical inferences, especially when such components and the
Index scores are interpreted in isolation. As stated in the opening paragraphs
of this chapter, the clinical interpretation of the WAIS—III and WMS-III
scores should be conducted in the context of the individual’s medical and
psychosocial history and in consideration of behavioral and other qualitative
observations.

WAIS-IIl
Description of IQ Scores
The VIQ score of the WAIS-III is a measure of acquired knowledge, verbal
reasoning, and attention to verbal materials. The items that compose the
subtests of this scale, even those included in the Stimulus Booklet, are pre-
sented verbally, and the examinee articulates the responses. The VIQ score,
apart from some content changes of the scale at the item level, is relatively
unchanged from the VIQ score in the WAIS-R.

199
Interpretive Considerations

The PIQ score is a measure of fluid reasoning, spatial processing, attentive-

ness to detail, and visual-motor integration. The tasks should be fairly novel
to the first-time examinee. The PIQ score, apart from some changes of the
scale at the item level, is basically similar to the PIQ score of the WAIS-R. In
the WAIS-III, however, the impact of bonus points has been reduced. The
Object Assembly subtest has been made optional and replaced with Matrix
Reasoning, which does not depend on quick performance. With the intro-
duction of Matrix Reasoning, the PIQ score is less speed-dependent and
places greater emphasis on abstract, nonverbal reasoning.
The FSIQ score is the overall summary score that estimates an individual’s
general level of intellectual functioning. It is the aggregate score of the VIQ
and PIQ scores and is usually considered to be the score that is most
representative of g, or global intellectual functioning.

Description of Index Scores

Like the VIQ score, the VCI score of the WAIS-III is a measure of verbal
acquired knowledge and verbal reasoning. The items of the subtests that
compose this index are presented verbally, and the examinee must articulate
the responses. The major difference between this score and the VIQ score is
the inclusion of Digit Span, Arithmetic, and Comprehension in the VIQ scale.
Because the attentional/working memory subtests, that is, Arithmetic and
Digit Span, are not included in the VCI, the index may be conceptualized as
a more refined, “purer” measure of verbal comprehension.
The POI score of the WAIS-III is a measure of nonverbal, fluid reasoning,
attentiveness to detail, and visual—-motor integration. All of the items of the
subtests that compose this scale are presented in the Stimulus Booklet; the
examinee responds by pointing, building block designs, or indicating a
response choice (orally or by pointing). Quick responding is less important
on the POI than it is on the PIQ scale because the subtest most related to
processing speed (Digit Symbol—Coding) is not included in this index.
Moreover, only one of the three subtests of the POI includes bonus points
for quick performance. The composition of the POI score makes it a more
refined measure of fluid reasoning and visual-spatial problem solving than
the PIQ score.
The Working Memory indexes of both the WAIS-III and the WMS-III are
highly correlated (r= .82). The subtests of both indexes include a range of
tasks that require the examinee to attend to information, to hold briefly and
process that information in memory, and then to formulate a response. The
main difference between the WAIS-III and WMS-III indexes is the inclusion
of Arithmetic and Digit Span (verbal tasks) in the WAIS-III index and the
inclusion of a nonverbal task (Spatial Span) in the WMS-III. The working

200
 Level of Performance

memory subtests composing the WAIS-III index include only verbally pre-
sented items, whereas the WMS-III index is equally represented by subtests
with auditorily and visually presented items.
The PSI score is a measure of the individual's ability to process visual infor-
mation quickly. Comparisons between the PSI and POI scores can reveal
possible effects of time demands on visual-spatial reasoning and problem
solving. Research has suggested that the PSI is also highly sensitive to many
different neuropsychological conditions (see Chapter 4).

IQ Scores Versus Index Scores

Unlike its predecessors, the WAIS-III provides options for two sets of scores.
The examiner should consider these options before deciding which subtests
to administer (see Figure 1.1 of the WAJS-III Administration and Scoring
Manual). When time permits, or the goal is to obtain extensive data about
intellectual functioning, all 13 subtests (excluding Object Assembly, which is
optional) should be administered. Doing so increases testing time from the
WAIS-R administration time by approximately 5 minutes but allows for the
computation of both IQ and Index scores. However, the requirements of the
clinical situation or time constraints may preclude the administration of all
13 subtests.
The referral question, purpose of testing, and practical issues (such as time
constraints) will determine which scores to obtain and which subtests to
administer. Some federal and state laws and regulations require IQ scores
and the analysis of the discrepancies between those scores. In such cases,
the 11 subtests that make up the WAIS-III IQ scores must be administered.
Of course, during the testing, the examiner can watch for any patterns in the
examinee’s performance to determine if the supplementary subtests (or
optional subtest) would aid in the interpretation of the results.

Optional Procedures of the WAIS-III

Discrepancies Between Digit Span Forward and Backward
The ability to repeat digits in the order of verbal (forward) presentation is
considered a measure of concentration, and the average adult performance
is 7 digits recalled, plus or minus 2 (Miller, 1956; Spitz, 1972). This task tends
to remain relatively stable with aging in normally functioning men and
women (Weintraub & Mesulam, 1985). Orsini et al. (1987) and Benton,
Eslinger, and Damasio (1981) have shown that only a small number of older
adults (approximately 8%) show decrements in performance. Digit Span
Backward, however, is more affected by aging and by impairment. Costa
(1975) has shown that a backward verbal digit span in normally functioning

201
Interpretive Considerations

adults is generally one less digit than the forward span. However, normally
functioning adults over 70 years old show a greater discrepancy, with a sig-
nificantly shortened backward span (Lezak, 1995).
To assess this effect of aging, E. Kaplan et al. (1991) have developed mea-
sures that highlight the differences between the forward and backward tasks
of Digit Span. First, the greatest number of digits recalled in the forward task
and the greatest number recalled in the backward task can be compared to
the performance of the normative sample (see Table B.6 in the WAIS-III
Administration and Scoring Manual). Second, the difference between the
two can be calculated and compared to the performance of the normative
sample (see Table B.7 of the WAIS-III Administration and Scoring Manual).
These two procedures can help the examiner determine if there are
attentional deficits, which become more apparent when the two compo-
nents of Digit Span are broken down. The information can also help the
examiner determine if further testing of working memory functioning is
warranted. For example, a significant discrepancy score may indicate that
the Spatial Span subtest from the WMS-III should be administered. This
subtest is the visual analogue of Digit Span and is also related to the working
memory construct.
Digit Symbol Optional Procedures
As E. Kaplan et al. (1991) have pointed out, there may be several reasons for
poor performance on Digit Symbol—Coding. In addition to tapping process-
ing speed, the subtest is also affected by motor coordination, short-term
memory, visual perception, and clerical speed and accuracy. Some of the
potential reasons for an examinee’s poor performance can be determined by
two optional procedures, which were designed to differentiate incidental
memory and graphomotor speed.
The first optional procedure, Digit Symbol—Incidental Learning, taps the
individual's ability to learn and remember the number-symbol pairing. On
the Digit Symbol—Coding subtest, if an individual must continuously check
the key before responding to an item, then his or her performance will be
considerably impeded. To determine if failure to learn or to remember
digit-symbol combinations is affecting performance on the subtest, the
examiner can administer Incidental Learning, which includes two tasks. For
the first task, Pairing, the examinee is asked to recall the symbols paired with
the numbers. For the second task, Free Recall, the examinee must recall the
symbols by themselves. As E. Kaplan et al. (1991) pointed out, if the examinee
has difficulty remembering (as indicated by a paucity of symbols recalled,
incorrect pairing of the numbers and symbols, and confabulations of the
symbol), then the possibility of memory impairment should be considered.

202
 Level of Performance

The second optional procedure, Digit Symbol—Copy, requires the examinee

to copy the symbols without having to match them to numbers. Perfor-
mance on this task can reveal whether or not the examinee has difficulty
writing the symbols (poor graphomotor production and speed).
Normative data for these optional procedures are presented as cumulative
percentages for the 13 age groups and the reference group and are included
in Appendix A, Table A.11 of the WAIS-III Administration and Scoring
Manual.

WMS-Iil
Examiners who are familiar with the WMS-R will notice that the index struc-
ture of the WMS-III has changed in a number of ways.
First, the number of indexes has increased from five in the WMS-R to eight
in the WMS-III. The WMS-III includes eight Primary Indexes and four Audi-
tory Process Composites. As their label suggests, the Primary Indexes are
intended to be the principal scores used to evaluate memory functioning.
The Auditory Process Composites are supplementary in nature and can be
used to evaluate various processes of memory when stimuli are presented in
the auditory modality.
Second, the WMS-III reflects two notable changes in the index nomencla-
ture. Because the same subtest content can be encoded in different ways
(e.g., verbal labels can be given to “nonverbal” materials such as figures or
pictures), the “verbal” label used in the WMS-R was changed to reflect more
accurately the modality of presentation rather than the index content.
Therefore, the label “auditory,” which is the parallel to “visual,” is used in-
stead of “verbal.” In some instances, however, the label “verbal” was retained
because of historical continuity (e.g., the Verbal Paired Associates subtest
dates to the early 1900s). Next, the Attention/Concentration Index of the
WMS-R was renamed on the WMS-III to the Working Memory Index. This
change reflects the content shift from relatively low-level attentional tasks to
relatively high-level working memory tasks. For example, Letter-Number
Sequencing (a complex working memory task) replaces the Mental Control
subtest on the Working Memory Index.
Third, the WMS-III indexes represent significant revisions of the content
of the WMS-R. The two subtests that contribute to the visual memory
indexes—Faces and Family Pictures—are new, as is the Letter-Number
Sequencing subtest. The Visual Reproduction, Mental Control, and Digit
Span subtests were retained in the WMS-III but are now optional subtests.
Additionally, whenever possible, delayed recognition tasks were included for
comparison with performance on the delayed recall conditions.

203
Interpretive Considerations

Finally, the method of calculating several WMS-III Index scores differs sig-
nificantly from that used in the WMS-R. The WMS-R Verbal Memory Index
and Visual Memory Index are analogous to the WMS-III Auditory Immediate
and Visual Immediate indexes. The WMS-R General Memory Index is most
similar to the WMS-III Immediate Memory Index. The WMS-III General
Memory Index consists of only delayed memory subtests, in sharp contrast
to the WMS-R General Memory Index, which is composed of only immedi-
ate memory subtests. This conceptual shift from immediate to delayed
measures to represent the global memory score (i.e., General Memory) was
based on two considerations: First, the relationship between immediate
memory and delayed memory is such that adequate performance on de-
layed subtests warrants the assumption of adequate immediate memory,
whereas the reverse is not always true. Second, the delayed memory mea-
sures conceptually correspond more closely to real-life or everyday memory
demands; thus the ecological validity of delayed measures is likely to be
greater.

Primary indexes
Auditory Immediate and Auditory Delayed Indexes
The Auditory Immediate Index is composed of the Logical Memory I and
Verbal Paired Associates I subtests, and its score is calculated as follows:

Logical Verbal Paired Auditory

Memory | Associates | Immediate
Recall Recall Index
Scaled Score Scaled Score Score

The Auditory Delayed Index is composed of the Logical Memory II and

Verbal Paired Associates II subtests, with its score calculated as follows:

Logical Verbal Paired Auditory

Memory Il Associates || Delayed
Recall ate Recall = Index
Scaled Score Scaled Score Score

During Logical Memory I, narrative stories are read aloud to the examinee,
who is asked to retell them from memory. The Verbal Paired Associates I sub-
test requires examinees to learn pairs of words that are seemingly unrelated.
With the exception of the first story of Logical Memory I, all information to
be remembered from these two subtests is presented at least twice. These
subtests require the examinee to recall (via free recall) the information later.
For Logical Memory II and Verbal Paired Associates II, the delayed condi-

204
 Level of Performance

tions, the examinee is asked to recall again the material that was presented
in the immediate condition. The immediate and delayed conditions are tem-
porally separated by approximately 25-35 minutes. The Auditory Immediate
Index and Auditory Delayed Index scores are calculated by summing the
appropriate scaled scores; the subtests that compose each index are thus
equally weighted.
These Index scores are measures of memory functioning when stimuli are
presented in the auditory modality. Low scores, relative to an individual’s
intellectual and attentional functioning, may suggest a verbal learning or
memory problem. A low score on the delayed index, relative to that on the
immediate index, may indicate a high rate of forgetting. The assessment
of delayed recall should always be made in the context of the immediate
condition because delayed recall (i.e., the amount of information available
through recall) depends on the amount of information that was initially ac-
quired. In other words, inferences about an examinee’s inability to retain
information may be inaccurate if relatively little information is learned
initially.

Visual Immediate and Visual Delayed Indexes

The Visual Immediate Index is composed of the Faces I and Family Pictures I
subtests, with its score calculated as follows:

Family
Faces | Visual
Pictures |
Recognition Immediate
Scaled Score Fecal Index Score
Scaled Score

The Visual Delayed Index is composed of the Faces II and Family Pictures II
subtests, with its score calculated as follows:

Family
Faces II Visual
Pictures Il
Recognition Delayed
Recall
Scaled Score Index Score
Scaled Score

For Faces I, the examinee is initially presented photographs of 24 target

faces. The examinee is then presented photographs of 48 faces, including
the 24 target faces and 24 new faces. The examinee must identify each face
as either a target face or a new one. For Faces II, the examinee is again pre-
sented the 24 target faces and 24 different faces and is again asked to iden-
tify each face as either a target face or a new one. For Family Pictures I,a
family portrait and four subsequent scenes involving the family characters
and family dog are shown to the examinee. The examinee is asked to identify

205
Interpretive Considerations

who was in each scene and each character's activity and location. Family
Pictures II requires the examinee to recall the same information without
again seeing the family portrait or the four scenes. The immediate and de-
layed conditions are temporally separated by approximately 25-35 minutes.
As with the auditory indexes, the visual indexes are equally weighted by their
component subtests.
The interpretation of scores for the Visual Immediate and Visual Delayed
indexes is similar to that of the Auditory Immediate and Auditory Delayed
indexes. These Index scores summarize overall memory functioning when
information is presented visually. Low scores, relative to an individual's intel-
lectual and attentional functioning, may reflect a memory weakness or
impairment when information is presented visually. The interpretation of
scores on the Visual Delayed Index should be made in the context of perfor-
mance on the Visual Immediate Index. It is also important for examiners to
keep in mind that the subtests that compose these visual indexes represent
both a recognition paradigm (Faces) and a recall paradigm (Family Pictures).
Consequently, meaningful differences between the scores on Faces and
Family Pictures, with a higher Faces score, may indicate retrieval difficulties.

immediate Memory Index

The Immediate Memory Index score is calculated by summing, and thus
equally weighting, the sums of scaled scores on the subtests that compose
the Auditory Immediate and Visual Immediate indexes:

Logical Verbal Paired Family Immediate

Faces|
Memory | Associates | Pictures | Memory
Recall Recall +} Recognition Recall Index
Scaled Score Scaled Score Scaled Score Scaled Score Score

The Immediate Memory Index score is considered the best global indicator
of immediate memory functioning. However, differences between the
Auditory Immediate Index and Visual Immediate Index scores can be
evaluated to determine if the presentation modality (spoken or visual) has
affected the examinee’s acquisition and recall of information. Low scores,
relative to intellectual functioning, attention, or delayed memory, may rep-
resent a weakness or deficit in learning or immediate memory.

Auditory Recognition Delayed Index

The Auditory Recognition Delayed Index is composed of the Logical Memory
Ii and Verbal Paired Associates II subtests. This Index score is simply a linear
transformation from the Auditory Recognition Delayed Index scaled score
(metric of 1-19) to the Auditory Recognition Delayed Index standard score
metric:

206
 Level of Performance

Auditory Auditory
Recognition Recognition
Delayed m4 Delayed
Scaled Score Index Score

The Auditory Recognition Delayed Index is the recognition counterpart to

the Auditory Delayed Index, which incorporates a recall procedure. The
Auditory Recognition Delayed Index is formed by adding the recognition raw
scores of Logical Memory II and Verbal Paired Associates II and converting
that sum first to a scaled score and then to an Index score. The Auditory
Recognition Delayed Index score does not represent an equal weighting of
the Logical Memory II and Verbal Paired Associates II scores. Compared to
Verbal Paired Associates delayed recognition, Logical Memory delayed
recognition has greater variance. Therefore, the Auditory Recognition
Delayed Index is more heavily weighted in the direction of Logical Memory.
Information retrieval based on recognition is generally easier than retrieval
based on free recall. Thus, low scores on the Auditory Delayed Index with
relatively better performance on the Auditory Recognition Delayed Index
may indicate retrieval difficulties.

General Memory Index

As stated previously, the composition of the General Memory Index of the
WMS-III represents a significant departure from the WMS-R. The WMS-III
Immediate Memory Index is analogous in composition to the WMS-R
General Memory Index because both are measures of immediate memory.
As shown in the following equation, the WMS-III General Memory Index
score represents a global measure of delayed memory and is made up of the
sum of scaled scores on the subtests that compose the Auditory Delayed
Index, the Visual Delayed Index, and the Auditory Recognition Delayed
Index:

Logical Verbal Paired Faces II Family Auditory General

Associates II ave Pictures I! Recognition Memory
Memory I!
modes) Daa he PRecal Recognition
+] Scaled |-F] “Recat |r Delayed: 1/72] 0 “Widex
Score Scaled Score Scaled Score Score
Scaled Score Scaled Score

Any differences between the subtest scaled scores that contribute to the
General Memory Index score should be considered in its interpretation.
To the extent that these component scores are substantially different, the
General Memory Index score may not represent a unitary global estimate
of memory functioning. The General Memory Index score, however, is
considered the best overall measure of the types of abilities that are critical
to effective memory in day-to-day tasks (i.e., memory of newly learned
information after delays during which intervening cognitive activity occurs).

207
Interpretive Considerations

Working Memory Index

The Working Memory Index of the WMS-III shares a subtest in common
with the WAIS-III: Letter-Number Sequencing. The WMS-III Working
Memory Index also includes the Spatial Span subtest, whereas the WAIS—III
Working Memory Index includes the Arithmetic and Digit Span subtests.
As such, the WMS-III Working Memory Index is equally weighted with one
auditorily presented task (Letter-Number Sequencing) and one visually pre-
sented task (Spatial Span). In contrast, the WAIS-III Working Memory Index
is composed of all auditorily presented tasks. The WMS-III Working Memory
Index score is calculated as follows:

Letter—
Number 4 Working
Spatial Span
Sequencing Scaled Score = Memeny
Index Score
Scaled Score

For nonimpaired individuals (i.e., the standardization sample), the Working

Memoty indexes of the WMS-III and the WAIS-III correlated very highly
(r = .82, p< .001). The WMS-III Working Memory Index also correlates very
highly with the WMS-R Attention/Concentration Index (see Chapter 4 of
this Manual).

The Working Memory Index is a measure of complex or high-level atten-

tional tasks that stress the ability to attend to information, to hold and pro-
cess that information in memory, and to formulate a response based on that
information. Low scores relative to overall intellectual functioning may
indicate a working memory weakness or impairment.

Auditory Process Composites

The four Auditory Process Composites—Single-Trial Learning, Learning
Slope, Retention, and Retrieval—were developed primarily on the basis of
theoretical considerations. In contrast to the Primary Index scores, which are
scaled on a standard-score metric, the composite scores are scaled on a per-
centile metric. Even though validation studies are presented in this Manual,
scores on these composites should be interpreted cautiously until further
validation studies are conducted. Although this Manual presents relevant
psychometric data concerning reliability and validity of the composite
scores, the clinician should understand the complex nature of these scores
(e.g., many are composed of difference scores, ratios, percentages, etc.). For
example, the Retention Composite score is a ratio of delayed memory to
immediate memory that is converted to a “savings” percentage score.

208
 Level of Performance

Moreover, the Retention Composite represents a relative measure of mem-

ory retention that controls the level of immediate acquisition. The level of
acquisition, however, modifies the interpretation of the Retention Com-
posite score. These conceptual and psychometric complexities should be
adequately appreciated by those who interpret the scores from these com-
posites. Due to the nature of the scores, interpretations based on the Audi-
tory Process Composites should be especially cautious, and conclusions
confirmed with data from multiple sources.

Single-Trial Learning Composite

The Single-Trial Learning Composite comprises the first trial of both stories
of Logical Memory I and the first trial of Verbal Paired Associates I, and its
score is calculated as follows:

Logical Verbal Paired Single-Trial

Memory | Associates | Learning
ist Recall 1st Recall Composite
Scaled Score Scaled Score Score

The Single-Trial Learning Composite is very highly correlated with the

Auditory Immediate Index. This composite is a measure of recall capacity
after a single presentation of material. Low scores, relative to attention and
overall intellectual capabilities, may indicate weaknesses or deficits in
immediate auditory memory. The Single-Trial Learning Composite also
forms the comparison standard for the evaluation of learning through repe-
tition (see the discussion of the Learning Slope Composite next).

Learning Slope Composite

The Learning Slope Composite is a measure of the relative increase from the
first trial to the last trial (i.e., last trial score minus first trial score) for Logical
Memory I Story B and Verbal Paired Associates I. The score is calculated as
follows:

Logical Verbal ; Paired Learnin

Memory | Associates | : g
Slope
Learning Learning Composite
Slope Slope ee
Scaled Score Scaled Score

High raw scores, that is, the difference score between the last and first pre-
Learn-
sentation trials, indicate a substantial increase in performance. High
ing Slope Composite scores indicate good learning performance relative to
from
the first trial, whereas low scores indicate a diminished ability to profit

209
Interpretive Considerations

multiple trials. The Learning Slope Composite should always be interpreted

in the context of initial performance (i.e., in the context of the Single-Trial
Learning Composite score). When initial performance is high, performance
cannot improve much over subsequent trials. For example, an individual
who recalls 7 of the 8 word pairs in Trial 1 of Verbal Paired Associates I can
improve his or her learning-slope score by only 1 raw-score point. Con-
versely, when initial performance is low, performance on subsequent trials
can greatly improve.

Retention Composite
Percent retention scores for the WMS were first proposed by Russell (1975).
Retention scores have been found clinically useful in differentiating
normally functioning individuals from individuals with dementia (e.g.,
Welsh, Butters, Hughes, Mohs, & Heyman, 1991) and may be useful in dis-
criminating among various clinical groups (Tréster et al., 1993). Normative
percent retention scores for the WMS-R have been reported by Prifitera
and Ledbetter (1992). Also, percent retention scores were shown to be
significantly related to level of education (Ledbetter & Prifitera, 1993).
The WMS-III Primary Indexes are measures of an individual's immediate
and delayed memory relative to a normative sample. Note that although
scores on the delayed indexes represent the individual’s performance at
the delayed point (relative to his or her peers), they do not represent the
individual’s efficiency in delayed memory relative to his or her own imme-
diate performance. In contrast, the Retention Composite is a measure of
delayed free recall relative to the individual’s performance in the immediate
condition. In effect, the individual is serving as his or her own control. The
Retention Composite score is calculated as follows:

Logical Verbal Paired

Memory II Associates || Retention
Percent Percent => Composite
Retention Retention Score
Scaled Score Scaled Score

Low scores on the Retention Composite suggest high rates of forgetting of

auditory information over a delay interval of 25-35 minutes. An adequate
evaluation of the retention of material at delayed recall must be based on at
least a minimal immediate score. An evaluation of delayed memory is diffi-
cult when initial acquisition is low. Therefore, percent retention scores must
always be interpreted in the context of initial acquisition. That is to say,
identical percent retention scores for two individuals may be obtained by

210
 Level of Performance

two very different levels of performance. A retention score of 50%, for exam-
ple, might be based on the examinee’s remembering two Logical Memory I
units at immediate recall and one unit at delayed recall. Another examinee’s
retention score of 50% might be based on his or her remembering 50 units in
the immediate condition and 25 units in the delayed condition.
Retrieval Composite
The Retrieval Composite score is calculated by subtracting the average of the
Logical Memory II and Verbal Paired Associates II recall scaled scores from
the combined Logical Memory II and Verbal Paired Associates II recognition
scaled score:

Auditory
Recognition LM II Recall VPA II Recall Retrieval
Delayed Scaled Score Scaled Score Total Score
Scale Score

The Retrieval Composite is a measure of the degree to which cuing increases

information retrieval beyond the amount of information available through
free recall. High scores suggest that more information may be available in
the examinee’s knowledge base than he or she can access through free,
unaided recall; that is, they may suggest that the person has a retrieval prob-
lem. Average scores suggest that free recall accesses most or all of the newly
learned material. Low scores suggest that recognition is poorer than recall
(possibly due to a ceiling effect). Identification errors during recognition
memory testing may reflect guessing or may suggest that nontarget stimuli
(i.e., foils or distractors) interfere with retrieval and that the examinee has
difficulty differentiating accurately between previously presented material
and incorrect alternatives.

Subtests
Information and Orientation
Information and Orientation is an optional subtest and does not contribute
to the WMS-III Index scores. Findings on this subtest can help the examiner
to
determine the appropriateness of a memory test or the examinee’s ability
be validly tested on the more complex and demanding components of the
such as
WMS-III. Relatively low scores may be caused by a variety of factors,
disorientation, aphasia, inattention, poor motivation, or a thought disorder.
take
To the degree that such factors affect test results, interpretation should

211
 Interpretive Considerations
UOHe}UIIIO
[PIO]

68-s8OO 6g Von ce g g g €

8-0800 eh OLR acl g € | | |

me
«=
GL-SL~—0Ok_——00k_—SsO0L_—S—s“00,—S
Solas fOr g € € |
UONeULIOFU
pue ]

LOL ee
zl eet, | L

=69-S9 meme b
g

—H9-SSHS-Sh
satsuanba1q dANL[NUIND) (saseyUad1e

fee esS |
“dnoiy

ss
JO

Ok—Ss«CES
re | |
aby

Hh-GE ee 8 p

—PE-DE«CL 2c 8 |

©
62-SZSs«CkSCtsC«CéiSCs«CéiS
cee g €

Pe-0Z Of
a
Aq asy

GI-BL Ber
sa109S

LI-91SC ce g L

—«-
ajqel"e°G

aug
vl el Zt LL OL 6 8 Z 9 g
jeJO) v € Z L 0

212
 Level of Performance

these into account. As shown in Table 5.3, individuals in the standardization

sample rarely obtained raw scores below 10 on Information and Orientation.

Logical Memory
With some differences in the content, administration, and scoring, the WMS,
WMS-R, and WMS-III have all included the Logical Memory subtests. The
WMS-III Logical Memory subtests contain a new story (i.e., Story B) and
revised administration and scoring procedures. The Logical Memory I Total
Score is computed by summing the recall units for Story A and the recall
units for both trials of Story B. The second trial of Story B was added for two
reasons. First, performance on the two recall trials of Story B can be con-
trasted (i.e., the Learning Slope scaled score can be obtained) to evaluate
any deviations from expected change from the first to second trials. Also, the
second trial of Story B increases the likelihood that the examinee will “learn”
or acquire enough of the material to be remembered later in the delayed
condition. This repetition helps to ensure that initial learning is maximized
because retention of material can be evaluated only in the context of the
material initially learned.
Low scores on the Logical Memory | and II recall measures may suggest
deficits or weaknesses in learning of or memory for conceptual material pre-
sented in the auditory modality. The supplemental thematic recall scores
can be compared to the literal or near-literal recall unit scores. Thematic
recall is a measure of the examinee’s ability to remember thematic informa-
tion, which is more general than the specific and literal information that is
scored for the Primary Indexes. For example, in the first thematic unit for
Story A, the examinee has only to indicate that the story has a female char-
acter to receive credit. In contrast, to earn credit for the story recall units, the
examinee must identify the character’s name. The Logical Memory II percent
retention score represents the examinee’s retention of material from the
immediate condition to the delayed condition.
Verbal Paired Associates
Results from studies of the WMS-R have shown that the hard items on the
Verbal Paired Associates tasks correlate more highly with other measures of
verbal memory (Macartney-Filgate & Vriezen, 1988) and are more Clinically
sensitive (Fisher, 1988; Trahan, Larrabee, Quintana, Goethe, & Willingham,
1989) than the easy items. Therefore, revisions of this subtest have included
the replacement of overlearned or easily acquired word associations (e.g.,

213
Interpretive Considerations

Baby-Cries) with novel and unrelated word pairs (e.g., Star—Ladder). For the
WMS-III Verbal Paired Associates, four learning trials of the word-pair list
are administered. For the delayed condition (i.e., Verbal Paired Associates IJ),
the examiner reads the first word of each pair, and the examinee provides
the second, or “associated,” word (i.e., cued recall). For the recognition task,
administered after the delayed cued-recall task, the examinee is read a list of
24 word pairs and must identify each word pair either as one presented in
the previous conditions or as a new one. Low scores on the Verbal Paired
Associates I and II recall measures may suggest deficits or weaknesses in
learning or memory for auditory material. In contrast to the material pre-
sented in Logical Memory I and II, the stimulus material of Verbal Paired
Associates I and II is semantically unrelated. Verbal Paired Associates there-
fore requires the examinee to organize the material more actively. The Verbal
Paired Associates II percent retention score represents the examinee’s reten-
tion of material from the immediate condition to the delayed condition.

Word Lists
The word list is a familiar paradigm for assessing learning and memory.
The WMS-III Word Lists subtest is optional (i.e., none of the Word Lists
scores contributes to an Index score). The Word Lists subtest incorporates
a full reminding procedure, as in the CVLT (Delis et al., 1987), but unlike
the CVLT, the WMS-III subtest does not organize the words into semantic
categories.
The Immediate Recall score, which is the sum of List A Trials 1—4, is a mea-
sure of immediate recall ability for unstructured material that is repeated
during learning. High scores indicate efficient learning and immediate
recall, whereas low scores indicate weaknesses or deficits in learning and
immediate recall. The Delayed Recall and Delayed Recognition tasks assess
information retention and retrieval after a delay of 25-35 minutes. The Word
Lists percent retention score is calculated by dividing the Delayed Recall
score by the Trial 4 score of Word Lists I and multiplying by 100. This score
provides a measure of delayed information-recall efficiency relative to the
amount of material previously accessible. Two contrast scores are also
calculated: List A Trial 1 versus List B, and List A Trial 4 versus List A Short-
Delay Recall. The first contrast score quantifies performance differences on
the first trial of each list. The second contrast score indicates whether List A
consolidation in memory is adversely affected by the presentation of a word
list that is similar in content and structure to the the first list, or whether List
B learning is rendered more difficult by prior exposure to List A (proactive
interference).

214
 Level of Performance

Faces
Faces is a subtest new to the WMS-III and was added because of findings
from research with tests using similar paradigms. Results from studies have
suggested that memory for faces is sensitive to right hemisphere deficits
(Carlesimo & Caltagirone, 1995; Newcombe, de Haan, Ross, & Young, 1989;
Schweinberger, Buse, Freeman, Schonle, & Sommer, 1992) and to right
temporal lobe and hippocampal lesions. Memory for faces has also been
associated with increased right temporal cerebral blood flow (R. C. Gur et al.,
1993) and right parietal metabolism (Berardi, Haxby, Grady, & Rapoport,
1991). The memory for faces paradigm has also been found to be sensitive to
the effects of right versus left temporal lobe epilepsy in children (Beardsworth
& Zaidel, 1994) and adults (Naugle, Chelune, Schuster, Liiders, & Comair,
1994). Low scores on Faces may indicate memory weaknesses or deficits
when material is presented visually.

Family Pictures
The recall of stories often involves recall of people, objects, places, and
events. The WMS-III Family Pictures subtest was designed to assess recall for
scene characters, character activity, and character location. Family Pictures
is a new subtest not only to the WMS-III but also to clinical practice and
research. Research with tasks that assess similar dimensions have indicated
that patients with schizophrenia exhibit impaired spatial location and con-
text (Rizzo et al., 1996). Persons with right hippocampectomy for treatment
of temporal lobe epilepsy have been found to exhibit rapid forgetting for
spatial location (M. Smith & Milner, 1989). The interpretation of Family
Pictures scores should take into account the relative weight the scoring pro-
cedure places on the three scoring elements (i.e., memory for character,
activity, and location). Scores are character-based; that is, credit for correct
activity and location is awarded only in conjunction with scene characters
that are correctly identified. Unlike with the Faces subtest, most aspects of
Family Pictures can be represented (and presumably encoded) verbally.
Visual Reproduction
Visual reproduction has been related to left hippocampal volume in individ-
uals with traumatic brain injury (Bigler et al., 1996) and to right versus left
hippocampal atrophy in female examinees with right hemisphere temporal
lobe epilepsy (Trenerry, Jack, Cascino, Sharbrough, & Ivnik, 1996). Results of
studies of other versions of this test have indicated that it is not differentially
sensitive to right hemisphere lesions (Chelune & Bornstein, 1988; Naugle et
al., 1993) but may reveal memory impairment in specific groups (Butters et
al., 1988; Fisher, 1988). Research has indicated that performance on visual
reproduction subtests may be confounded by constructional dyspraxia in

215
Interpretive Considerations

certain clinical groups (Haut et al., 1994) but that visual memory deficits can
be detected if these effects are controlled (Haut et al., 1996). Also, experience
indicates that this is one of the more difficult and time-consuming WMS and
WMS-R subtests to score.

Retained from the WMS and WMS-R, the Visual Reproduction subtest is an
optional subtest in the WMS-III. The WMS-III subtest includes two new
design cards (A and E); one WMS-R card (Card B) was deleted. In addition to
the immediate and delayed recall conditions, the WMS-III Visual Reproduc-
tion subtest includes a delayed recognition condition followed by a direct
copy task and a discrimination condition. The scoring criteria were also
changed on the basis of several studies that compared responses by clinical
groups and matched normal control groups. In most cases, the scoring
criteria now allow for partial credit.
These changes now allow examiners to make comparisons between delayed
recall and recognition. The direct copy score can be compared to the recall
measures to assess possible motor-control effects on drawing ability. The
discrimination condition can reveal visual-perceptual distortions that might
have adversely affected the examinee’s learning and memory. Table 5.4 pre-
sents the cumulative frequencies for the discrimination condition for each
of the 13 standardization age groups. As shown, raw scores below 5 are rare
at any age. The Visual Reproduction percent retention score is a measure of
the examinee’s retention of material over a delay of 25-35 minutes.

Letter-Number Sequencing
The Letter-Number Sequencing subtest is a measure of auditory working
memory. The development of this task was based, in part, on the work of
Gold et al. (1997). Research has shown that this task is differentially sensitive
to a variety of neurocognitive disorders. Also, the ability to perform the task
may be spared in individuals with anterograde amnesia (see Gathercole,
1994, and Kopelman, 1994, for a review). The WMS-III Letter-Number
Sequencing subtest requires the examinee to order sequentially a series of
numbers and letters orally presented in a specified random order. The exam-
inee must first remember the numbers and letters and then reorganize the
numbers into ascending order and the letters into alphabetical order.

Spatial Span
Spatial Span is the visual analogue of the Digit Span subtest. Previous ver-
sions of spatial span tasks include those designed by Corsi (1972) and by
E. Kaplan et al. (1991). Administration is similar to that of the WMS-R sub-
test except that a three-dimensional board is used rather than the two-
dimensional card. The WMS-III Spatial Span Board is a modification

216
 alge“WG safuanbary VAe[NUIND) (saseyUed1eg
Jo [enst, UoHonposday uoneUTUTLsIq]
[210], $3.109¢
Aq a3y
aby dnoiy
/RIOL a09S I-91 GI-BL F2-02 62-SZ PE-OE r-SE pPS-Sh 9-SG 69-S9 pL-OL G6L-SL 8-08 68-S8

Ls 9
00l 8
OO! Bl L
O0| cc L

OO! co v
OOl 61 4
00| ve €
001 62 6

001 Le é
OOl 6€ i.

OOL 9€ Z L
OO! cv 5) L
00} c& 6 S
L
OO! €9 62 LL €

& v € é L
0
Level of Performance

217
Interpretive Considerations

of the one designed by E. Kaplan et al. (1991). Spatial Span taps the exami-
nee’s ability to hold a visual-spatial sequence of events in working memory.

Digit Span
Digit Span is now an optional subtest on the WMS—III and does not con-
tribute to any of the indexes. The Digit Span subtest is composed of Digit
Span Forward and Digit Span Backward, which, as noted by E. Kaplan et al.
(1991), may tap different functions. Digit Span Forward is a measure of
focused attention, whereas Digit Span Backward demands more effort from
working memory. Because Digit Span is optional, normative information
only for the combined Digit Span Forward and Digit Span Backward is
reported. (For information on evaluating discrepancies between Digit Span
Forward and Digit Span Backward, see the discussion of this subtest in the
WAIS-III section.)

Mental Control
The Mental Control subtest of WMS-III is also optional and does not con-
tribute to any of the indexes. The WMS-III version reflects expanded content
and changes in the scoring criteria. This subtest is a measure of the exami-
nee’s ability to retrieve overlearned information and to mentally process
information. For example, the examinee is asked to say the alphabet and to
perform novel multitasking skills, such as alternately saying the days of the
week and counting by sixes. Subtest scores reflect both accuracy and speed.
Bonus points are awarded for quick, perfect performance.

Patterns and Profiles of Performance

Historically, two primary schools of thought have addressed issues of inter-
scale variations that are commonly used in the interpretation of test scores.
Spearman (1904, 1932/1970) argued that most cognitive abilities (e.g., those
measured by most WAIS-III and WMS-III subtests) should be approximately
equal and reflect a general cognitive factor, or g. In contrast, E. L. Thorndike,
Lay, and Dean (1909) argued that “scatter” or ability differences (e.g.,
subtest-to-subtest or index-to-index) are the norm in healthy individuals.
Thorndike et al.’s arguments should caution clinicians not to overinterpret
ability differences or scatter. An excellent discussion of these issues can be
found in Matarazzo (1990). A brief description of examining intrascore varia-
tions, or “score scatter,” in the WAIS-III and the WMS-III is presented here.

218
 Patterns and Profiles of Performance

Performance on the WAIS-III and the WMS-III can be interpreted in terms

of the patterns of an examinee’s various scores. Comparing an individual’s
functioning in one cognitive area to his or her functioning in another area
can generate hypotheses about spared and impaired cognitive abilities.
These ability comparisons are based on pairwise statistical comparisons
between IQ and Index scores, which can help the examiner identify poten-
tially meaningful patterns of strengths and weaknesses, a process that is
important in describing functional impairment and for planning rehabilita-
tion protocols.
The most appropriate use of profile analyses is the generation of hypotheses
that are, in turn, either corroborated or refuted by other evaluation results,
background information, direct behavioral observations, or additional eval-
uation. Profile analyses occur mostly at the IQ and Index score levels and
are based on statistically significant differences between these scores and
the clinically important frequency (or base rate) of the score differences
in the standardization sample (Matarazzo, 1990). Interpretations based
on profile analysis are often stated in a manner similar to the following
example:
This individual demonstrated a [mild, moderate, severe] weakness in
{name or description of the IQ or Index] relative to [mame or description
of the other IQ or Index]. The discrepancy is [both statistically significant
and rare in nonimpaired persons, or statistically significant but not
uncommon in normally functioning persons].
Although statistically significant pairwise differences occur in many im-
paired populations, the same ability differences may also occur frequently in
the normally functioning population (Matarazzo, 1990). Therefore, it is criti-
cal for the examiner to know the frequency of discrepancies between scores
in the WAIS-III and WMS-III standardization samples (e.g., see Matarazzo &
Herman, 1985, for a discussion of this issue pertaining to the WAIS-R). Base-
rate information provides a basis for estimating the rarity or commonness of
the examinee’s obtained difference within the normal adult population.
A discrepancy that is statistically significant yet frequent in the standardiza-
tion sample probably reflects normal variations in an individual's abilities.
In contrast, a discrepancy that is both statistically significant and rare in
the standardization sample could represent a meaningful and substantial
decrease in at least one of the abilities being compared. All relevant explana-
tions should be considered, including the possibility that the ability is im-
paired due to brain dysfunction. In general, the larger the discrepancy and
the less frequent its occurrence, the less likely it can be explained as normal
variation.

219
Interpretive Considerations

The following example illustrates the possible interpretation of a statistically

significant and infrequent difference between two scores:
A 74-year-old, high-school-educated female examinee with no prior
history of neurologic or psychiatric difficulty obtained a VIQ—PIQ score
difference of +33 points. This difference is statistically significant at the
p< .05 level, and the base rate for this occurrence is 0.5%, meaning that
a difference of this magnitude occurred in less than 1% of the standard-
ization sample. Because of the statistical significance and high infre-
quency of this difference in the general population, an appropriate
hypothesis might be that this individual’s nonverbal intellectual func-
tioning is impaired relative to her verbal intellectual functioning.
The next example illustrates the possible interpretation of a statistically
significant but frequent difference between two scores:
A 20-year-old, male college student with no history of neurologic, psy-
chiatric, or academic difficulties obtained an Immediate Memory Index
score that was 16 points higher than his Working Memory Index score.
This difference is statistically significant at the p < .05 level and occurred
in approximately 29% of the standardization sample. Although this dif-
ference is statistically significant, it occurs with such frequency (i.e.,
nearly 1 of 3 individuals obtained this difference) in the general popula-
tion that it is neither unusual nor, in and of itself, indicative of impair-
ment. In the absence of other evidence, this discrepancy does not
necessarily indicate a deficit in working memory relative to immediate
memory.
The following descriptions of selected WAIS-II] and WMS-III differences are
suggested interpretive guidelines for comparing selected pairs of scores.
Several other interpretations might be possible in addition to the ones sug-
gested here. Any interpretation should always be considered an hypothesis
to be evaluated in the context of a thorough clinical evaluation.

WAIS-IlI Discrepancy Analyses

Several types of discrepancy scores can be calculated within the WAIS-III.
They can be obtained at the subtest level, the IQ level (i.e., differences
between VIQ and PIQ scores), and the index level. Discrepancy scores can
also be obtained across test batteries. WAIS-III and WMS-III discrepancy
scores are provided in this Manual (see Appendixes B and C) and can be
recorded on the WMS-III Record Form. Discrepancies between the WAIS-III
and the WIAT-II achievement scores can also be calculated for examinees
aged 16-19 and recorded on the WIAT-II Record Form. The tables for these
discrepancy analyses are listed in Appendixes B and C.

220
 Patterns and Profiles of Performance

Subtest Score Patterns and Discrepancies

Most individuals have areas of relative cognitive strengths and weaknesses.
It is, in fact, very uncommon for a “normal” person to function at the same
level in every ability area.
The WAIS-III Record Form provides a section for determining the individ-
ual’s strengths and weaknesses at the subtest level. These strengths and
weaknesses are calculated in terms of the Full Scale or in terms of the Verbal
and Performance scales. The examiner must choose whether to use a mean
score of all the subtests that were administered (the overall global mean) or
to use the mean scores of the Verbal and Performance subtests. If the latter
method is used, the mean of all of the Verbal subtests administered is the
base against which each Verbal subtest score is compared, and the mean
score of all Performance subtests administered is the base against which
each Performance subtest score is compared. Table B.3 in the WAIS-III
Administration and Scoring Manual includes both the significance of the
difference (at .05 and .15 levels) as well as the differences obtained by vari-
ous percentages of the normative sample. Using this frequency data, the
examiner can decide how rare the obtained difference is in a normative
sample.

IQ Score and Index Score Discrepancies

In constructing the IQ scores for the Wechsler—Bellevue, Wechsler (1939)
placed most of the emphasis on the FSIQ score and believed that an exami-
nee’s FSIQ score is always an average of the person’s performance on all of
the subtests (Wechsler, 1944). However, Wechsler realized that occasions
arose when the VIQ and PIQ scores must be viewed separately, usually for
persons “with special disabilities who need special consideration” (Wechsler,
1944, p. 138).
Since the publication of the Wechsler—Bellevue, the VIQ-PIQ difference score
has become a more common method of determining when to modify the
interpretation of an FSIQ score and to examine the VIQ and PIQ scores sepa-
rately. With the publication of the WAIS-R in 1981, Wechsler included a table
to show the minimum differences between the VIQ and PIQ scores required
for significance at the .15 and .05 levels of confidence for each age group.
Generally, a difference score of 10 points was required at the .05 level, anda
difference of 7 points at the .15 level. These values became “rules of thumb,”
and clinicians started applying them clinically with their examinees.
differ-
Matarazzo and Herman (1985) first documented how frequently these
n sample and that a relativel y
ences occurred in the WAIS-R standardizatio

221
Interpretive Considerations

large difference may not be rare in the general population. They demon-
strated the need for examining statistical significance as well as clinical
meaningfulness (base rates) and cautioned the clinician against overinter-
pretation of VIQ-PIQ score differences. When comparing discrepancy scores,
the examiner must review other variables (e.g., the psychosocial history,
education level) in addition to the statistical significance and clinical mean-
ingfulness of the test scores. When a significant and meaningful difference is
found, psychologists might interpret the FSIQ score differently, in a way that
reflects these differences. A variety of detailed interpretation schemes has.
been suggested to explain meaningful differences (e,g., Kaufman, 1990, 1994;
Sattler, 1992).

In addition to the VIQ-PIQ score differences, the WAIS-III includes norma-

tive discrepancy information on all possible pairs of Index scores. The VCI-—
POI score comparison is similar to the VIQ—-PIQ score comparison except
that Arithmetic, Digit Span, and Comprehension are not included in the VCI,
and Digit Symbol—Coding and Picture Arrangement are not included in the
POI. The VCI-POI discrepancy score may be useful for testing hypotheses
about the effect of these excluded subtests on the respective VIQ and PIQ
scores. For instance, if the examiner suspects that the examinee has relative-
ly lower ability on tasks requiring attention, working memory, or processing
speed, which would affect VIQ and PIQ scores, he or she can examine Verbal
and Performance differences using the VCI-POI comparison, for which these
abilities are not emphasized.
The VCI-WMI score comparison can reveal differences between the individ-
ual’s capacity to hold and process information in memory and his or her
acquired knowledge and verbal reasoning skills. Similarly, the POI-PSI score
difference can reveal differences between an individual's visual-spatial and
fluid reasoning skills and his or her ability to process information quickly.
Because the subtests that compose the VCI and WMI are all verbal and the
subtests that compose the POI and PSI are nonverbal, these discrepancy
comparisons may provide meaningful information.
Tables B.1 and B.2 of the WAIS-III Administration and Scoring Manual pro-
vide the statistical significance and frequency data necessary for analyzing
these various discrepancy scores. Because Matarazzo and Herman (1985)
have demonstrated that frequencies of score differences can vary by ability
level, the frequencies of differences for various levels of FSIQ scores are
provided in Appendix D of this Manual. The user is cautioned, however, that
when an individual has a neuropsychological disorder or condition that may
affect cognitive functioning, the individual’s current ability level may be
lower than his or her premorbid functioning (see Heaton et al., 1978;
Parsons & Prigatano, 1978). In such a case, the frequencies of differences by

222
 Patterns and Profiles of Performance

levels of FSIQ score reported in the tables may be misleading and should be
used with caution.

Directional Discrepancy Scores

The WAIS-III Administration and Scoring Manual presents cumulative fre-
quencies for composite-score discrepancies irrespective of the direction of
the difference (Table B.2). For example, ifVIQ minus PIQ equals —10, the
value reported in the frequency table is 10. It is not possible to determine
how often the VIQ will be greater than the PIQ or vice versa; however, it is
possible to know how often the VIQ and PIQ differ by 10. The cumulative fre-
quencies provide an estimate of the base rate that the degree of difference in
performance is observed in the general population. This method of comput-
ing discrepancy scores presumes the clinician has no a priori hypotheses
regarding the clinical profile of the respondent. Tulsky, Zhu, and Vasquez
(1998) reported that the use of discrepancy scores that do not indicate the
direction of the difference may overestimate the frequency of the difference
score when the clinician has specific expectations regarding the relationship
among the IQ measures. For instance, the clinician may expect that for an
examinee diagnosed with Huntington's chorea the PIQ will be lower than the
VIQ due to relatively preserved verbal versus visual—perceptual skills in this
clinical group (see Table 4.31 of this manual). Tulsky et al. (1998) reported
that the distribution of difference scores was normally distributed and
presented a methodology for approximating the cumulative frequency of
directional discrepancies for WAIS-III IQ and Index scores. Sattler and Ryan
(1999) applied this methodology to the WAIS-III IQ and index scores,
making the frequencies available to clinicians. Tulsky, Rolfhus, and Zhu
(2000) evaluated the accuracy of the methodology for estimating the base
rates for directional discrepancies. The results of the study indicated a close
approximation of estimated versus actual directional discrepancy score base
rates (Tulsky et al., 2000). The cumulative frequencies for the actual direc-
tional discrepancies are presented in the article. Clinicians should use the
standard discrepancy tables for evaluations for which they have no expecta-
tions regarding the performance of the examinee, as will be the case most of
the time. In situations where the examine already has a diagnosis, suspected
condition (e.g., dementia), or known medical condition and the clinician
expects the examinee to display higher or lower scores on specific measures,
then the examiner should use the new directional discrepancy tables. The
tables of directional discrepancy-score base rates for the WAIS-III 1Q and
Index scores (Table D.6) and the WMS-III Index scores (Table D.7) are pro-
vided in Appendix D of this manual.

223
Interpretive Considerations

Other Score Discrepancies

The introduction of new indexes and the co-norming and codevelopment
of the WAIS-III and the WMS-III have made possible the comparison of a
variety of scores, at both the subtest and index levels. Because these com-
parisons are new, their clinical utility and meaning are speculative. They
are, however, fertile ground for research and clinical investigations and
may prove to be meaningful and important. These score differences should
be interpreted cautiously until further research establishes their meaning,
validity, and diagnostic utility. Providing a detailed, systematic, and “com-
plete” method of protocol analysis is beyond the scope of this Manual (see
Kaufman, 1990, 1994, and Sattler, 1992, for detailed guides on interpretation
strategies for other Wechsler intelligence measures). Nevertheless, examiners
are advised against taking a “shotgun” approach to interpretation (Kaufman,
1994). Until more experience has been accumulated and research per-
formed, the examiner should have a clear reason for calculating a difference
score, and that reason should be based on the examinee’s history, the referral
question, behavioral observations, and other test results. It should also be
remembered that a difference between two scores may be clinically mean-
ingful for one individual, whereas the same difference for another individual
may not be.

WMS-Ill Discrepancy Analyses

Primary Index Score Comparisons
immediate Versus Delayed Indexes
The comparisons between immediate, or short-term, memory and delayed,
or long-term, memory are perhaps the most common in memory assess-
ment. Relevant WMS-III comparisons include those between the Immedi-
ate Memory Index and the General Memory Index, for an analysis of global
differences, and between the Auditory Immediate and Auditory Delayed
indexes and between the Visual Immediate and Visual Delayed indexes,
for an analysis of modality-specific differences. Low delayed performance
relative to immediate memory performance (considered in the context of
intellectual functioning and attentional abilities) may indicate weaknesses
or deficits in the examinee’s ability to retain previously learned material.
Auditory Versus Visual Indexes
Differences in performance on the auditory and visual indexes (i.e., Auditory
Immediate versus Visual Immediate and Auditory Delayed versus Visual
Delayed) may indicate lifelong strengths and weaknesses or acquired deficits
in memory processes when information is presented in different modalities.

224
 Patterns and Profiles of Performance

Other cognitive abilities (e.g., attentional abilities, receptive and expressive

language abilities, perceptual organizational abilities, and vocabulary
and articulation) may also influence differences in auditory and visual
presentation.

Working Memory Versus Immediate Memory Indexes

The Working Memory Index score can be compared to the Auditory Imme-
diate Index, Visual Immediate Index, and Immediate Memory Index scores
as a means of evaluating possible differences between complex attentional
abilities and the acquisition and encoding processes (i.e., learning and
memory), which depend on attention. A low Working Memory Index score in
the context of a low Immediate Memory Index score and relatively higher
intellectual functioning may indicate that attentional abilities are affecting
the examinee’s ability to learn the material initially.

Auditory Delayed Index Versus

Auditory Recognition Delayed Index
The comparison between scores on the Auditory Delayed Index and Audi-
tory Recognition Delayed Index is essentially a comparison between delayed
recall and delayed recognition. A profile of scores in which performance is
adequate in the immediate conditions but low in both delayed recall and
recognition conditions suggests that the examinee was not able to retain
previously learned information over an interval of 25-35 minutes. Because
retrieval through recall is more demanding than retrieval through recogni-
tion, a low Auditory Delayed Index score relative to the Auditory Recognition
Delayed Index score may suggest some type of retrieval weakness or deficit.
Again, it is important for the examiner to consider all recall-recognition
comparisons in the context of initial learning.

Auditory Process Composite Score Comparisons

Single-Trial Learning Composite Versus
Learning Slope Composite
the Learning Slope
A low Single-Trial Learning Composite score relative to
ate auditory
Composite score suggests that although the examinee’s immedi
due to anxiety—check
memory may appear to be weak or impaired (possibly
es over multi-
measures of attention and working memory), memory improv
Learning Composite
ple learning trials. Low scores on both the Single-Trial
memory difficulties
and the Learning Slope Composite suggest immediate

225
Interpretive Considerations

and a decreased capacity to learn from subsequent learning trials. Compari-

sons between the Single-Trial Learning Composite and the Learning Slope
Composite scores should not be made when the Single-Trial Learning
Composite score is very high. When the Single-Trial Learning Composite
score is relatively high (i.e., the examinee has learned much or most of the
material on the first trial), performance cannot improve much.

Retention Composite Versus Auditory Delayed Index

A comparison between the Retention Composite and Auditory Delayed
Index highlights the difference between intraexaminee and interexaminee
comparisons. A high Retention Composite score in the context of a low
Auditory Delayed Index score indicates that the examinee’s delayed recall
performance is low compared to that of the normative group (interexaminee
comparison). This pattern of scores also indicates, however, that the exami-
nee could later recall the information, at a similar level of performance, that
was learned during the immediate condition (intraexaminee comparison).
This examinee would also be expected to have a poor Immediate Auditory
Index score, confirming weak learning but adequate retention. In this case, a
relatively good recognition score (compared to the Auditory Delayed Index
score) would suggest an additional retrieval weakness.

Differences Between the WAIS-III

and the WMS-III
Discrepancies between intelligence and memory are sometimes used to
evaluate memory functioning. With this approach, learning and memory are
assumed to be underlying components of general intellectual ability and, as
such, to be significantly related to the examinee’s performance on tests of
intellectual functioning. Because of the relatively high intercorrelations
between intellectual functioning and memory (see Chapter 4), the exami-
nee’s IQ scores become an index or estimate of his or her probable level of
memory ability. Discrepancies between the estimated memory performance
based on IQ scores and the examinee’s actual memory performance form
the basis for the discrepancy analysis. The discrepancy score then provides
a global indication of whether or not the examinee’s ability to learn and
remember new material is commensurate with what would be expected on
the basis of his of her intellectual functioning.
Some researchers have proposed that a discrepancy in which IQ scores are
appreciably higher than memory scores may be suggestive of an acquired
memory impairment (Milner, 1975; Prigatano, 1974; Quadfasel & Pruyser,
1955). Others (e.g., Butters, 1986) have argued that IQ-memory discrepan-
cies may not be sensitive to various patterns of memory deficits that are

226
 Patterns and Profiles of Performance

observed in particular clinical populations. Bornstein, Chelune, and

Prifitera (1989) compared the base rates (frequencies) of 1Q-memory score
discrepancies obtained by a mixed clinical sample diagnosed with memory
impairment and by the WMS-R standardization sample. Their results indi-
cated that the discrepancy between the FSIQ score and the Delayed Memory
Index score has some clinical utility. Specifically, a discrepancy of 15 points
was obtained by only 10% of the standardization sample but by about 33%
of the clinical sample. Bornstein et al. noted, however, that the direct inter-
pretation of the IQ-memory score discrepancy may be confounded by other
factors. They correctly pointed out that the IQ-memory score discrepancy
does not take into account overall level of performance, estimates of pre-
morbid abilities, or the possible effects of neurological disorders and other
specific neurocognitive functions on intellectual and memory functioning.
An examinee with dementia, for example, might have experienced a global
cognitive decline, that is, across a broad spectrum of many abilities that
includes memory. In this instance, a low IQ-memory score discrepancy does
not indicate the absence of a memory deficit but, rather, indicates only that
the examinee’s level of memory performance, relative to global cognitive
abilities, is commensurate (i.e., low). Alternatively, the examinee’s level of
memory performance may indeed be in the impaired range. The IQ—
memory score discrepancy, therefore, should always be interpreted in the
context of current level of intellectual and memory functioning and esti-
mated or known premorbid functioning. To the extent that an examinee’s
dis-
overall intellectual functioning has remained relatively stable, a large
crepancy score in which the IQ score is greater than the memory score
of
suggests a more focal memory impairment or weakness in the context
both intellectual
relatively intact overall abilities. In other instances in which
discrepancy is
functioning and memory decline, the absence of a significant
where memory is one of
consistent with global neurocognitive impairment,
many cognitive abilities that have declined.
memory function-
The interpretation of differences between intellectual and
the interpretation of
ing follows the same logic and methodology as used in
a and rationale for
ability-achievement discrepancies. (The specific criteri
evemen t differe nces can be found in The
evaluating these global ability-achi
and later in this chapter .) An essential cri-
Psychological Corporation, 1992,
score compar isons is the compar ability of
terion for the appropriateness of
measur es. They must be highly compara-
the normative data for the two test
II were co-normed,
ble or the same. Because the WAIS-III and the WMS-I
same normat ive data can be made.
direct comparsions based on the
d for WAIS-III and WMS-III
Two methods of discrepancy analysis are offere
the predicted-difference
comparisons: the simple-difference method and
the predicted-difference
method. Although both methods are presented,

227
Interpretive Considerations

method is generally preferred because of two primary considerations. The

formula for the predicted-difference method not only takes into account the
reliabilities and the correlations between the two measures but also corrects
for regression to the mean.
In general, the FSIQ score should be used as the best estimate of intellectual
ability. However, if the difference between an examinee’s VIQ and PIQ scores
is 10 points or more (p < .05), the higher of these two IQ scores could be ©
used instead of the FSIQ score as the best estimate of intellectual ability.
Furthermore, the General Memory Index score is usually the best estimate of
an examinee’s memory functioning. In some cases, however the direct inter-
pretation of the General Memory Index score may be obscured by significant
differences between component subtest scores. For example, if the differ-
ence between auditory and visual subtest scores is significant and meaning-
ful, the modality-specific Index scores may be more appropriate than the
General Memory Index score (i.e., the concept of “overall” memory becomes
less meaningful in this case). This Manual provides discrepancy-score tables
for comparing the WAIS-III VIQ, PIQ, and FSIQ scores and the WMS-III
Primary Index scores.

Simple-Difference Method
With the simple-difference method, aWMS-III Primary Index score (usually
the General Memory Index score) is subtracted from the WAIS-III IQ score
(usually the FSIQ score). This method is sometimes selected because it is
easy to explain to examinees, families, or other health professionals. How-
ever, as Braden and Weiss (1988) pointed out, the simple subtraction of stan-
dard scores assumes a perfect correlation between the two scores. Further,
unless the simple difference is tested for statistical significance, measure-
ment error is also ignored.
The proper use of the simple-difference method requires the examiner to
determine first if the difference is statistically significant, and if it is, to deter-
mine how frequently a difference of its size occurred in the standardization
sample (Berk, 1984). These two steps should be familiar because they are the
same ones used to interpret the difference between WISC-III VIQ and PIQ
scores (Wechsler, 1991).
First, the statistical significance of the difference between two scores
accounts for measurement error and allows the examiner to conclude if the
difference is a “real” or a “chance” occurrence. That is, the difference must be
of sufficient size to minimize the probability that it has occurred because
of unreliability in the measures. Formulas for determining the statistical
significance of simple differences are identical to those for calculating the
ability-achievement difference scores (The Psychological Corporation, 1992;

228
 Patterns and Profiles of Performance

also see Reynolds, 1985, 1990) and were used for the construction of the
tables in Appendix C. Tables C.1-C.3 provide the differences between the
WAIS-III FSIQ, VIQ, and PIQ scores and WMS-III Primary Index scores
required for statistical significance at the .05 and .01 levels. The difference
score (calculated according to the steps in Chapter 3 of the WMS-III
Administration and Scoring Manual) must be equal to or greater than the
listed value to be statistically significant. For example, for a 63-year-old indi-
vidual, an FSIQ—General Memory Index score difference of 11 is statistically
significant at the .05 level because the difference is greater than 10.2, the
listed value (Table C.1). This score difference, however, is not statistically
significant at the .01 level because it is less than 13.4, the listed value for this
age group.
The second consideration for interpreting the difference between two scores
is the frequency of that difference within the general population. That is, the
difference must be of a magnitude that is relatively rare in the sample that
links the two measures (i.e., the co-normative sample of the WAIS-III and
the WMS-III). Even though a difference is statistically significant, it may
occur frequently. Tables C.4—C.6 list the differences between the WAIS—III
FSIQ, VIQ, and PIQ scores and the WMS-III Primary Index scores obtained
by various percentages of the standardization sample. It is important to note
that Tables C.4-C.6 present percentages of the standardization sample who
obtained WMS-III Index scores lower than their IQ scores. For example, an
FSIQ-General Memory Index score difference of 28 points occurred in 2%
of the WAIS-III—WMS-III standardization sample (Table C.4), indicating
that it is quite rare to obtain a discrepancy of this size in the general
population.

Predicted-Difference Method
Shepard (1980) was one of the first to advocate a predicted-difference
method based on the correlation between the two variables. For discrepan-
cies between ability and memory, the ability score is used in a regression
equation to calculate a predicted memory score. The prediction formula
used for the discrepancies between WAIS-III and WMS-III scores is identical
to the ability-achievement formula used by The Psychological Corporation
(1992). One noteworthy limitation of the predicted-difference method is that
scores
when correlations between measures are low, the range of predicted
is restricted. Berk (1984) summarize d the disadvanta ges of the predicted
-
method, pointing out the limitations of imperfect correlations (discrepan
cies are due to prediction error as well as true differences).
n the
The predicted-difference method is based on the difference betwee
the memor y score actuall y
memory score predicted from the IQ score and

229
Interpretive Considerations

obtained by the examinee. As with the simple-difference method, in order

for differences to be meaningful, the difference score should be statistically
significant and rare. Tables B.1-B.3 provide the estimated (predicted)
WMS-III Primary Index scores based on the WAIS-III FSIQ, VIQ, and PIQ
scores, respectively, for all ages. In Table B.1, for example, the General
Memory Index score predicted from an FSIQ score of 87 obtained by a 63-
year-old individual is 92. This score is subtracted from the General Memory
Index score obtained by the examinee. This difference is the discrepancy -
score. Tables B.4—B.6 provide the differences between predicted and ob-
tained Primary Index scores for the FSIQ, VIQ, and PIQ scores required for
significance at the .05 and .01 levels. The discrepancy score must be equal to
or greater than the listed value for the relevant age group to be statistically
significant. For the 63-year-old examinee who obtained an FSIQ score of 87,
according to Table B.4, a discrepancy score of 17 is significant at the .05 level
but not at the .01 level. The frequency of the discrepancy score is then deter-
mined. Tables B.7-B.9 provide the percentages of the standardization sample
who obtained various discrepancy scores. It is important to note that Tables
B.7-B.9 present percentages of the standardization sample who obtained
WMS-III Index scores lower than their predicted WMS-III Index scores.
Table B.7, for example, indicates that a difference between the predicted
(i.e., from the FSIQ) General Memory Index score and the actual General
Memory Index score of 8 points was obtained by 25% of the WAIS—III—
WMS-III standardization sample, whereas a difference of 27 points was
obtained by 1%-2% of the sample.

Differences Between the WAIS-III

and the WIAT-II
The WAIS-III provides a means of comparing an individual's general intellec-
tual ability level to his or her level of academic achievement. Comparisons
between intellectual ability and academic achievement have served as a pri-
mary criterion in determining the presence of specific learning disabilities
since the enactment of the Education for All Handicapped Children Act
of 1975.
Two methods for comparing intellectual ability and academic achievement
are presented: the simple-difference method and the predicted-difference
method (the WIAT-II Examiner's Manual provides the rationale for choosing
these methods and the statistical procedures involved.) The method for ana-
lyzing and interpreting ability-achievement discrepancies is very similar to
the one used for the VIQ-PIQ score discrepancies discussed earlier.

230
 Patterns and Profiles of Performance

Simple-Difference Method
With the simple-difference method, a WIAT-II standard score is subtracted
from the IQ score. This method is often selected because it is easy to explain
to parents and school board members.
Table C.7 reports the minimum differences between WAIS-III FSIQ, VIQ,
PIQ, VCI, and POI scores and WIAT-II subtest or composite standard scores
required for significance at the .05 and .01 levels. The differences between
WAIS-III IQ and Index scores and WIAT-II subtest and composite standard
scores obtained by various percentages of the linking sample are provided in
Tables C.8—C.12. The percentages represent the portions of the samples who
obtained WIAT-II scores lower than IQ scores.

Predicted-Difference Method
The predicted-achievement method takes into account the reliability of the
ability and achievement scales as well as the correlations between them.
Thus, the interpretations based on this method should be more accurate
than those based on the simple-difference method. Tables B.10—B.14 provide
predicted WIAT-II subtest and composite scores based on the WAIS-III FSIQ,
VIQ, PIQ, VCI, and POI scores. The discrepancies between the predicted and
observed achievement scores required for significance at .05 and .01 levels
are listed in Table B.15. Tables B.16—B.20 report the differences between pre-
dicted and observed achievement scores obtained by various percentages of
the linking sample. The percentages represent the portions of the sample
who obtained WIAT-II scores lower than IQ scores.

231
 q

= ke f *y ’ igiRi2

" i thi iW
at, hia

= a f wing ¢

* ray he
—~ : , Pry i fete |
mS =
d vy, } i
ey atl

> > i ri LA ~
= fh, |
sd Bilal

Se i on ‘p= we top
=a

oy tisil on ethpet wher’

. fhe Meee: oyecdee te m weed htm (dng =
oryait ie Rat) ‘ ~ Te envehdes bi gia |
ro > (reeeg ereign! lations Loptyye-stge pdb fagees |cect ele Gyealiyi ee auHl
seis pare biG 0 > dere Sona Tite fy 491 sinter eaogd: parts
0) 1 Ni le Dasistiom ae tains babtoewediyss aire
tre eory eng”. if) nerd du peraob ane[1°
neydeel Sei,
howe 19. ect Diee egal gis 406 eich ns dk vinarle
“0 omer? ep onatir ntl yhanreet sd a1rode) 240" wade bos
id
* ior dehigiers IN Re REO Seto 2 nerve ie deorreede
ng
Siyneee edt a agolrind add Insane: eal mene y ele
Ciittor cus * ater all 0-0 mide o
aug tre uw <a
The WAL») Pile hipogcone tte mania
: ia 7 os
Nm | i at Siew
'
 APPENDIX A

Intercorrelation Tables for the

WAIS-Ill and the WMS-III

233
 ajgel"y
“| SUOTefeII0I
Jo 1I9}UT
[]I-STVM3S91qNS payeos sa10dg
pure sung
Jo payeos sa109g
10J O] SafBog

234
pue :saxopuy]
Sasy 21-9 sIBaX
Aseyqns
[2|89S
xe@pu| A Ss Vv sd
Appendix A

| 4) N1 dd ad dd YN Vd Ss WO [eqeA “Hed IInd dOA_ Od WM Sd

A 28° 88 06
S) €9° Ye VeiesseLe
v SS Sv LL el c8
sd Le Se Se 6S" gy G8"
| el eS) Soe Le 8 8h 88>
0) cL LS Sy «OE 6G 08" vl
N1 ce es Oo US) ve Se" 98°
Od ey CavoRLe cv CE ce vL €9° 8&
do Sv Oo a ) Lg" 8e" ca CCm ve co 6S" Le
dd or" Zt Coe BE 8p seeLG Le" Hak: cL 98"
YN 8y" Sir: OS iL’ 6h OGY 6c 8e" oe 6G gL LL 18°
Vd Ly LE 6c" «9 66 8600” Ok 6e" &= ge ey 89° 6S"
Ss Sv’ COO 6h oo ly 6a= ve Sv Way Ae «BE 6c 06°
vO Ww OF= ve 9c 9e- SE «= 00 or 8=&8 65 6p
[Eqa/\ 6L V9) Vo Oe” cL OL LG or" 6= 6S 6S SV VS LV, cL’ G6) CO OMe OCLs
‘Hd cg CSI LG oya oO GG vy 9G; 68 b9' Og On) dC 6 20) CO 860g" 69
In jib Sy Sey BY! CL {SIR Hig vs" Ore So 79: 6h co oo OG. Vom OL «= 99°
OA VE gi OOo OE el L se Sv" oy VS hCG’Sy €9° ohh OS
Od 9s" cS”eV LS OO cy 0s" Or Oo Oe Va eGCo 1 67 IS
WM Ly’ Ss iy Wy Rie «6S Lo Le" 8psCOgE: bh bv’ kk or
Sd 0S" BE86 eS Ly’ vv WE 86 ey 79 ly OCHO19 OV
vUeey OO} +0 OOF OOF 66 “6i6 OL 6:6 00h 66 66 Zor 10l ODL 86S 66h 860l 662 862 LOE 0c
as 0'E OF Ge 6¢ O€ 6¢ 8c 6¢ moe S6c 86S “O08 Ge. 08 Sel SOL Geo 82. tL JES Wash
 235
Appendix A

=
ESS
FO ic 0 6c GE 6¢ oe OF Ve Fe O8 as
gs 08 J <8 Wye SSOLM Br fo ES 6c
cOl 101 66 66 OO! cor FOL OOL ZOL cOL FOL UeeW
ooz zoe zoe oe soll 10S 209 col FOL 66
Gy gotivet(i ti ittiESCi‘é‘i SCO Sd
se Ctidest(‘c:*«C«CCO CG ik WM
Ly 8s Lg (Ac 19 (Ac) 8s Od
9 4g ack gS 19° Cv Lv 9¢°
> 99°
ww ov es 6L gL 8" OL gL 18" OA
eS eS 8s Lg 8S
es 19° cick
co PoC etCtaSC<ia”SC*~<i DSCC SC COtCi ing
ig 8 8 16 Of GF 99 ‘ed
Lr SB eee a ee 6 oo +s #so if ec Gch ef ef ww ss
g9° 6V ov (ach gL gl es gl Bl €8° |eqis/
8s 08” ol cio 96° gL 4g Lg 8S g9°
vv €9° ce ev ee 6 ev 6° 6 LS” 6y vO
Lv 6e°
a «GG cS 2 SS Ss
16 9 4 a 6 & Vd
tone os or of ceo 2 eo
aro Be 6s 6Cee’(i‘éresés«O«CCS (tsi (iG WN
1g (iéia:CCO STG ag
cl oi ee i OelCOtCSCCC
6s ¥" o¢ ge Oh ce se er se oF ao
16 Of 9¢ ge Se ¥e" Od
ZL oc, 89 ly ce
1g o Se sy i mw @ NI
0g" 8 SS 2c 7 9
06 = 08" cl a: an ee) 8 l
99 19 6s 6s sa
68° 19-89 v
8° 29 eg"
06 = «18° 9g" el S
ze OtC 68" A
Sd WM Od OA ld ‘Hed eueA VO SS vd YUN Ga Gd od NI 9 Ft SG YW $S A _ XePUI
faleog
Asayqng
n SE
nn, Se
sIBIK GI-GI Sesy :soxopu] pue
saeag OJ 10J Sa109g payeIg Jo suINg puke S210Ig PafeIs 3SA}4NS []I-SIVM JO SUOeJetIOsIaU] “Sy age
 age] “E"y SUONLIa1I0
JO D1I9}UT
TII-STVM 1891QGNS P2TBIS sa109s
pue suns
Jo paeds $a109¢
10} OJ SaTBIS

236
pue :saxapuy
Sa8V FZ—-072 SIv93X
jseyqns
ja|eog
xapul a
Appendix A

oe = (ye <Sah Po 9 MN Md dd Ga HN Wd SS WO lealeA ‘Hed IInd OA Od WM Sd

A 68° gg" e6
S isi 9 BL” 06
Vv cc 1g 1g eg"
sd getité'
= 6g 69" 19" 9g"
l yy" eg 6g ee gg" «BL 06
D eee aeOG —— DC. OL) LAg GL
N1 ge ce 6¢ Of ¢ a 18
Od ee ee COGWy OGG 6S) or 6L:
(ale) eee te be Gyie 06
Ge. Sg. eo 06"
ag ChtCGH’C(“‘é‘«iGT:=C“‘
eek éSCO*COStCQH’CCGG br 9g"
HW meer ee 9G9 2a Wee ce fr 26 68 o7 68! 6f 18
Vd yee ceo) Be) Sy 62 Gh Sf ue oS ar oo
ss or ye ye br oF by BF if io ec is . ey : 06"
vO ereumenee:
ee Seu Gyn Ol On = ee:C«dG(
ON GS ge gg
Eee 1 ede Rie a as a a a a YR aa ce ae Se Ce
"Yad ee ee eG OSS LCC es Of yp ce: 9 Po) 10m be
iin ner ee7 an 0s LG GG fh OR OL BS 89 1g BS Come 2 Co
OA ee gcc gS’ iS eg ic) ieee
Od yh,ye Se CR Me i i a A gg’ gg
WM fone.
© ge OG Op ie 6c $F 20 19 Of 2 Ob 6S"

Sd
66
gp 86
OF 66
ey 66

ge 10

ec ZO
io LO
0S LO
ee 10

as Ois O08 be

ecuy OO!
OF
jo OO

0€ COLE, sanoOr

O00}
br OOF
be be O€
og OOF

te Ve
OG

main Lame)
c VC Scie
20S

66S
/02

CS
LEZ
O0€

LOL

ueeW
zeamuecr
gg
 237
Appendix A

ln
EEE
Po orG: Ole fe sce) bic Le Ve oe O€ OF ds
ERie w Ve ac whi WA Gis Wis Ve
-FOl FOL €0b 66 COL FOL Zot OO OO cot 66 OOF UeeW
doz zoe soe ooe Hit ZOS v09 Ol FOL
WR ae PR Ri 9€ Gr SV ey = oS yS ww lv Sd
We
dp RR Js AR GRE Ye Ash Key hsy LL WM
gs" eep eee
6G €9 °° 99° 66) 9G 20 2S: 9¢ co G9 ZG Od
og 868 69 29
8G 6G ES SEs Pe COOLS Oli LG Chine Ll OO) OA
6h 19° vv’ [S'_.
502 Zoey, Co) SEs as SZ td) syhe NSVE 8° Wing
LY §=88 ys «6B G97) (69) 5 699
9" 09 ae 00 same cir, 2S Vig erdch 0995) g9 8S—iD” ‘P98d
GG cl ep vas May co 86OL
8S Ce ee ee ae Gy 89s 6 19" Gl 82s 98) JEqaA
by 28: 89, OG 86: el 6p
6b 0 Sh iG sé 66 LOW | an 48 9% 8 6lftCéCET vO
69°
Ly 6S 0S) Oe 07 05) 2: yor es 60 dv $s
26"
oS 8 bE 6v ee IER vo = ov cs 6 8h COG Vd
6S el Hs
4: a 4 6y (‘i oS" LS" 09,” SES Nears: YIN
ys" ee 08"
ve" ag «=O 8 dy 0S) 6s bs qq
$8" el 6L
9c" Cy oe: SS elles eet Ach, OIE ao
co 9° €9°
ty 6 gesCéE Gy 95 Se Be. Od
18° co bl
sey IIS yo = BY €9" N1
corel
89" LS" 997 ene O)
8° 98"
8y 69 69 O08 |
6 = 08 98"
ss HAS sd
€8 OL el
9 86999 Vv
c8 ac ee"
LE S
68 = 6L Se"
16 A
vo -SB
VO SS Vd YN dd dd dd N1 0) | sd Vv Ss A xepu|
Sd WM Od OA IIM4 “Hed lege,
[a|e9S
ASeqns
n o——————
SIB9X 6Z-GZ S8BY :sexapuy] pue
“HY a1geL
saeag OI 10} $9109g payers Jo sUINS pUk S910Ig Payedsg 1S91qGNS TII-SIVM JO SUOHefe1tOd19,UT
 age] “Gy SUONL]ALIO
JO IIAIU]
ITI-STVM 3s9}qN¢ payeds sa10dg
pure suing
Jo payeog sa109g
10} O] Sa[eog

238
pue :saxapuy]
Savy PE-OE SIBaX
Aseqns
[29S
Xepu| A s Vv sd | 0)
Appendix A

N1 dd ao Ga YN Wd SS VO 18424 “Hed IInd OA Od WM Sd

A 68" G8 G6.
Ss) jt, 98° 18 Le
Vv Soacgs Le Le" $8",
Sd cmosCo 19° 6S" $8"
| (5VLi 75 Ve or 88° ES 6
0) Ls SEL LE Ye 7 OL 18 18"
N1 8h 09 19° ov Sv 98°
Od 66 te Cyc Bye LY GL 29: c8
do OSs eee ey OF= ve Oey CC 6S" cg 88"
dd es OS= 6S OG) ec ES: 1/27. oo) ve 08" bl v8"
YW os 2 89 Zr Go 87 ov’ «OE = oS gL bl 08"
Vd ComuneCV 8d IQ eo Cy Vi EB 6L SL
SS ev’ BE = oy cy OLE 6 ly BECO
| cb ¢ LB
vO 8h cy CyB ie ie BS VG Col 2 ec)
0): 7A
Jeqle/ Ge Ae BE ey eeeCGY oo RR Msp 08" eG SO C2 AAST ase
‘H8d 0L ER: WWE Coos89 69 6S OOD, ee 9) en 9 CO LO) GeolTOs COOLS
Ing 18° By We 0g Be os Soi U9 or, Ome DO OI 9S) bo] 06 48 ey ee
OA v8" 4908Ly’ 18" cD 98: 9S ZG) 89 ee 7 oy /RY 69) OF
Od I) (el i Gy ‘oi rash ee 6S" Oo mG GeeneC 0) VG 220 OSes
WM 19° ey wey 20) BS) USAyy Ly 86D 9 cs
Sd Sv’ 866 IS Lv’ ey’ WAL SWE 9¢" 6h GE OF tS ey
Ueey -O} LOL @6 LOL OO ZO 86 86 OO ZO OOF OOF ZO 10! €09 10S EOL COE OOF O0€ c0c
ds be O€ ke 6¢ he se Gin Orgs or far ‘Be OP Ge Gh 6yi “SOL ‘Sve $2 o. 08 cS
 239
Appendix A

a6 cs 8¢) O& oe O8 sme AUS MORE as

+8 ty 8. Pee “Stl “67. oc ue Fo) “Oe Gia
cG OOF FOL 86 OOF 66 uBe\\|
LOS 26S OO FOr Ol LOL Ol LOL FOL OO OOF
z0z lez soe g6c voll
Lv ev ev Ly 9° 8y ov 1g” Sd
Lv 69° ce 8v es 69°
ve 69° ov 79 09° 99° 4g ci WM
92° ge" es €9° or 0g
LS 9S” 19° Od
vs 99° 89° lv 8S" 8 6S vs Le 19°
8S 4S 89° 4g
ov 87 rach 08” BL vv g9° (Wh: €8 OA
9° 9 6v 09° 09° 0s
0S 99° H4
9 67 4g 89° gL gL 67 ol pl 18"
6L ice 06° 8S 09° 79 CE
99° 19° 19 “Hd
vs 19° 69° Lv 19° vs £9 6S" ov 9
GL, 6S” v6" 69° Le" 99° Ach
vy 6r OL gL 6L 6 69° gL €8" [eqs
49 S6 S6 pl OV (Ace 19° 99° 9S"
ys Se A(0)
8" gy 99° 6¢° \¢° ce ey 8" vl 6° ev vW
vy
6S cy ov ov ov ic LV ey 67 Ss
te 6y 0s”
06° es gs Vd
0s’ Sv (Ag ev ace es ZS €e lv
LL el
LS es ey gS es 6° Ach cis 9S” HW
ice gl 08" 9
vv ys vv 0S’ cv 60° GS Lv 0S (ajs|
$8 oL 18"
er OWE OR OR Bg cee eae ao
68 gs" 6° Go ge ww er Od
62° vv
£8" 99° ]#8
oe g@ ss oF eH NI
sg Q- Se ar ee 9
18° 98 69° ill |
ov’ €9
98"
06" 08°
€9' zed sa
28 gc ss 09) v
1g BL 0g’ SL S
06 6’ y8 A
6 S@ 68"
Od OA ind ‘Hed Wah vO Ss Vd YW de co Sod NT Sy J) Si vos css eeee ae
Sd WM jaleag
aseyans
O o———$——$——
SIRAK HH-GE SIBy :soxopu]l pue
IIIJUT «“*g"Yy BIGeL
saeag OJ 10J $910Ig payeIsg JO SUING PUL S9109g P2BIS 3S93qNS ITI-SIVM JO SUONL]A1IO
 ajqelLy SUONEIa1I
JO 0010qN]
[]]-STVM 1S91qQN¢S payers sa10d¢g
puke sung
JO pafedg Sa109g
10J OJ Sa[BIS

240
pue :sexepuy]
Sasy P-GP SIvIX
asayang
/aleog
Senn Ae) “S= ve sa’/ rf 9
Appendix A

NM m4 ad Ga YW Wd SS WO eA ‘Hed IInd OA Od WM Sd
A 8g" 26"
S QL 8" «(2g 06
Vv cee 08" gL’ 62’
sd er or cy zo gg’ zg"
l oF 69 oy ie se" 682g"
s) ee?) ely 1° geet Yeo) 9g" 1g"
NI ee er UC csti‘icw|!;COCe se’
Od ec eB Of $5 IS o ae aye ze"
ao oe ep ee be by 6k Sh ay IG 26°
ag ee: eee Gee Se =7y Of iy eg ey 18"
YIN +9 9 og se 2 08 ec 8S 6h 69 eee" 18°
Wd Te Ge ae eG SSeS Ole 6 2 eg Ce
Ss Ce OG eg OS ey 26°
vO 6SisCié‘«iG:C“‘<‘é‘
eg Ic HSCOC*«éH*C<‘éiSC*«C:C*é‘éHSOC*«#G’*C;«*«GSG
ee oy reg a esoiy ee ca Ce er 8 oo
‘Yad OS) Vy eh ie Le PCCM a a a: eg GE 06 99° 1
ling ee re Gr eg eG Or ZF 99 eo o9 Gare: Oe
OA ae uk fo eae eee a a: a a 49os
Od Ge Ga CG's” eo 29 so
WM esKE Gio SCR | a | 8g"
Sd te I GC IS a wr ce a a i
uee; 66 Lo OOL OO ODL 66 E01 +O LOL OOL LOL 101 LO! OOF O09 v0S POLL FOE €0€ 80 H0e
me eolNOcuD
Ge De OF Be Oe 6c We OF Oe 6c OF SH Bll OG 28 92 vl ye
 241
Appendix A

ene
O€ OF be Le re O€ fe ce ve oe OF€ as
rs Ci 92 ¥8 CSc “Gi™US. Ge Oe ke
66 FOL 66 LOL ZO FOL OOF OOF OOF OO 66 ued
66L o0€ oOo0€ 662 OO OOS O09 OOF FOL FOL
Is 69 or 9¢ os 69 ts 9S co er 9 GF 69 Sd
8v Ly 69° 79 99° vs gS €9° oL WM
Sion gS’ lv 8S" 0S’
Lg
oo + wm iz so os iw oS +0 So gS €) )yo sees Od
zy gg" 08° 18” vs 89° 08" 8 OA
(An 19° 9° 69° 1S 0s 9S" (Ach
09° el 69°
vg 9° OL 6L 08° 8s 17 62 cs (IN
cs 88° c6 9° el 89° LE rach
el “Ped
pl €9° ace v9 09° 19° 69° LS” 79 69° OL
BL cg c6 gL c6 OL pl 79
GS \¢° 8S OL 08° 18 6S el 6L 8 [eqis/
eg 98° el co 96° Hoke es gg 69° Ab
Lg’ 19° Lv 6S 6c Ace vv Le ov es ov vO
an es
es oo cs 6 + of wo ws Sy ug so OS Ss
26 tél CCC Vd
La, 29 «Oghtt(‘ie:SOCSt(‘iES:CG:s(‘i‘i:SCtiCiESC(
ze téW" o¢ gr us ew wooo ow ¢e + OF YW
98 ag
78 re ee Oi«zG:—“‘éthé®étOUOeCOhCUCSHCUt’'
zo sé le IS I MP eS ee ao
Z6 Of Ob Ge eee dd
eg ie Be ly \ sc)
ig 6 $6 Se Gee NI
8g" Bf — XG Gd Oe ar 9
eg" 1g |
z6 OC" lg oe o& ie
99 ZL is os = et sa
QL 9 «19 v
1g 6’ 28° S
16 68" 9g" LL
e6 SB 68 A
OA ld ‘Hed e@eA WO ss Wd YN Ga Go od NT 9 t Sd YW $S A xOPUI
Sd WM Od jajeos
aseyans
Sy Se
a. SIBIK F9-GG sesy :asdy Ag Sexepuy
J1GeL
pure sayeog OI 10J Sa109g payeds JO SUING pu sat0dg PaTeIg }S9IqGNS Jo SUOPP]a110910}] "9'y
 A age] ‘ov UT SUOTP[91109.
jo 10}
[TI-STVM
1S91qNs Paes saz109S
pure suns
JO Pafeas S9IOIS
1I0J OI SIfBIS

42
pue :saxapuy
Say 69-G9 SIB9X
eee
Aseiqns
EEE
[9/29
xepu| A Ss
Appendix A

Vv sd I @) NI Idd dd ds YN Vd Ss WO [eqeA “Hed IInd OA Od WM Sd

A Le" 88= v6
iS} 18" 68° C6218
V 69) EZ 98° es" 18
sd LS" BF = 6S 89° eg eo"
| ce ae OF os” 68" cB
6) 18° Lill” lis Ly BL" 18° G8 cL
N1 OE)Sie 8609 6S LS" 98°
Od o9" 9” = eg is Ci 99 cy 18" Leds $8"
do HG, EE ee ine 1667 ~ LY 87 8S cS yes kL’ €6"
dg SIS) 48S) 6S 8E Ly 6S 86 ty ey vs €8 OL: 28°
YW OE) vay eh BY > oS 19° GY 8G LS” G9 8 6L 98"
Vd MGS MSE sep CE 6b gS" fs" JAS OS iB oc 08° cL Sky
Ss 8g <c9 eo hig Gy 86g 6h 9 umes oc CSCS ee"
YO 87 oS ww 8 OF ae Mag hee ag 89" GG GG 6G -
[Eq 98" ees! SZ OS ys 18" OL is) C6 oo 99GB cs BL RS AS HWE eh Hey
‘8d cl ae OL BY 86 e9 aan EO 69° ele) a AW YR yh (a 0c) «99 €8
In4 8" 8 6L SS” 8 18" LL’ LL’ Oe)2 ee S/O)2 ce 8 06 8 BL
OA Coes: SL EG co eaS8' 829 SO GieeneSi TB 90) ZS 369) OG el Veils
Od 69° EL = 89 Higa 4 $9 09 IZ aie Ve WWE Lg: OZ
WM bL 69= 8S cdi 89 19° eee) Cy LG’ BhCtséBY’
68S le 09"
Sd 69° Coro!OV eS 9G = 9G 09° EZ FQ 69 eo €2 09
uUeaY OOF OOF 66 66 OOF OOF OOF FOL 66 66 OO OO OO OO GES 66h 860! ODE OOF FOE 66l
as be see Sir. OC Ge ig Ee: be € O88 ke aie Se FOE oe a Ose F5 G2 92, eo
 243
Appendix A

a
EEE —E——E—EOEEE—E—E—_—E e
"OS Ue C6e we ee 6c ve tet 6¢ 6c 66 66 6¢ as
GG “SZ VL 62 EFC [Link] “OS
OO OOF OOF FOF cOl FOL 66 66 OOol FOL +O cOk cOl UeeW
eoz 662 10e E0e QOL COS £09 OOF
69° Lv’ eg Lg 6S" 9g" cies 9° vs ve (Ace ys vs Sd
\g°
9¢° 99° vw 8c Lv 19° gS ys lv (Ach ly 8S WM
LS Le es
09° €9° 69° es 19° 19° y9 v9 ce 09° L9° v9 Od
99° 9° 19° g9°
Ac) 19° 09° 87 09° \¢° 18" gL Ov g9 gl v8" OA
09° Lg OL: es 6S
v9 OL 69° vs G9 v9 gL BL wv OL EES 18” a
GL gL 88° co v9 oe
8S g9° pee \S° 99° 19° 89° 19° 9¢° rac) kL 89° “8d
6L 9° c6" gL 6° kL GE
€9° rach 99° 09° 0S 8S” 19° 6L 6L cy 69° bl €8 Jeqs/\
€9° LES el S6 co LE An
87" 09° 0s v9 ev 9S” (Ao 6v" ly €o 9¢° 1° Lv vO
eg es 9g" 6S gS’ 9S 8S” gs ve 6y ys es Ss
88°
0s’ eg Az Qs 9¢° gs’ cis be" Cv 8S’ 9c Vd
be pl
19° ov’ es 09° 9S" 8s se 09° es 9° YW
3 gL 6L
Ly 69° or es (An vo LS’ 8° es dd
18° gL £8"
Sy ov’ ev Iv Lo ev ev cv ao
06° €9° OL
vy vs (Ac 02° Lv 6S es Od
8 ae 6L
or 0S’ ov 19° 9¢" (An N1
sg’
27ig 8c" LS £2 pl is)
£9" 98°
Le g9 89° gL |
68 sé 98"
87 62° vv sd
gL es 19°
28" gL’ 6L 1s 09" v
06 «ze eg" 81: S
e6 SB" 68" A
VO ss Vd UW ag aod Od N1 4) | sa Vv S A Xepul
WM Od DA IIN4 = “Hed =[eq18A
Sd
faleog
asaians
ee
SIBIK FL-OZL S88VY :saxopuy] pue
Se
saTeag OI OJ Sa109¢ pafeIsg JO SUING PUP Sa10Ig PafeIg 1S91qGNS []I-SIVM JO SUOHe[er10919}UT “OLY 31qel
 ajgey
Ly“| SUONLJaIIOII9}
JO UT
TII-SIVM
3S91QNS payeds sa109g
pue sung
Jo payeds Sa109g
10J OJ Sa[BIS 2

244
pue :saxapuy
Sasy 62-GZ SIB8X 5
asaiqns =
[aje9Ss
x@pul <x
Suen va [Sats 9 NI Md ad G8 YN Wd SS VO Ie “Hed IInd OA Od WM Sd
A 88" 18 C6
S pit 88" Ze"
Vv «6 9g bl 89° 18
Sa GG" 2g os 02" 69" LL
| ee SG" Ly 88" CE 26
3) ges—-g-——- 07 oi'—- 18" LL
N1 Sa 7 a a: i BL"
Od eames
5 egg te ce 2 ce GL" 9 Ds
ao eres Gr aa 1p 88 EE 8S Ch 60 oo 68"
aa aU ae es a”| | An 4 oF Bo 18"
HN Ce,Ge Fe ee fe? ce Os OL oo Le
Vd eee fe CO Och ee cL 107
Ss Gee iC CC CHC de sel cy 26
vO rue nau a Mens ne: oe an a 3
re PO |: oe) a i) a jr 6G 28 ym
"Hed cx pis Tip CE Rem CMC a Sn n° a)
a: 7
In4 as rant Sis Ome TMNT air a nn:a, nn nn i op : Ca
OA rc <: | << a AO”
| 4 A| | <A A 6S ez os
Od (iéG’:*Ci‘iG:CO;é:‘éeS’:=«é«
is’ ghsi(‘itiaCtC«‘aKQSCtié«OSGk
CiLTCGH’Ci( eS H’'
S66 6s 2g LS
WM ee eo Bae OR Be a gs’ Og 6st) Sf 09"
Sd rl ese Apt nee) A | a a) a
ueey COL +O EOL OOF OL COL OOF OO 66 ZOl ZO 66 OO! OO 609 +0S Ol POE €0€ cOE 661
ds Ve 0S Bic 8c 6'¢_ 98 66 0€ wha hae Mia AOE SO Oe ak AU, UIE GR OW dA Gi
—
 BIGeL “ZI"y SUOTepa1I0I10}UT
JO []I-SIVM }sa}qn¢ payeds sa10Ig
puke suINg
Jo payeog sa10Ig
10J OI SafeIg
pue :soxapuy
Sosy PZ-08 SIBIA

/aleog
xepur __A § WV SO | 9 NTI %d ad GS YN Wd SS WO IeeA ‘Hed iliMd OA Od WM Sd

asaiqns
A 06 Sg’ v6
S 0g 9g" es" 16
v 4g" 19 18° QL 18°
sa a 09° es" 1g"
l ea Fe 6 8" LL 06
9 —eF —~-c9' 0g’ Se 69° eg 6L
NI ereGe JG! ge OG 9g"
Od S| (a a 7 an 2 BL’ g9 18°
ao ee a ce Te 98> SB Sg’. Q 92 ig 16
ag ee a 2 ae oe ne a BL = 99 98
YIN tee Ge SS Gr OOS Se oo Fes r pho y8
Vd ee en a: a a a: oye gi)
Ss OY > a en; es a a: nn Ce 16
vO Rr an a a a a a a a
[eq G8 62° Ua ov BE gl g9 1S” 1g” 8 09° che es 6c 69° v6" c6 9° $8" ZS
‘ed ky eSee en: nn a) a a
\IN4 18" BL 69° vv (ihe ez 69° 19° 69° 8S 89° 99° g9° ov 68° v8" v8" CE
OA 8 08° 49° lv 8 gL gS 0S 87 Lv 4g GS ov ke rach 89° LS
Od
19"
so
ee
fee
gn
CCGG
OSC’
ao)
we
cy
WM v9 99° 19° €9° 8S" 19° 79 19° Lg 8° 9S" 6v g9° Lo 19°
Sd Lg” Lg” 67 or 6¢° Ov g9° gS 99° 9S 1g” Lg” 99° ve

uesy
860!
S09
Eby
96c
s0€
661
10E
FOkL
cok
LOL
tOl
LTO
OO
£986
+0
86
66
ds Ope, Ocoee © Ca 88 8c Cama: Tr CCl manic COC Oe SC (OF OF) yre GH S€ O08 FL 62 ¥S

245
Appendix A
 age “Ey SUONL[eL10010}UT
JO [IJ-SIVMsa1qng payeos sa10dg
pue surng
Jo payers Sa109g
10J OJ Sa[BIS

246
pue :saxapu]
Sosy 68-G8 SIVAK
Assyqns
[P89S
Appendix A

xepul A s Vv sd I 0) N1 dd dad dae YN Vd SS WO leqeA ‘8d IInd OA_ Od WM Sd

A 98" 02 06
5 OL 18" 62 88
V or 6E° 69° (BS) 18°
sd Se Le ov 09° eS co
| (OE: Sh vA ee €8" aR: LO
©) LL EG 6& «6S oo Le vl
N1 8c soe6” Le ka 8"
Od 6YtiéiH”
8 nor OS 8= eo Oz 89° BL
el) ey ova Cyan iti 6h Le" eee oi gl 69° c6
dd 8" oS ao of8 Ww OW co Sv’ 8y BL cL es
YIN or She Se CES cy 8G Vic LG: Lv cs OL vo BL
Vd ee: eS GV 86ce8 eo 8= le Sip COO «66h 66 gL bl
ss ov oS ey <a rn| ly Se LL LG ly vs €6"
YO 6E" 6 8k 80° ~— Ze vv 60'0- 8e eC nn.) Oy ee Cec «
[Eqa/ JE NE SI vv. VL EZ eS GG 8S kG: 7 eS ey bl OMNI!
ee CO we sy
‘H98d 6S" SIS) sp LE’ 89s se ey 09° @9 = £9 yt amenCo cL LV cb: CL VOs 8S"= 08
In el Sac 6S EY ZL (eh) ap 09° 19° AS ie CMG!cls CL
OA c/aCo Le: elo sez) Fo. eh es oo POO, Vamsi (ssh as) HSS
Od oy 8h Le" 09° mneyOC a 2 AS a res SS et go
WM vv ISK MS (655) He OS go Le’ 6G cy OU vk Sig)
Sd or" 6b «GE eG WE 9p es’ TE My ry OS be 6"
UPS OO} OOF FOL -OL OOF +-FOL +OL Ook 66 3:6 LOL 66 OO OO c09 L6rp 660! OOF 66c cOE 8S6l
ds OS OtG es eZ 18a BiG. 8¢ O€ O€ Camm CameO Cee,
1 Ve bs fail AIIM Ciera WANE RE db eM
 247
Appendix A

62 oe oe oe oe Le 82 oe Ls Le oe as
O'OL 66 66 0'0L O'OL e'0l ool 0'0l (oye) O'OL 66 uee|\
02’ v2 LO’. zg Ze" Ce ZV 9h 90° 6S ~—-eq UONIUBooey Aoypny
L0° 02 ee bh Sb". v6" ZV iz 82° }299Y || Id We4
Ol ZV €0° Le" 90° 69° €0°- 80° E984 I WA
gl’. 90° Lp n 02 Ets OL Le uowuBbooey || se0e4
ec ce ee" 3 20'- v8 He98Y | WI
Ov’ 60° 70° v0'- 92 ueds jeneds
Ql. Go [Re ve Hulsuanbas N-7

[12984
wey
|Wd
Go
8
Lh
she St IT298Y |
VdA
Q|°. uoiubo08Y | see

12994
W1
|
Ee jyeoay yje0y Hovey jye994 ueds fulsuenbes jjeoey jye904 Hooey jyeoay 3s9}qNS
sey PNY || Wd wes || VdA || S084 | WI jeneds N-1 | Wd wey | WdA | sa0e4 1
sIBaX 21-9] SeBy :sa109¢ payeas 1saIqQns AreUILI [II-SINM JO SUONR[A1I0II9jUT “HL"y Bigel
 o1gel “GL'W
UT SUOT}ejf91109.19}
JO III-SINMAIeUulLIgjS9}qns PI[BIS *S8IOIS
sasy 6I1-SI SIBIX

248
|W seoe4
| WdA
| wey id| N-1 jeneds aWWI] S99e4
|| eee ee el SS
WdA
|| wey Id|] PNY 98H
ysejqns yyeo98y Hovey jyeoay jjeday fulouenbas
ueds yjeo0y foosy yjeo8y yyeoay jeG

|
Appendix A

NT
seoe4
| uouBboosy gL

W984
VdA
| 2984 89° JS
wey Wd| JE98Y 0¢ iG Ql
N-7 Bulsuenbes vy’ 60° Z¢: Le
eneds ueds oe Qh 8" 92° 8g"
WI|] 2998Y v8 6k OL 8c ov 9c
se0e4
|| uoqluBodeY BL GL Ge" 9c" 6k vi 6"
VdII 98H 8y" 0€" gL 80° 9c" ke Ly Le’
Wey Wd|] JE98Y Ms 80° vo 68° ev ve ee 0¢ Sl
Auoypny uonuBooeYy
[eq gg" Gl vw 8o 0g" GG 09° 6h JAS: 8e"
ueal/| LOL 00 0'01 001 001 L'6 oxo) 001 001 66 LOL
as o€ ee Le Le Ge Ng o€ Oe 62 8'2 82
 249
Appendix A

oe Ze 67 Le Le 62 82 Le oe oe Le as
O'0L 0°01 0°01 0°01 LOL 0°01 0°01 66 0°01 0'0L O'0L ueey\
82" 12” er’ 99’ €0'- ee" 82 9b" 82" 99° —-Jaq UONIUBooeY Aioylpny
Or’ Ly’ 6h" 2 St’ 96" ee ly 1S JEOOY |] Wg we4
82 er’ ee" le 82 el ve se’ 1299 |] VWd/A
Or’ €0' 12” ge" 8h" 69° ee uojuBooeY || seoe4
vl Se’ Sy’ 9¢° ge’ 8" 2984 | NT
0¢ aly ea y0'- ol ueds jeneds
Au ze" 20° 62 Hulouenbeg N-7

|VdA!E98Y
we4
92°
|Iq
Ly’
ey
12994
ee
Aa
G77 uoiubooey | seoe4

e994
|WI
\2q yjedey {]2094 Hovey je998y ueds ulsuenbesg |jed0y 1/2904 Hooay jje08y ysajqns
day PNY |/qwejy = || WdA || se9e4 I7 jeneds N-1 | Wd wey | WdA | se0e4 | 1N7
SIB9IA VC—-OC sa3y :S91OI§ pajeIs jsajIqns Arewlid III-SINM JO SUOTILfaIIOIIIIUT “OLY 3iqgel
 ajgel
|"y“7 SUOT}e
J9 fa110D19}UT
ITII-SINM Areulligjsa}qns Poses *S9IOIsosy
S 67Z—-SZ SICIK

250
11 se0e4
| WdA
| wey Jd| N-7 jeneds IW Se0e4
|| WdA
Il wey Iq|| pny dey
ysajqns “yWeo04 Booay y1e094 “yye084 Bulouanbas
ueds jje9904| Booey yyeoay yyeo9a4 Eva]
WI| WO8Y —™=
Appendix A

seor4
| uonjuBoosy +0" ee
WdA
| E98Y 99° €0'-
we4 Iq| JE99Y se’ eu Gh
N-7 Bulouenbes Ly Ale ly OL
jelyedsueds ph vl 02° 92° 1S
WI|] e99y sg” Ae 99° og" 9¢" vie
se0e4
|| uowuBooeY OL BL’ 80° Qh le 6k Ge’
WdA
|| E98Y lg 80° 68° ea 8c’ sl 8S" Hise
wey Ad|] E99Y 1 60° Qh" 26° el ee og" vl 8z
AuoyIpny UORIUBooeY
Jaq 92° 80° 19° ce’ a 02 el’ He or 9¢°
ues 8°6 66 O'OL ek LOL €'0l 8°6 O'OL LOL LOL 8°6
as 6 62 oe PALS ae oe re Re Le Le Ze
Appendix A

251
82 Ze Le oe oe Le Ze oe oe Le oe as
yO 8°6 LOL LOL Z'0l O'OL 8'6 66 66 Z0l €0l urea
ve" ly’ Lb el’ v2" Le" ee" er’ gb 69° —-[8 UONIUBooey Aioypny
ev 8h Ly 8c Le v6 ve ve Ly []E99Y || Wd wey
v2 gs" 9¢° vs" vy €8° 6h vs" R98Y I VWdA
60° ray i 6h x Wa be uoniuBooay || seoe4
62° Le 6e° 9s" 40° 68° e994 {1 1
os" 62° se" el 82° jenjedsueds
Le" os co ev Hulsuanbas N-7
9g" Gc 8e" [E98 Wey Id|
ca GS” 12984 |
VWdA
10° uoubooayY | seoe4
- WEO8Y | NT
jag jje08Y [ye00Y Hovey jjeoey ueds ujouenbes |jedey je084 Booey \je09 ysayqns
dey PNY || Ad wey = || VdA || Sa0e4 | WI jeneds N-1 | Wd wey | WdA | Se0e4 |W
SIBIK PE-OE S8BVy :$9109¢ payeos 3saiqns AreulAd [TI-SINM JO SUOHRIaLIOII9jUT “QL"y BIGeL
 o1geL “6L'V
UT SUOTPja1109
JO III-SINM
19} Aeullig jJSsoIqns Pyleas *S9IOIS
sassy PH-GSE SIBIA

292
N17
| s90e4
| WdA
| wey 9d| N-1 leneds IN se9e4
|| WdA
|| = wey Wd|| PNY dey
}s9}qnS jje090y Booey jyeo0y jjeoay fulouanbesg
ueds jje00y Hooay yyeoey yyeo0y Eyal
WI| E99Y
Appendix A

see
| uoniuboo8y 80°
WdA
| IE98Y ge" 62°
we4 Iq| JEDBY 92° see Or
N-7 Bulouenbes Oy” OL’ vr’ ay
jeyedsueds Ge gL ec 6e" Or’
WI|1 E99Y 88° SI or se’ cs’ oto
se0e4
|| uoubooeY 02° GL Ge’ ve a QI Lo’
VdA
|| 98H ve" 92° 83° ly eS" rAse ov ve
wey Iq|| [E99Y 0g" Go’ Ly v6" oy Ly’ ey Lz or"
Auoypny uoNuBodey
f8q_ —s ZZ OL eg’ 9¢° 0s’ ey OE v2 rAch oy
ues v'0l O'Ol LOL LOL oKoy O'OL v'Ol LOL v'Ol el LOL
as 82 ne Le 62 82 62 82 oe Oe 62 Ze
 253
Appendix A

Le 62 ral oe Le le as

0°01Oe
Ze oe Oe
LOL 66 O'OL 0'OL (one) €'0l 001 O'Ol O01 ure
Se ee" el ZL’ jeq uonjuBooey Auoypny

Ly 9e"
le Ly ot" re Le"
Ze" Le" vs" 2 ee" 06" ee vs |JEOOY |] Wg we4
82 ve" be 7g 298 II VWdA

Bl 88"
QL es’ Ls
ve Qt 62° se OL 62° uonluBbooeY || se0e4
0¢" by Le" 68° 298 |] W

8c ee ss’
se
2 8h gt 9¢° ueds jeneds
se" oo ee Hurlouanbes N-1
0g Ly FED8Y | Aq wey

6°
6h (Ac 1129984 | WdA
0c uonubooey | se0r4
e994 |W

|WdA112984
yye904 jye904 Booey [ye994 ueds Bujouenbes = jjedey Booey = 12984 1s2}qNS
pny19d
| Wd wey = || VdA || se0e4 1 jeneds N-1 | Wd wey | Sa0b4 | WT
99
a ———————————————
SIvok FS-SP sa3sy *S81OI§ PafBIsg }S9}qnsg Areullig III-SINMJ° SUOTLIA1IOII9}UT “OZYY 3/qeL
 3]ge] Z"y
“1 suoneyj
JO aiio0s1
III-SI
9}Uy WM AreulLig ysaiqns payeos :sa1oag
Say P9-GG SIB8X

254
WI| seoe4
| WdA
| wey Id| N-1 jeyeds |W Se0e4
|| WdA
|| = wey Iq|| PNY 99y
ysaj}qns jyeoay Booey yyeoay jjesay fulouanbas
ueds jyeoay Hooay TRESS] yyeoay 19q
WT| EO8Y .
Appendix A

seory
| uoniuBbooey OL’
VdA
| E984 Cv ve
wey Iq| E994 0g Ov ww
N-7 Bulsuanbes ey’ €@ og 92
jenedsueds ov’ 92" a 92° 8r
1 |] "2984 €8" LG eg" Sv vy ge"
seoe4
|| uowubooeY 90° 9g" Le 8" Wa €e HG;
VdA
Il 2984 0g" oe €8° 6o° 8d" oe ev ve
wey Iq|| I]B99Y 6c en ev Le" Zé 0c Cv LG" ov
Auoypny uoMUBoo8Yy
l9q oL eb vs oe" ev 9g 89° Co ev 6c
ueey\| 00 001 001 00 0kZ 001 0'0L 001 66 0'0L 001
gs 6¢ Le ie Le tS 62 las 62 ove oe Le
ee
 259
Appendix A

as
oe O€ 8c

Le
Oe Le O€ 6c Le Le be
0°01 LOL 0°01 LOL 66 LOL 0'0l O'0L 0°01 O'0l 66 ues
es" 0s” 12 29" ee" sy’ Sy QP" 02’ 89° Jaq uoH|UBooey Auoypny
es" €2" 0s’ 9¢" Or’ 26" lg 1 6S° JE9OY |] Wq we4
62° 6h’ 92 6e" 0s" 1g" St eg" E98Y I] Vd
Le" el ZL €a 87 69° 6c uoniuBooey || seoe4
62" le Lg 6h’ So’ gg’ WeO8y |] WI
er’ Le" 92° LO’- ee ueds eneds
le Or’ 20'- 1S" Bulouenbeg N-1
lg’ 02’ 29° JE998Y | Dd We4
iA 19" E98Y |WdA
vd uoiniuboo8Yy | Seoe4
me 98 | WI
190 TBE): | jye004 Hovey jye00y ueds Bujouenbesg |jed0y jye09y Booey 1ye994 ysayqns
day PNY || Iq wey || WdA || se0e4 1) W7 jeyeds N-1 | Nd wey | WdA | se0e4 | NT
eS
sreaxk 69-S9 sasy *S81IOIS PafeIs SVIqNsS Arewid III-SINM JO SUOTILfsAIOIIOJU] "ZZ BjGeh
 3/qe “EZ"Y suoneja
JO ti0siaj
III-SIW
uy M AreUlLIg }saiqns payers $9109SeBV YL-O2 SivaX

256
N71
| seoe4
| WdA
=| wey [Sq N-1 jeneds HW seoe4
|| WdA
|| = wey Nd|| PNY 98H
yseyqns [je094 Booey jyeoey jjeoey 6ulouanbas
ueds yyeoay Booey jyeoay yyeo0y 19G
WI| E99Y .
Appendix A

ssoe4
| uoniuBbooey c0°
VdA
| 298Y ly Z0°
wey Iq| FED@Y 6g" 92° 6g"
\-7 Bulousnbes vs" Lz’ le ge’
jenedsueds 82° Go’ gL 9¢" vs’
7I E99Y 19° 0° Lr’ es’ 1S’ Ge
seoey
|| uomuBbooey 0° 6S" LO’ ale €z aS 80’
VdA
Il E98Y ly Ol 1g" ge" 6° ve ey Ol
wey Iq|| TE99Y Ly v2 Ly’ 68° 6¢° se" LS 6h ev’
ANoyIpny |8q_UONJUBOD
—- 89° eY SO" OV’ 82’ vy 62° 7S" ol eS’ Le"
ueey\ oye) O'OL oxo) 0'Ol 66 L6 LOL ove) O'OL LOL 66
gs Ove Ze oe Le ve Le 62% 62 Le Le 62
 257
Appendix A

Le 6c 6c Le Le 62 oe as
Le Oe Oe Oe
0°01 c Ol 001 0°01 0'0l O'0l 0°01 ueey
0°01 00 0'0l oe
Se" 62 bh Lv ov" €9° ee" 69° jeq uontuBooey Auoypny
ey’ eS"
GG" le ov’ 68° 8" Se 60° [E09 || Wd wey
Le 8e"
60° LS’ 02° Se" 9¢° 98" ve" Ov" EOOY || VA
Sv 02° 60° Sv’ Le" 09° 92 uoniubooey || se0e4
Lo 0s" 19° 99) 9€" 28" EO9Y || WI
Le 6h oo S's 02’ ueds jeneds
6e° Se" ee se Hulouenbes N-1
ov Lv 87 |[EOOY | 4 wey
Go Lg’ WE08Y | VdA
1Z uonjiuBooay | se0e4
|WW
2984
——_——_
eee
Ee jje094 \je004 Booey jje004 ueds Burlsuanbas = jjeoay yjeooy Booey ~—Ss 2084 \sayqns
|| Ad WeF Il WdA || $9084 I] NT jeneds N-1 | Wd wey | WdA | |W
904 PNY EE aa aa
"72" 3/GeL
ET
SIBAK G-GL SAB :S8109g payed 3saIqns AreUlLIg [TI-SINM JO SUONRJat10919}UT
a arate ESSSTISSE ISSN SSTRS SSS EEE ENE SSIES
Neen ae errr ee eee eee eee a ee ee ec
 Bfgey “GZ"y SUOT}e[e1
JO I0I19}U]
[TI-SINM ATeUILIg sajyqng payeos :S910IgSey F8-08 SIVA

258
|W seoe4
| WdA
| wey Wd
| N-1 jeneds IANIL S904
|| WdA
Il wey Ii|| PNY 99y
ysayqns yye99y Booey jje90y jjeoay fulouanbas
ueds jje090y Booey |je9ay yyeoay Eyal
WI| e984 bs [Link]
Appendix A

seer
| uoljluBbooay be
WdA
| E98 es’ 43
wey Iq| /E99Y ly Ac 8
N-7 Bulouenbes es" Le" Go ey
jeyedsueds og" Lh 20'- QL" OV
WI|i E98Y 6g" 8c 1S’ Sr" ly’ ve
seoe4
|| uonluBooey St 19° €@ 1S’ le rh lL
VdA
| IE98Y 8g’ Or’ ay 6y ee 00° 29" ve
wey Aq|| TE99Y 0S Sy Key 26° cy 92° vs" Sr vs"
Auoypny uoNIUBooeY
|8q GZ Ls 9g" Gg €9° 62° 69° 82" 6S" ey
ueal\ 00! 00! 0'0L 001 66 exer oxo) oxey exer (exe! 0'0l
gs i 62 Le oe oe ve Ze 62 Le Le Le oe
ca —eee
 259
Appendix A

Le Le oe ore Le oe oe Le oe oe ze as
0°01 0°01 LOL 0°01 0'0L 66 201 LOL O'0L O'0l 0'0l ues
es 1S” ze" eZ’ 82" OL Sr’ ye" rid €Z' eq UoNUBoo8y AuoyIpny
Ov’ Ov’ 2L' 6 80’ se’ 82 Or 19° E98 || Di we4
rl Ly’ $0'- 02 ve’ LL €0° Aa E98Y Il VdA
62" 62 Ol er’ €0° ote Ze uomuBooey || se0e4
02 10° 09° 62 92° 06° We98y | WI
6e° 70° Sl’. 92° Le" ueds jelyeds
ol 6h Sl 8 Hulouanbes N-7
62° ey 0s’ eOOY | Iq we
Z0° l2 WE98Y | WdA
Col uoljiuBoo8Y | Se0e4
e984 | WI
[/2994 Hovey je094 ueds ulouenbes |jedey jye90y Gooey i1e994 : ysa}qns
I]299y
pny120
99
|| Wd wey WdA || Sa0e4
Il | WI jeneds N-1 | Wd wey | WdA | 11
SIBAK GY-GB SaSy :Sa10IS peyeds 3sajqns AreUNLId [[I-SINM JO SUONe[sI1I0I19}UT “QZ"y 31GeL
eee
 ajqey “7Z"y suonejas
JO 10319}UT
TTI-SINM saxapuy
pue :sayisod
SosyuioDLI-9T SIB9K

260
Aye saxepul Asoyipny sseo01g seysodwog
soppy SIA Ss wu}= phy SIA pny ues) BY= jel-ajBuig
= Bujwes]
ee
Pq s
2042d wey way= Bulusee] adojs UONUAa}eY
=: JeAaY
Appendix A

Aiewiid sexepu
Aoypny ayelpewiw|
jens! ayelpawiwy| Z0°
eyelpewiuMowe OL’ el
Aoupny pekejag GL ve gy i
jens, pefejaq 10 eae; 65 er
Auoypny uonjuBooey
pakejaq Ly ye Or cee OL
je1ue5 Mowe 9g" 6S" 1g” 6L gg” 89°
Bupom Nowe Ge’ Ole a se GO. ge" 92
Asoyipny ssa01g seyisodwog
-e/6uls
je) Bulwee v6" 00° 09° 09° 00° ly ly 8c
buluee7adols Le 00 GZ- ereyG: Os gO- €0°- SO’ gg’.
UO!Us}9Y ve og" Sy QL’ Le’ Be" OL’ QI id 90'-
[ASLIOY L0°- ea: ec fc. 60° 89° el 80° SO” ZO ae
ueay\ 661 O02 666 661 661 0'OL 86 €'02 661 0'02¢ 0°02 20
as oF Bb L9 9r ob 62 L8 oP oF Ly 6'P hed
 261
Appendix A

nnnLE EEE SUES

66 cid Ly o's v8 Sy 9S as
Ea vv Ly vs 6's
L614 6'6Y LOL 661 661 O0r 661 02 ure
20 861 Z02 0°02
00° 8S" 10° rc. Gk t= 0z’- jeAeuOy
8e"- QI ole 90'-
19° 0g 0€° gL Sv Lo ov UON}US}OY
bE ce Le
Se’. €0'- SO’ S0’- 20’ oO oo. 10 Ol. edojs Buiwee]
vv LL vs ee gL 6L Le v6 Hurwee7 jeu -e|6uls
seyisodwoy ssad0jg Aioyipny
8h" Le" ge" cyt Le" Ly’ Mowe Suro
aL’ QL’ 8" 68" go’ Lg" Mowayy je1eues
re 0 GG—as SG 6S" peAejaq uon|uBooey Aoypny
Le ok 98° 6° peahejag jensi\
LY ce 68° pefejag Aioypny
BL 98° Kioweyy ayelpewuu|
9¢° ayelpeluu| |eNnsi/\
aye|peww| Aioypny
sexapu| Aewiid
ado|s Bujuseey Wwe we |9q 994 \9q \eq Wei wu] wu]
JeAalJay —UOI]UB}OY
Bujwes] ~=jetuj-ojbulg = BM ue5) pny SIA pny Wy SIA pny
lta deh ee nee NE hie
saysodwo) ssev0ig Aoyipny sexepuy Aiewid
I ae “gZ"y B/GeL
SIBaK GI-8I Sey :saysoduroD pue sexepuy [TI-SINM JO SUOHRIa110I19;U]
 BIqe] “GZ"Y SUONL[A1I
Jo 0IIOIU]
[I[-SIWM saxapuy
pure :Sa}Isodu
SA8yroDF7-0Z SIB9X

262
— oe
Arewtid saxepuy Auojipny ssao0ig sayisodwog
pny SIA wy pny SIA pny uas) BM= eu-ejGuig
= Buywiee]
wu]
~ “WU
| wey fed “ed 294. ‘eq way Wat Bujusee] adojs UOI}Ua}aY
= JeAS}94Y
Appendix A

Asewlg sexepu
Aoypny ayelpeww|
Jens! ayeIPawWly| ly
eyeipewuNowe, 98° 98°
Auoypny pefejag a: gy
sensi, pehejag 8° 68° 98° gS ;
Auoypny uoquBooey
pakejaq €9° rss efor 6" ge"
jesus Nowa 98° el £6" 28° fel 99°
Bupoy Moway ve Le 9e" Le Lo 60° ce
Asojlpny ssa90ig seyisodwog
l-e|6uls
jeu Bujwesy 26" 9¢" GL GL’ 8y rAch GL’ 62°
Bulweeyedojsg 20" 80° €0° 6L L0° Z0'- 60° ae eon
uoua}ay ey’ Ge" ov OL ee el eg" ee ge" €0°-
[PABUIOY 90’- oo. Oe le. vO: 29° 00° vl LO’ ra - er’
ues 661 66 g66 102 ooze 0'0L 0'0S 0°02 L'0z 2:02 0'02 L0-
as 7s 2s 8°8 L's e's oe bbb 6b 6'P Ly oP oe
ee ee ee
 263
Appendix A

a
Le es Lv 67 vs ZO0L ce LY gg EL SV vS as
S61 e'6l 861 £02 L6v 86 02 66 66€ L'02 Z 61 uesy\
c 0-
ce - ke- 20° 90’- SO’ es" 00° €dg-, 60) SO L0’- [EASH}OY
OL oS Go 08° 1S” Sy 98° OL by €9° UOI}US}8Y
Cre 0d vo Ol ) e 62° 6h 80° 02° adojs Bulweey
ve LL gQ° 2 Eas bee Sl 68° Huiuee7 jeu -aj6uls
seyisodwioy ssao0jg Asoyipny
Ly 02° ce’ 9¢° Ly ve 6¢° Koweyy Bupoi
gL ok 88° €6° ach 18" Moway je1eue5
ve 19° ol’ ve bl pafejeg uonuBooey Auoypny
LE’ UR L8° 80 pahejag jensi\
18° 0" 06° pefejeg Aioupny
oe ce Kioweayy eyeipewiy|
; 02° eyelpeww| |ensi/,
eyeIpewwu) Aoyipny
Aiewiid sexapu|
JPAQ}OH =—-UOTJUN}9Y adojs Hujusee) we wai j9q 994 Ese] \20 Wei wu wu
Bujwseey = jetuj-ajuig = BM ues) pny SIA pny Wt SIA pny
saysodwoy ssed0jg Asoyipny sexepuy Aiewiidg
i
s1eIK GZ—-GZ SeBy :Sa}IsoduioD pure saxapuy I]I-SINM JO SUOTeIa1I09I10}0] “gE"y aiqel
 Bjqel “LE"y SUONP[AIIO
JO IIOIUT
[TI-SWM saxapuy
pue :soyIsoduroy
Sosy DPE-OE SIBOK

264
e— eooe.-—
Ajewid sexepul Aioyipny ssad0lg sayisodwiog
sopny SIA wu}= pny SIA pny was) By= jelu-ajGuig
= Bujueey
ede
Pq 994 jeq way way= Buluee] edois uoquayaY
©: JeAewiay
Appendix A

Alewidg saxapu|
Aioypny eyelpewiu|
JeNsi/\ o}e|PSWW] Js
ayelpewuuy|Asowsy v3” 1g"
Aoupny pahejaq 6" «6 = og
jensi/, pahejaq ov 98" gL wy
Auojipny uonluBoo8y
pefejeg 99" ve LQ’ 69° Se
je1eues Aowayy rach 89° le 88° disk Le
Bupon Mowayy AcE ogi LG Lg Lh ve" Lg
Asoylpny ssad0Jq seyisodwioy
jeuf-e|6uisHulwee7 98° bea oL vl co 6S" 69° ev
Bulwee7ado|s sy ol 9¢" 8" bh Goi oe 0¢ - Cl
UONUS}OY 9 8d Sv GLa 60° oy 6S" 8c Ly 20°
JEASIIOY 6ge HO) (4g Me 60° 9¢" - cl COi- Piles kom 6e"-
ueal/| 022 sE0d Ve07 0G 66h vO Z0S 8'6l v02 02 661 €0
gs eG SY €8 eG BY 8°¢ vOl Gs 67 os oP Ge
 265
Appendix A

22
VC ev LY 87 87 90l Gc 97 67 08 87 67 as
€'0- 60¢ €'0¢ £02 L'0¢ e1s LOL y0¢ 8°02 9°07 L°0¢ S02 uesy\
4 Be Gc'- LE vl lo 6S" 20° Se 140) LO’. 80° JBASIIOY
140) oP" Lg 19° ev 9¢" OL Ac 9¢" 0s UO!}US}9Y
ets Le 82" St 82" tes” Fa 12 02" edojg Burweey
6e° SL 89° 8e" 18° GL ce 06° Bujwee? jeu -e|Buis
sayisodwoy sse00iq Asoypny
9’ gg’ Ly’ 9 «=—oos—(s«é 0s" Kowa, Bupjonr
08" 6L 06° 6’ 89° v8" AKioweyy jeseued
Ly’ ye rach Les roy peahejag uonuBooey Aioyipny
1S’ 6L 88" Sy pahrjaq jensi,
28" Sy’ 26" pehejag Aoupny
€8" 78° Kiowayy ayeipewu|
6e° eyelpawty| jensi,
ayelpewiwu| Aoypny
saxepuy
Aewid
JPAQJ9H ~=—-UOI}U9}9Y adojs Hulusea7 wa Wwe [aq 984 190 leq Wwe wu] Www]
Bujweey = jeluj-ajbuig = BM ues) pny SIA pny ww} SIA
sayisodwo4 ssed0ig Aioyipny sexepuy Auewiidg
SIBOA FPP-SE sa3y :sa}ISOdui07 pue soaxapU] [II-SINMJ° SUOTea1IOD19}0T “"ZE"y ajqel
 a1gey “EE SUOTefa1I0sIa
JO jU]
III-SINM soxapu]
puke :sajyIsoduroy
Sosy FS-CPp SIBaX =

266
aC—mX————
eg,
Alewid sexepuy Asoypny ssed0ig soysodwog 5

=
=

SIA
SIA

wy
jew

pny
wu}

pny
pny
eq
lq)

ua)
994 jeg wey

BM
way= Bulusea] adojs UoTJUa}aY
=: JeAau}ay
Ba
<

-ajGuig
Auewld sexepu]

Bulwes]
Aioypny ayeIpawiw|
jens! ayeIPewuy| Sr
Aiowsay| ayeipewiuuy| 18° G8"
pehejag Aoupny €6 LV €8
pahejag sensi, LS” 8° 62 6y
pehejaq uonlubooey Auoypny cole) 87 GS v9 oe

We
BL
cL
13°
ce"
68°
Suro Mowe On" noe Gy gr Ly” 8y"

jesauesy
Asowayy
Asoypny ssed0ig Sa}isOdwioy
-aj6us
jeu, Bulwe97 16" O8ts«éH
= 6L ly 1g QZ’ Ly’
Bulwesyadojs oe 92° ve Sv ve co. 6 OL 00°
UONUS}9Y OL lo’ vs €8 0d Lv €9° bee 8S" Le
JEASUIOY lo- ore - CC ee Bie 6v 90'- v0'- opr €d'- 0€"-
ues 661 00¢ 66€ 661 0dc 00l 0'0S €'0¢ L'0¢ €'0¢ 0'0¢ 00
ds es 0's 88 es 0S 6c ZOL 6Y 67 BV vy 9°¢
Appendix A

267
9¢ Sv 0S os os Z0k be LY og Z8 os bs as
00 661 0°02 0°02 2 02 6 6V 0'0l L°0¢ 661 66€ 661 0°02 ueel|
fc - 9}'- 80° vl 10° gS PE. Lo- OF- 0¢'- 10) [eASH}9Y

sa}isodwoy
€0° Lv 6c" IL ov cv 8 LS ov 9S" UONUS}9Y

Aloyipny
ssad0ig
ee'- £0°- HNN 10) Ot" 8b 60° 40° 60° edojg
Bulweey
Ly’ LL 99" ge" 7 eee 4 18’ Hujwes7 jeu
-2|Buis
6y ov ¢ vv 6 SC’ Ly Bupjon Mowe
ZE 6L’ 16" 68" 19° €8" Mowe jesaues
se’ 89... ‘2. 18 ive pefejaq
uonjuBooey Aioypny
€" bl 18° by pafrjeq jensi/\,
ach 0s" L8° pefejeg Aioyipny
v8 vg" Mowe ayelpewuu|
er’ 9yeIpewu| jensiA
ayeipawiu| Aoypny
saxepuy
lewd
|BAG}94 ~=—-UOI}U9}04 adojs Bujusee7) Wa wa |@q 984 j9q 1eq Wal wu] wu
Buywes] = etuj-ajulg = BM way) pny SIA pny ww} SIA sip
sayisodwo4 ssadoig Asoypny sexepuy Arewidg
SaBy P9-GG SIBIK
:sayIsoduro7) Jo pue
I]I-SINNM sexapu]
SUONea110010}U] 31/ge] “7E"y
 96 6v se vs oS L Lh Ove LY es 68 Sv cic ds
mt) 0'0z 66k 0°02 0°02 10S 0'0L g0e «66. 666 e66l waco! ueo
9¢". Or. Ae Sl €0'- 9g" 60. 62. o2to ac6= el [eASHIOY
ral ly oe" gg’ 6e° ve ik: n:| 1S’ uoUa}ey
Al co Ly €e 6E° LE aROr Wee ge" edojs Bulwee7
ov 1D 1g gg’ oss 1S" €6° Hujwes7 jelu,-e|6uis
sayisodwoy ssed0i/g Aioyipny

uonubooey
Ach rAch 6€" Sy it 0S" Kowa Burom

Auoypny
=paAejag
GL 78 L6 . 68 Pee og Mowe [e4aues

6S"
19°
ey
ly
19
; 19° Ach cree €9° peAejag sensi,
08" 0S" 98° pefejeg Aoypny

9ye|pewy|
48° 16 Kowa) ayeipewu|

/ensiA
6S"
eyelpewwu| Aoypny
Arewidg
soxepuy
= [eAeeY UONUD}aY
«©—-
= adojs Huluseey Wel Wal 99Y J9q j2q leq wei wu] Wu]
3 jeu-ajbuigs
= Bujwes] BM= uas) pny SIA phy Wy SIA pny
5 sayisodwoy Auoypny ssed0jg Ajewid sexepuy
Qo

268
= S8BV 69-G9 SIBaX
:sajzIsoduroD JO pue
[[I-SINNM saxapuy]
SUOTIEAIIOIIOU 3jgel “GE
 269
Appendix A

LO ——@—
ee
Lv 8'Y Ls LOL 6¢ ov 0's 18 0s LS as
9¢ vv
L°0¢ L°0¢ S61 20S 66 c 02 0¢ OO” 661 L'02 uesl|\|
L°0- 0°02
S0’- 60°- LO" LO’- Ach L0'- 8¢'- vl - él - Ol’- JEASLIOY
CV -
lo ve ve S9 LC: cto 6L Ace As 0S” UOI}US}9Y
ra gL’ Ach Or’ ee" is =O LZ ey’ adojg Bulwee7
ey’ 19° es" 92° 69° 69° ae 88" Buiwee7 jeu-e|6uls
sayisodwoy ssad0/q Aioyipny
ys" Sv’ Ov sr Ss’ ey’ or Kioway Suppo,
el gL 88" 98" 6S" gL Miowayy jes8ue5
ee 29° 9° ez 99° pefejeqg uolnlubooe8y Auoypny
ey’ 69° Wks Se pehejeaq jensi,
pie HAS) 98° pehejeq Aioypny
08 18" Miowey eyeipewity|
62 syeIpawity| |eNsi/\
ayelpewitu| Aioyipny
Aewiid sexapu|
jeAeu}ay —uoN}Ua}aY adojs Huluiea7 wel we jad 99Y Ee Pq wey wu wu
Burueey = jetuj-ajBuig = BM ua) pny SIA pny WH] SIA pny
sayisodwoy ssed0jg Asoyipny saxapu| [yew
————————————_—_—_—_$_$_$_$_$$————KQq@«
errr
UT “of
steak FL-OL sa3sy :sayIsoduioy pue soxapU] [II-SINM JO SUOT}efa1109190} o10e1
 age) Ley SUONeIaLI0010}
JO u7
[II-SIWM saxapu]
pur :soysoduioD
Sesy 62-GL SIBAX

270
Aiewid saxepu| Aoyipny ssad0ig saysodwo9
opty SIA Ss wu}= pny) SIA pny uak) By jeluy-ajbuig
= Bujwes]
wu= wu} wey ease 984 jq well wey Bulwes] adojs UOJUa}aY
=: [eAaey
Appendix A

Arewld sexepu
Aoypny ayeipewiw|
jens, ayeIpewiu Gy
ayeipewu|Nowa, 998° ve"
Aoypny pahejag 88" 1g” 28" a oe i
jensi, paxejag Sy 68" pL a
Aoyipny uonjubooey
paxejeg pl 6" ge BL LS
|e18ues) Mowe 08" HE 06" 16° 1g” v8"
Burom Nowa 6e" Be" Gy ly Ly le 6h"
Asoyipny ssad0/g seyisodwog
-e|6uis
je Bulwsee7 06° Ly BL’ el’ ge’ 69° 89° 92°
Bujweeyadojs Sy ge 8° 8S" Ly Sy" lS’ ee re
uolUua}oY oS” Be" vs" 28° oy 19° LL by Or Ov"
[PABLOY Z0'- 90° 20 Pe. O° vy 60° 80'- 00° Go’. 0z’-
uea|\ O02 O02 666 O02 002 oxo! 0'0S L'0z 661 v'02 0°02 00
as vs iis 169. eG. ON Le bh ly ral er L's Ze
 age] "BEV
UT SUOT}e]AI110010)
JO [I[-SINM saxapuy
pur :saysoduroD
Sesy /8-08 SIBI
SSS c
Arewid saxepu| Auoypny ssed0ig sayisodwoy
pny SIA wy pny SIA pny ues) BM jetuj-ajBulg)=—
= Bujwes]
wu] Wu] wa \2q \9q 99y J9q Wal way Bulusea7] adojs UOT]U9}0Y
=: JPAe}aH
Aewilid sexepul
Aoypny ayeipaww|
jensi/ a}eipawuy| es"

28°
88°
Mioypny pakejag 68" ag ie
jensi/, pefejeq

Moweayy
ey Ss gL 8V

eyeipewi
Aioypny uonlubooay
pefejeg GL es gL OL LS"
jeseues Moway €8" 6L €6" 88" 1g" 28"
Bupyon Mowayy ge" NV OOO we eb" ss’ os’
Aioyipny ssa90iq seyisodwoy
-ajuis
jeu Huluee7 L6" 6r 62’ Lg" rAd OL LL ge"
Buiweeqado|s es" 9a" Sy 1S” ig oy or" 02° i
UOI|Ua}OY os" Ly rc gL oy ge" 99° ec ly ZV
jeAaUIOY 90°- OL 90! Szim 80° 6y" SO” Le’ 10° 60’- ge".
ues 002 O02 OO0r O02 02 (one)! 0'0s 86h 0°02 86h 202 00
as 2s es 26 7S rae) oe oma 9's gr ov BP v2

271
Appendix A
Ss B/ge] “Gey SUONeie110d19
JO )UT
TII-SINM Saxepu]
pue :sayIsoduroD
SeSy 68-G8 SIB9X
WN &
=F
Alewid sexepuy Aioyipny ssad0ig sayisodwog 5
pny SIA wy pny SIA pny uae) By= el-ajBuig
=
as

Buluee] 4
wu] wu] wal) jaa jad 99y Jaq waif wall Buluses] adojs UONUa}OY
==: JEAdL}aY <
Arewld sexepu]
Aoyipny ayelpewiw|
/eNs!A ayeIPSWy| 6e"
ayelpewu|Mowe, eg" v8"
Aioypny paxejag 68° vy 6L
jensi, pekejag ey 6L° bl ole
Aioipny uowiuboosy
peejag L9 9g" 19° te Sy
je1eues Mowe BL 99° 98" 68° v8" BL’
Bupoiy Nowe 02 02" ve Qh ve ve ve
Aioyipny ssed0ig sayisodwog
-e|6uls
jeu) Bulwee7 BL’ vy ae ye Gy’ vs" 19° Ol
Bulwee7adojg 09° LO’ 6E° Lg’ Sl LY’ 6e° 0° Zl
uoue}ey 6S" oy 09° 6L° 6y gS" cE li 79° 90°
[PASUOY Ele AG: 0 cy 6y 60° Ale 20% 60'- ee
ues|\| “0d “10 TOr 0 66. 001 0'0S 102 0°02 L'0z 0°02 L0-
as 6'P me) 8 LG ze) Le bb 0's er os 0's e?
eee
eee —_——
 age) "On'Y [[I-SWM sSUOTe[e1109
JO Areurig
19)U]pue [ejusuIa;dd
3saIqng
ng pae9g *$9.109¢
Sasy 67-91 sea
SSS
e
ONT «=| S8898J
«1 WdA
| wey «(19d TIM
= 1HA N-7 jeneds
= jeneds jeneds jeyuew
= W5iq
4sa}qnS jjeday
= ayy
= Hodey
= eoay j}e0904 jjeo0y jjevey Bulouenbes
ueds pmjueds pmygueds jou0d ueds
AT| SNEWS] 18"
sack
| uolilubooey 80° 10°
VdA
| IE98Y 8Y" ey’ ch
wey Wd| IJB98Y Se" 8c co Zi;
IM| IE98Y Gv 9° vl es" 0"
HA| IE98Y 6° 0g" gl Vhs 92" Ov’
N-7 Bulsuenbes teemy ce" el ve er’
jeyedsueds oo lo 90° co 02° 02° Ov’ 0S’
jenedsueds puemo 0c 6L v0" 6 Le io) 1 Sv 98"
jeyedsueds puemyoeg8k 6L SO" 6° oi Lo Le 6E ce oP
|e}US/\| |ONUOD 82° ve OL 92° 6L 6c ee" LS) LE 8c 9¢"
161g ueds 0¢° Go Le 02° Sk ve 9¢" €9° cv 9¢" 9¢" Or’
WI|] 2984 v8" OL fA Lg ve 8v" 6e° 9¢" od’ 6l Ld So 92°
WT|| | ONHeWSY gL gL a4 Lv So 8" 82° 92° 02° QI" 0c Jat ral
seoe4
|| uouBooey 60° 20° el 02° So 6L co eis EEe 60° 60° io) ck
WdAIl IE98Y ey Le: St gL Ale Lv 6¢° Le 6k QL" 8k Le 6k
AJoyIpny UONIUBODey
[8= 99° cs" vl ov Ke Sv LS Lo 53 }a\ 6h ol co co
wey Iq|| I]E998Y 8" oe 82° €¢ 6 Ge ve’ Gl Go lo co ko SL
IM|| E984 Lv A 0c or [Kes Ach Ov’ Oe" 6h A lo 62° €o-
7M|| uouBooeyY 9¢° Le Bl 82° eo or 92° Go 80° 140) OL Lo co
YA|] E98 8c ve St Lv’ So ov 6S 6e° oe" 90° 90° 0g ve
HA|| uonuBooey ee eC ck ce €¢ LG 9S" Lv 9" Lv 8e" ve ce
HA|| Adoo Ql [ye 00° €¢) 8h 82 os" hee 6c" lo Le Go vo
ueel\ 66 66 OOl 0'0L 0'0L 001 001 001 LOL 001 0'OL v6 LOL
as Oe O€ Oe Lis 0'€ O€ O€ OS Oe O€ CS 8c 62

273
Appendix A
 a1ge1 "O7' [I]-SIWM, suUO}e[911
JO 0919}U]
Areurlig
pue [eyUauIaT
}sa}qng
ddng Paes *S9109§

274
Sesy GZ-9I SIVaX (panu1juo2)
eS
WI WTESE Se024 IWdA PMY Say wey 19d TM HIM YA YA HHA
s9}qnNS yeoay
= Way
)=s Booey
= e994 i290 ~=—sjeoay eoey
= jyeoey jjeoay Boosy Ado)
Appendix A

AT| OHEWSY]
seoe4
| uoniuBooay
VdA
| B98
wey Iq| JE99Y
1M| 12984
YA| 12984
\—q Hulouenbes
jeyedsueds
jeiyedsueds puenuo4
jenedsueds puemyoeg
[e]Ua/\ JO4JU0D
u16iq ueds
IW 2984
WT|| oeWSY] €8
seoe4
|| uonluBooey fi 80°
VdAIl 2984 ov Le lo
Auoypny UoNH!JUBoDeY
|8q 99° vs 6l Aas
wey Wd|| B99 LS 0g" “Gi eo 0g
IM11 29984 es ce 6L 6y ov Gb:
1M|| uouBooey 6e° Se vl Le Le 92° 4g
YAIl I1299Y 6E° ve ve ev Be 62° ov 0¢"
HA|| uonuBooey ee LG ae 0g" Gh 80° 0¢" 02 Ach
HA|| Ado ia 20° 82" Qh Lz 6k el er le
ueay\ 66 66 O'OL 0'0L 0°01 0'0L OOr OO 0'Ol 0 COL
as Oe Le Oe O'€ 0'e 0'€ Oe Oe Le Oe Vs

——
 age] Ly'W
* []I-SIWM suONe]a110
Jo 9193U]
Areull1g
pue JeyUsWIaTd
389}qGng
dng Pag *S910IG
Sa8y F9-0E SIVA
nSnpaSSESnsnsssssessssnesesseeseseneeeeess
SS
———— =
TWIT =|WT «6188084 WdA
| wey «(19d TIM IYA jN-7 jenedS neds jeneds jejuey
— ubig
4s9]}qNnS Jeoey
= Way
)= Booey
= eda
= 2984 jjeo0y jjeoey fBulouenbes
ueds pmyueds pmygueds joju0d uedg
AT| oWeWOY! €8°
seoe4
| uonlubooey vl ale
VdA
| E984 9" BE” 86s 9
wey D4| ]B99Y Se zs.a Ley or
1M| E984 ys" 8S" ve es” ve
HA| IE98Y 6° 0 tc by Se Le
N-7 Burouenbes Ge iy ge" ee" ge" ey”
jenedsueds ce co 6 60° 8c Gc Sv ev
jenedsueds puenuo4 le 7k eh 1A gl (AS Le Se 8
jenedsueds puemyoegte 8 ANS Le ve ve ev Se" v8" vy
JeJUS/\ JO4JU0D vv 0g 02° Go ve Le 8c es Le Lo ce
y161q ueds 82 6l Oe 4 60° ve 0¢ Le" 8s LE re aoe a Yi
WI|] 2994 18° at 0c cs Lv sg Ov Ly ve €o Ase 6° 8c
AT|] oWeWOY! gL eg eb Ly Le cv" co ve Go io) £c Le oa.
seoe4
|| uomluBbooeya eb 89° Le 62° oe" Lo ve 8h ts vi co co
VdA
|] 2994 ev LS Ge se" lv 6" Ly ov 8c ) 82 82 Go
Auoypny UONIUBodeY
(eq 2Z 19° Oe 27 Lv GG: ov Se LE: ve Lo ee Ov 8c
wey Wq|] JE99Y 6° or 8c ov c6 ve 9g €¢" 8c lo lo co vo
IM|] T2994 8v ov 0c 1g 6¢° 6S" Se ve Ka 6L ve ve BL
7M|] uonuBooeY Lv LE: eo: Lv Se 09° 0 Go 9} 60° 9} 60 Sl
HAIl HB98Y LS) 2 02 Se ve Se g9° 6¢° Se’ 9c SG 6h Go
HA|| uonuBoosy Le og" ea ov vy LG. L9° ov vy ve" Le Les Le
HA| Ado 9° Ze A jae Ol Le Lv 82° lo’ 6L 6L Co co
UeS|\ LOL dO OO OOL 001 001 001 LOL 0'0L 001 oe) v6 LOL
as Oe be AS Ove O€ Be O€ be 62 6? Oe 8? 6?

275
Appendix A
 a1ge] Ly'W
* [[I-SWM, SUOTe91100
JO 19}U]
Areurlig
pue [e}UsUIaTd
1saIqQng
dng Paes "$9109

276
Sasy F9-O|E STVIK (panu1juod)
WNT WT Se0e4 I¥dA pny Sey wes 9ld1 WIM: HIM NYA YA HYA
ysa}qns jjeoey way), Booey jjeoay j2q peaay
= jjeoay
= jedey jyed0y Boosey Adog
Appendix A

WT| SNeWSY]
seoe4
| uoljiubooay
VdA
| E98
wey Iq| E994
1M| 2984
YA| I298Y
\—-q Hulouenbes
jenedsueds
jenedsueds puemuo4
jeedsueds puemyoeg
Je]US/\ [OuJU0D

WI|| e99Y

y6iq
ueds
WI|) onewey! gg’ :
seoe
|| uoljupodeY 7° ZAIe
VWdA
Il E98Y Sr Ors Go’
Aloypny uoNIUBooeYy
JQ Ss LZ = 4g 6h vy’
wey Ad|| TEDOY Sr’ Or ve Or’ ce"
IM|| E994 or” ty og" 0s" ay’ 6°
TM|| UoIUBOd8Y sr" 6 62° 6" vy ve’ 9¢°
HA|] E984 EG Lee 6c LE Le’ 9g" ty’ ee"
HA|| uoliuBo0o8Y «GE ee v2 oy Or" ey" Le" dg: G9"
HA || Ado 86 ve Lh 8 6l vl 62" 9¢° ge Be"
ueel\| Geo Vor 0) ee 0) LOL LOL Sting Soo “CO” QU COL”
as OF oe Oe oe oe Le oe Le Le oe oe
 alge "any [II-SWM
UJ SUOTeII110
JO Areurig
919)pue Je}usuIa;dd
3saIqng
ng pae9g ?S9109¢
Sady 68-S9 SIBIK
SS
eee STE
IW JWT 189984 |WdA wey $1%d LIM YA1 N-1 jeneds leneds jeneds jeyuew
= w6ig
ysa}qns =jeoey way
)= Booey
= yeoey
= 2084 jjevay jjeoay Bulouenbes
ueds pmyueds pmygueds joju0d ueds
WI| oneweYy! v3"
seoe4
| uoliubooeY 6L 02
VdA
| ITE98Y 1 ov St
wey Og| JE98Y 6y" vy Ge ev’
IM
| 12984 os 6¢° €o° (Ace 6e°
HA| E98 ge" ve La: 9¢° SY Se
N-7 Burouenbes Be = fe ~ Oo" ze" Se ob ve"
yeNyedsuedg 82 gl SI cl ce 6h ce ey
jeyedsueds puenuo4 oc vl te 80° vi io) 02° ge" c8"
jenedsueds puemyoegLe’ 6l vl Sb Go 02° Se 8c" €8° OV"
|eUs/\| JO4JUOD ve So. 20° ZG i) €¢ Gl ey 62° 61 AS
y161q ueds bz’ sae" Ol”) ce Qh 02 lL 8h" ey" Ze" i eb"
WI|] E9984 v8" A Sy eo 87" gS 6v Or OV Go co tc €¢° ve
WT|| SHeWSY! ll gL Lo cv Lv vy 9¢" ve’ 02° Ze gl co co
seoe4
|| uoniuBooey 02° 8k 19° 8h Stam Lo ce" co 1b LL 14% 60° LE
VWdA
|| E994 Lv 8° 8k €8° 8c" eS" 8e° ve Gl 60° 8k vl 8h
ALoyIpny UoRIUBODaY
8 OL 9S" Go 0S" wy Ach ce Sv Go’ co 6a Go 8c"
wey Sq|| IJE98Y 0S vy ino vv 68° Sv vv’ Ge 82 8h Le io 6h
IM|| T2984 ee ve ZV Se 60° 67 HA lo 60° o) 90° 60° SO"
7M|] uouBooey ev ce" lo ov ee 09° 92° oe eh oS cl Sl Ae
YA|] IIE98Y Le" Le oo ve" Lv LG cg 60° 02° Sk co fe 60°
HA|] uonuBooay Se 9c eo Ly LG: ctom g9° Se ke £ci 0g 6L° pa
HA|| Adoo wate “ee cc. LZ 0¢° 92" vb’ ee" 0€" he 82" QL 0z
uee/\ OO OOF Ook 0'0L OOL 0'OL LOL 0'0L 0'OL oe) O'OL 96 LOL
gs O€ O€ Oe O€ O€ O€ Oe GE Oe O€ Oe bit?
| 6¢

277
Appendix A
 age] “Zh"Y ITI-SWM suUOTe[9110
JO AreulL1g
pue
910}U] [e}UsWIa[d
}S91qGn¢
dng PII *S91OIS

278
Sasy G68—-S9 STeaX (panu1juod)
IIT JIWT [189084 |IWdA PNY 99y wey [3d HIM IM Il YA =TIYA YA
ysaiqns jje904 wey, Bovey ~—syeooy eq ~=—Sseaay eda
= jjeoey jjeoay Gooay Adog
Appendix A

WI]
| oWeWSyL
seek
| uonluBooay
VdA
| IE98Y
wey Wd| JE98Y

|
1M
YA| I1298Y

2984
N—-q Hulouanbes
yeiyedsueds
jeiyedsueds pJemio
jenedsueds puemyoeg
|e]US/\ JO4]UOD

WI1 e99y

16iq
ueds
WI|| oneweyl 8g"
4 Se0e
|| uonlubooeYy 97 970° ¥
VdA
|] E98Y 6 eb LZ’
Auoylpny UONIUBooey
}8q =649°~—s= 6S 92° eS’
wey Iq|| E994 6b Ze" ey Ly
IM|| 994 eS LE 9¢" 8c" og"
TM|| uouB098Y Gv ge Oz ay 6Y' 9¢" 09°
HA|] E994 Ge fe 8a" ce’ Ze ey ee Se
HA|| uonJuB099H ze ee" Ly’ LY’ le’ 62 ce" 1g
HA|| Adod el Ge’ ve 0g" AN el ve ie 9¢°
uesy\ oor =Unr ODL “OOS COL LOL “oorrrcor
OUD COOOL
as OF€ o€ oe Oe oe Le Le Oe oe oe oe
et pes pnnnes
nee
 APPENDIX B

Discrepancy Score Tables Based on

Predicted-Difference Method

279
Appendix B

Table B.1. WMS-III Index Scores Predicted From WAIS-III FSIQ

Scores: All Ages
WAIS-III Aud Vis Vis Aud Gen Wkg WAIS-III
FSIQ Imm Imm Del Rec Del Mem Mem FSIQ
45 77 73 67 45
46 77 74 68 46
47 78 74 68 47
48 78 15 69 48
49 79 75 69 49
50 79 76 70 50
79 76 71 51
80 76 71 52
80 Wf 72 53
81 77 72 54
55
56
57
58
 Appendix B

Table B.1._ WMS-III Index Scores Predicted From WAIS-III FSIQ

Scores: All Ages (continued)
WAIS-—III Aud Vis Imm Aud Vis Aud Gen Wkg | WAIS-III
FSIQ Imm Imm Mem Del Del Rec Del Mem Mem FSIQ
100 100 100 100 100 100 100 100 100 100
101 101 100 101 101 100 100 101 101 101
102 101 101 101 101 101 101 101 101 102
103 102 101 102 102 101 101 102 102 103
104 102 101 102 102 102 102 102 103_| 104
105 103 102 103 103 102 102 103 103 105
106 104 102 103 103 103 103 104 104 106
107 104 103 104 104 103 103 104 105 107
108 105 103 105 105 103 104 105 105 108
109 105 103 105 105 104 104 105 106 109
110 106 104 106 106 104 105 106 107 110
111 106 104 106 106 105 105 107 107 111
112 107 104 107 107 105 106 107 108 112
118 108 105 107 107 105 106 108 109 113
114 108 105 108 108 106 107 108 110 114
15 109 105 109 109 106 107 109 110 115
116 109 106 109 109 107 108 110 111 116
Ti 110 106 110 110 107 108 110 112 117
118 111 106 110 110 108 109 111 112 118
119 111 107 111 111 108 109 111 113 _
120 a
121 ie
122 a
123 a
124 aT
125 i
126 iar
127 ioe
128 aa
129 Ti
130 Aes
131 ne
132 saa
133 1a
134 hae
135 136
136 137
137 138
138 139
139 +40
140 144
141 142
142 143
143 144
144 145
is 146
146 147
147 148
148 149
149 150
150 151
151 152
152 153
153 154
154 155
155

281
Appendix B

Table B.2. WMS-III Index Scores Predicted From WAIS-III VIQ Scores:
All Ages
WAIS-III Aud Vis Imm Aud Vis Aud Gen Wkg WAIS-III
VIQ Imm Imm Mem Del Del Rec Del Mem Mem VIQ
48 48
49 49
50 50
51 51
52 52
Ee) bas
54 54
55 55
56 56
57 57
58 58
59 59
60 60
61 61
62 62
63 63
64 64
65 65
66 66
67 67
68 68
69 69
70 70

72 72
73 73

76 76
77 rae
78
78
79
79
80
80
81
81
82
82
83
83
84
84
85
85
86
87 86
87
88
89 88
90 89
91 90
92 91
93 92
94 93
96 95
97 96

99 98
100 99
101 100
102 101
102

282
 Appendix B

Table B.2._ WMS-III Index Scores Predicted From WAIS-III VIQ Scores:
All Ages (continued)
WAIS-Ill | Aud Vis Imm Aud Vis Aud Gen Wkg _| WAIS-il
viQ Imm Imm Mem Del Del Re Del Mem Mem ih viQ
103 102 101 102 102 101 102 102 102 103
104 102 101 102 102 101 102 102 102 104
105 103 102 103 103 102 103 103 103 105
106 103 102 103 103 102 103 103 104 106
107 104 102 104 104 102 104 104 104 107
108 105 102 104 104 103 104 104 105 108
109 105 103 105 105 103 105 105 106 109
110 106 103 105 105 104 105 106 106 110
111 106 103 106 106 104 106 106 107 111
112 107 104 106 106 104 106 107 107 112
113 108 104 107 107 105 107 107 108 113
114 108 104 107 108 105 108 108 109 114
115 109 105 108 108 105 108 108 109 115
116 109 105 108 109 106 109 109 110 116
117 110 105 109 109 106 109 110 111 117
118 a
119 a8
120 at
121 igs
122 ge
123 a
124 i
125 ie
126 =
127 ee
128 ie
129 ae
130 “
131 a
132 ‘i
133 ie
134 dees
135 -
136 =
137 a
138 oa
139 aie
140 ve
144 ye
142 re
143 ee
ae
145 145
sae
146 ve
147 aie
148 va
149
150
150 151
151 152
152 153
153 154
154 155
155

283
Appendix B

Table B.3. WMS-III Index Scores Predicted From WAIS-III PIQ Scores:
All Ages
WAIS-III Aud Vis Imm Aud Vis Aud Gen Wkg WAIS-III
PIQ Imm Imm Mem Del Del Rec Del Mem Mem PIQ
47 UP 79 71 74 TO UE 70 66 47
48 Te 80 72 74 THA 77 71 66 48
49 8) 80 Wie 15 78 78 71 67 49
50 74 81 73 75 78 78 We 68 50
51 75 81 74 76 78 78 WS 68 51
52 75 81 74 76 79 79 1s 69 52).
53 76 82 75 Tl 79 79 74 69 53
54 76 82 as) Wate 80 80. 74 70 54
Bis) UY 82 76 78 80 80 75 71 55
56 77 83 76 78 81 81 75 71 56

284
 Appendix B

Table B.3. WMS-III Index Scores Predicted From WAIS-III PIQ Scores:
All Ages (continued)
WAIS-III Aud Vis Imm Aud Vis Aud Gen Wkg_ | WAIS-III
PIQ Imm Imm Mem Del Del Rec Del Mem Mem PIQ
103 102 101 102 102 101 101 102 102 103
104 102 102 102 102 102 102 102 103 104
105 103 102 103 103 102 102 103 103 105
106 103 102 103 103 103 103 103 104 106
107 104 103 104 104 103 103 104 105 107
108 104 103 104 104 104 104 104 105 108
109 105 104 105 105 104 104 105 106 109
110 105 104 105 105 104 104 106 107 110
111 106 104 106 106 105 105 106 107 111
112 106 105 106 106 105 105 107 108 112
113 113
114 114
115 115
116 116
117 117
118 118
119 119
120 110 108 111 110 109 109 111 113 120
121 111 108 111 111 109 109 112 114 121
122 114 109 112 111 110 110 112 114 122
123 112 109 112 112 110 110 113 115 123
124 112 109 113 112 111 111 113 116 124
125 113 110 114 113 111 111 114 116 125
126 114 110 114 113 111 111 115 117 126
127 114 111 115 114 112 112 115 118 127
128 128
129 129
130 130
131 131

i e
132 132
133
135 135
136 136
137 137
138 138
139 139
140 140
141 141
142 142
143 143
144 144
145 we
146
147 -°
147
148 148
149 149
450 150
ed 151
fee 152
= 153
15 154
= 18

285
 alge]“27° SsaduaIaIJIUsaMjag
G pa}dIpaig
pur poule}qQ [TI-SINM
xapuy sa109¢ parmbay
10j [eoNsheI

286
aouedyTUs
Aq a8yIS dua1a}71G-pe}dIpaig)
INI-SIVM
(OISA
—Popey|
eaeee eee
nn Sen CO
III-SINM ues)
aouRdIUuBIS aby dnoly
xepu| SeBVIIV[0Ae7 GZ-SZ_pz-OZ_-sS«GI
= E-DE
SS -BL_SCZI-91 §=69-S9_H9-SG)HG-ShCph-SE
=6Z-GL_pL-OL
8-08 68-S8
Appendix B

Aoupny SO" SO 860A SG wel Gh McKee eh Gras CRS Bae Exp Gel OK S06
8yeIPewiu| 10° Siz Cue Gl Oe SIG alte Fl ode Rie aka ia 661 8S2 692
fensiA SO" CASH sith Exe 8°91 L6L ez Gu @6: GOL Hel Sig Aish oe o9l
8}e|Pewiu| 10° Ge tise GOeé§ ike L'v?e ecumcale
war gle Gel Bah Ob aie eis
9}e/POW| SO" ZO ye:
Gs |eC10 222 EeeS ers, 9 Zn, CAL Esk Oh hob SL
Nowa LO’ oOF G82 Gol wie Gee wie ORL Ge KS eo cio! COP Giccume hECo
Mowpny SO" ee ie meee ep Gee er cli OSI Zeke -E'Sh. SOL. 2b» 28_ OZ)
pahejeq 10° SESiok Soir eich PZG EL Gehl Cet AGE EAL LL’ BLOCH'OSSC*é«'DG
CG ES 69h cee
jensi, SO" LOL 8h PB). YEG| rif eid MAN eich Ce ial asi eel wish OME
pehejeq LO’ ecoerciO
eee OC eo 0Gae CG) s95 LB= Gide Vile 0c) ERTL IIL (yL- (ove Boke
Moupny
uonilubooeY SO" AL GAL
= SO idk OSE eS e6 Scie Tem Fon SL ESL Ak Gut
pehejaq LO’ GGL LSE Leb LL 698k 79k Geen)oer LoL gel QSL S82b 66
Je1euas SO’ meter eo ch See cetSe GOL OV Fel r ciclo
Seo Merch eS SZ viele
Mowe LO" rE mane mene Gere eiee ey vcsS Spel “SOL” G0= eéb Loe “Get G6L vee seo
BuroM SO" LL 6 GirAll eral oN L'6 8'LL ecIOl mG: OG LE Lh 8 roa Cal: Vee
Nowe 10° r Sie eS SGC VL OLE aT eA IL OMS HES HA ASIN SWANS YARNS
 287
Appendix B

ee
set os: oot OL St OL 99k ek (@9L GSE Gb eereeneee ee Mowe;
ost zk oe Cel oh 62t S8t s0b et Der Gal cheers cor ee Bupom
rSe Gz cle ize le che 6t2 “e2l “Sol Gee Pol Coleen Mowe;
S64 Of FOL gor “zh ol Se. SE Fel) .O8. SSt | ele eee eee jesaued)
60z zil 99 Ov St rk 8SL SIL St OGL wOl | GTILM Gale aci ee pekejeq
rol eek) «6SEL 06 OCOOLL SC isCSECCi‘éiS:C (CC CCCOCSE «CER RO ee eee uonuboo8Yy
Aioypny
Ge e02 o0c 0c 802 le Gee See Obl Eee Whe | Cec m eee pehejeq
ris 29l ook oor 99: OZt -G2L Pat Si F2L GUL pole Ge eo ee fensi
Ose vee 02 s02 zel SBl otc opt FZL O02 Gel PO eles oc pehejeq
‘et St oo: Sok St Obl <BOL FUL Yet Bet Bebe Orie Aoupny
ez sez 681 Soc. 962 zee Ove’ kOe OGL Wee “Sie F Gabe ee ee Aowew
S66: pel e6Pl col Wet zoek eZ eS OS: G8l O2l ) Ori eect ee 9}2|PEWWU
zez fig Let 80c Ole Oecd @0e Gf Gle Ste Gec 8 OOcu mcr cee eye|Peww|
esi 6OL OL QOL OO GZL. fol El S8ZE EGR 22 9 Ont oer eee lensi,
Oz. VS2 fe gle ole oes ree Voi fer Fis eyes e Pie meee ee ce eyeIPewUU|
olZ S6l Zt OL WLZb @ZL S8L GE .OSL —BOLP SSE N:Gep 2 et eee Aoupny
68-S8 8-08 G6/-SL pl-OL 69-S9 9-SS S-Sh br-SE EOE 62-S% 2-02 GIB LI-9t sebvily jeneq xepul
dnoip aby ueay eoueoyiubis HI-SWM
(OIA ILI-SIVM—POtPe AuUarayIG-peyotpeid) asy Aq souBoyTUsIS
[eonsneis 103 paambay se1oog xepuy [TI-SIWM Pauteigo pure peyolpaid Usamjeg SeoustayIG °G"g 3IqeL
 age] "9°g saduarayyIG
Usamjag pa}dIpeig
pure Peulerqgo [1I-SIWM
xapuy sa1osg parmbay
10J TeINsHeIS
aaueoyTUSI
Aq a3y S

288
eusI19jJ1-paistpaig)II-SIVM—
(Old PoyeW
}_}\}k}®}x[]{x—_—[—__—_———--- —_—_e
II-SWM souRrdylubis
uray) aby dnol5
X@puj joneq seBviIV JZI-9L GI-8l 2-02 G62-GZ E-0E Ph-SE S-Sh 19-SS 69-S9 pl-OL 6L-SL 8-08 68-S8_
Appendix B

Aioupny SO’ O's Vk eet jhJi vesl ost 8S LOL 69L L6L Sst Pel Cle Gig
ayeIPSWLy| LO" Oe Wee ed ee 86¢ C836 Lt Le Ge Cie Ke CREOIOR CiCa
jensiA SO’ rae) 681 89OL LOL Z8L OLL 9L6 9'9L 9°91 6'rk Lv vt Ssh Fel
ayeIPpawy| LO" UZ Gee OA 4e He de Cle We sug G8l GL eb v6l vol
9}e/PSWLW| SO’ LOL 9°81 691 rome} LL6 ZL vSl 79 L9L jijJi LS cel vsti 9°91
Mowe LO" ile Gite Cie Old ee Cale SGé6L 902 €le Geo L'6L 991 9€¢@ le
Aioypny SO’ ZGL 9'EL 6EL Gvl 9'9L Z9L 9'EL oSh sel 691 Z91 rl6 6hZ Su
pehejeq LO” Z6L Z9L ayill LS Oke Paice OL} 76h (20 A Onis BBL
= Sc Lee
jensi, SO’ ZGL O'8t 69L ZGL 69} GL VLi 6'9L Sh 9EL Gel O'S} Srl OVI
pekejaq LO" 961 G7 cle sai!1) Gilg Gel Ore Vie L6L 691 LOL 9'8l G8 GL
Aioypny
uomubooey SO’ VEL O?L Lib 6S} 8'rk O'vk Eau vel Jis{ Srl Gel vel outer Q'S}
pekejaq LO” y9OL vrvl O'rk 661 vst oLk eel O'S| Leh ree oSh SOL €6l S6l
[esJoua5) SO’ O'9L LOL Lvl 79L Ze GOL vl 9'Sl Srl Lh e Srl G’Sl fomey Gull
Mowe, LO’ £023= c0e TAGE OCmEE Camm
Si CeeOi: 78h 8'6L 981 Occ 9°81 961 €€ 6le
Bupyonn SO’ Vel eel irs Gel G8 OL Vth Ol Gel d ciel OL GL cel 0'91
Moweyy LO" V'Sh GOL Sl 691 OlZ v'vi O'rk Zvi 691 6S| v'rk O'rk GOL 702
 Appendix B

Table B.7. Frequencies (Cumulative Percentages) of Differences

Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III FSIQ)
WMS-III Percentage
Primary Index 25 20 15 10 5 4 3 2 1
Auditory Immediate 8 11 13 16 20 21 23 25 28
Visual Immediate 10 12 15 18 23 24 27 29 32
Immediate Memory 8 10 13 iz, 21 22 24 26 28
Auditory Delayed 8 10 13 16 20 22 25 27 32
Visual Delayed 9 11 14 17 22 24 25 28 34
Auditory Recog Delayed 9 11 14 17 21 24 25 27 30
General Memory 8 10 12 15 20 21 24 25 29
Working Memory 8 9 11 14 17 18 19 21 24

Table B.8. Frequencies (Cumulative Percentages) of Differences

Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III VIQ)
WMs-—Il eee eee crceniage 2
ee eee ele ee ee IN SF.
Auditory Immediate 9 11 14 16 21 22 23 25 27
Visual Immediate 10 13 15 19 23 25 27 30 33
Immediate Memory 8 11 14 17 22 23 24 27 30
Auditory Delayed 8 11 13 17 21 23 26 27 31
Visual Delayed 10 11 14 17 23 25 27 30 33
Auditory Recog Delayed 10 12 14 17 21 23 26 28 32
General Memory 8 11 13 16 21 23 25 26 31
Working Memory 9 10 12 15 18 19 20 23 26

Table B.9. Frequencies (Cumulative Percentages) of Differences

Between Predicted and Obtained WMS-III Index Scores:
All Ages (Predicted-Difference Method—WAIS-III PIQ)
WMS-III Percentage
Primary Index 25 20 15 10 5 4 3 2 1

Auditory Immediate 9 11 13 16 21 22 24 26 31
Visual Immediate 10 12 14 18 22 24 26 29 31
Immediate Memory 9 11 13 16 21 23 24 26 30
Auditory Delayed 9 11 14 17 21 23 26 28 32
Visual Delayed 9 11 14 17 22 24 26 28 31
9 12 14 17 22 23 25 27 31
Auditory Recog Delayed
8 11 13 15 20 22 23 24 29
General Memory
8 10 12 14 18 19 20 21 25
Working Memory
anna ee
EaEaaa EE

WMS-III Index scores

Note. Percentages for Tables B.7—B.9 represent the portions of the sample who obtained
i

lower than their predicted WMS-III Index scores by the specified amount or more.

289
Appendix B

Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
FsiQ
RD MA WL OL Total |Score

290
 Appendix B
Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores (continued)
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
FSIQ
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total
101 101 101
101 101 102 102
102 2 102 102 103
102 103 103 103
104

= Word Reading PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 201
SP = Spelling
Appendix B

Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores (continued)

College Students
Subtest Standard Scores Composite Standard Scores
FSIQ FSIO
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

292
 Appendix B
Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores (continued)
College Students
Subtest Standard Scores Composite Standard Scores
FSIO FSIO
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
=
101
iv
102
103

correlations for the five WIAT-II composites are .63, .70, .46, .62, and .71.

= Word Reading PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 293
SP = Spelling
Appendix B

Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
FSIO. FSIQ
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

294
 AppendixB
Table B.10. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III FSIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
FSIQ
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total ae
101 101 101
1 101 102
be 101
102 102 103
104 103
105 103
106
107
108
109
110
111 107 109
112 | 108 110
113 | 109 111
114 | 109 111
115 | 110 112
116
117
118
119 -|O
ARWNH
DIAN
OD
120
121
122

124 130
125131

127
| 128
128
129
130
(sine ce
132. 139.
140
14

in the order
Note. Based on the correlations between the WIAT-II standard scores and the WAIS-II| FSIQ. Presented
.62, .73, and .49, and the
shown in the table, the correlations for the nine WIAT-II subtests are .66, .82, .76, .66, .69, .78,
correlations for the five WIAT-II composites are .86, .82, .63, .67, and .86.

PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD = Word Reading
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 295
SP = Spelling
Appendix B

Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
via via
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

296
 Appendix B

Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores (continued)

Ages 16:0-19:11 (High School)

Subtest Standard Scores Composite Standard Scores

Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total se

-aS
SS RAN
=|O
ODU
120
121
122
123
124
125
126

160 | i ee : : rs
scores and the WAIS-II| VIO across the four age ranges.
Note. Based on the correlations between the WIAT-II standard are .72, .70, .72, .65, .60,
subtests
Presented in the order shown in the table, the average correlations for the nine WIAT-II
composites are .78, .75, .65, .77, and .83.
73, .57, .81, and .44, and the average correlations for the five WIAT-I|
PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language
WRD = Word Reading
MR = Math Reasoning RD = Reading
NO = Numerical Operations
WE = Written Expression MA = Mathematics
RC = Reading Comprehension 297
LC = Listening Comprehension WL = Written Language
SP = Spelling
Appendix B

Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores (continued)
College Students
Subtest Standard Scores Composite Standard Scores
via via
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

298
 Appendix B

Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores (continued)
esses

College Students
Subtest Standard Scores Composite Standard Scores
via
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total a,

and the WAIS-II| VIO. Presented in the order shown in

Note. Based on the correlations between the WIAT-II standard scores
for the nine WIAT-II subtests are 52, .50, .55, .45, .47, .62, .38, .66, and .34, and the correlations
the table, the correlations
for the five WIAT-II composites are .66, .64, .44, .61, and .68.
PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language
WRD = Word Reading
Numerical Operations MR = Math Reasoning RD = Reading
NO =
WE = Written Expression MA = Mathematics
AC = Reading Comprehension 299
LC = Listening Comprehension WL = Written Language
SP = Spelling
Appendix B

Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
via via
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

300
 Appendix B
Table B.11. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
via via
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
101 101 101 101 101 101 101 101 100
102 101 101
102 102 102
103 102 103
104 103

ee
dard scores and the WAIS-II| VIO.
— standar aa
Presented in the order
en the WIAT-II

PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD = Word Reading
Numerical Operations MR = Math Reasoning RD = Reading —
NO =
RC = Reading Comprehension WE EGY= Writtenchor
Expression ;
MA== oe :a 301
: LC = Listening Comprehension WL = Written Languag
SP = Spelling
Appendix B

Table B.12. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III PIQ Scores
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
PIO PIO
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

302
 Appendix B
Table B.12. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III PIQ Scores (continued)
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
PIO PIO.
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
100 101
101 101
101 102
102
102

= Word Reading PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 303
SP = Spelling
Appendix B

Table B.12. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III PIQ Scores (continued)

College Students
Subtest Standard Scores Composite Standard Scores
PIO PIO
Score|WRD NO RC SP PD MR WE LC’ OF RD MA WL OL Total |Score

304
 Appendix B
Table B.12. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III PIQ Scores (continued)
College Students
Subtest Standard Scores Composite Standard Scores
PIO

WAIS-II| PIO. Presented in the order shown in

Note. Based on the correlations between the WIAT-II standard scores and the
.29, .59, .38, .48, and .39, and the correlations
the table, the correlations for the nine WIAT-II subtests are .32, .55, 45, 30,
for the five WIAT-II composites are .42, .63, .39, .51, and .55.
PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language
WRD = Word Reading
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
= Listening Comprehension WL = Written Language 305
SP = Spelling LC
Appendix B

Table B.12. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III PIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
PIO PIO
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

306
 Appendix B

~ Table B.12. WIAT-II Subtest and Composi

posite Standard Scores Predicted From
WAIS-III PIQ Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores

Score}WRD NO RC sP PD MR WE LC OE RD MA WL OL Total pe

160 _ [iGO rear

in
scores and the WAIS-lII PIQ. Presented in the order shown
Note. Based on the correlations between the WIAT-II standard
the correlatio ns for the nine WIAT-II subtests are .57, .67, .63, .51, .67, .65, .54, .64, and .44, and the correlations
the table,
for the five WIAT-II composites are .77, .68, .51, .57, and .72.
OE = Oral Expression OL = Oral Language
WRD = Word Reading PD = Pseudoword Decoding
MR = Math Reasoning RD = Reading
NO = Numerical Operations
WE = Written Expression MA = Mathematics
RC = Reading Comprehension 307
LC = Listening Comprehension WL = Written Language
SP= Spelling
Appendix B

Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
SO
WAIS-III VCI Scores
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
vcl vc
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

63 67 70 63 62
64 68 71 64 63

308
 Appendix B
Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VCI Scores (continued)
———

Ages 16:0-19:11 (High School)

™ Subtest Standard Scores Composite Standard Scores

Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total ae

101 101 101 101 101 101 101 101 100
101 101 101 102
102 102 102 102
102 102 103

160 [eee
the WAIS-III VCI across the four age ranges.
Note. Based on the correlations between the WIAT-II standard scores and
in the order shown in the table, the average correlations for the nine WIAT-II subtests are .70, .63, .70, .59, .54,
Presented
and .42, and the average correlations for the five WIAT-II composites are .74, .65, .60, .74, and .75.
62, .52, .80,

PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD = Word Reading
MR = Math Reasoning RD = Reading
NO = Numerical Operations
WE = Written Expression MA = Mathematics
RC = Reading Comprehension 309
LC = Listening Comprehension WL = Written Language
SP = Spelling
Appendix B

Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VCI Scores (continued)
College Students
Subtest Standard Scores Composite Standard Scores
vel vel
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

310
 Appendix B

Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VCI Scores (continued)
College Students
— Subtest Standard Scores Composite Standard Scores

Score |WRD NO RC sp PD MR WE LC OE RD MA WL OL Total aes

100 100 101 100 100 101 100 101 100
101 101
101 102
101 103
103

scores and the WAIS- II VCI. Presented in the order shown in

Note. Based on the correlations between the WIAT-II standard
42, .52, .40, .42, .51, 35, .64, and .27, and the correlations
the table, the correlations for the nine WIAT—II subtests are 47,
for the five WIAT-II composites are .60, .53, .40, .55, and .61.
PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language
WRD = Word Reading
MR = Math Reasoning RD = Reading
NO = Numerical Operations
WE = Written Expression MA = Mathematics
RC = Reading Comprehension 311
LC = Listening Comprehension WL = Written Language
SP = Spelling
Appendix B

Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VCI Scores (continued)

Adults
Subtest Standard Scores Composite Standard Scores
vel vel
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

312
 Appendix B
Table B.13. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III VCI Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
vcl
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total rhe
101 101 101 101 101 101 101 101 100
101 102 101 101 101 101
102 102 102
103 102 103
104 103 103

160 Seo aealaan

in the order shown in
Note. Based on the correlations between the WIAT-II standard scores and the WAIS-III VCI. Presented
.45, and the correlations
the table, the correlations for the nine WIAT-II subtests are .61, .76, 72, .64, .55, .74, .60, .70, and
for the five WIAT-II composites are .81, .76, .61, .65, and .82.

PD = Pseudoword Decoding OE = Oral Expression - OL = Oral Language

WRD = Word Reading
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 313
SP = Spelling
Appendix B

Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores
Ages 16:0-19:11 (High School)
Subtest Standard Scores Composite Standard Scores
PO! PO!
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
40 40
a 4
42 42
43 43
44 44

314
 Appendix B

Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores (continued)
Ages 16:0-19:11 (High School)

| SSS ae
Subtest Standard Scores Composite Standard Scores

Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total con

160
and the WAIS-III PO! across the four age ranges.
Note. Based on the correlations between the WIAT— Il standard scores
shown in the table, the average correlation s for the nine WIAT— ll subtests are .38, .70, .33, .47, .37,
Presented in the order
the average correlation s for the five WIAT-II composite s are 45, .80, .52, .70, and .77.
79, .44, .50, and .67, and
OE = Oral Expression OL = Oral Language
WRD = Word Reading PD = Pseudoword Decoding
Operations MR = Math Reasoning RD = Reading
NO = Numerical
WE = Written Expression MA = Mathematics
RC = Reading Comprehension 315
LC = Listening Comprehen sion WL = Written Language
SP = Spelling
Appendix B

Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores (continued)

College Students
Subtest Standard Scores Composite Standard Scores
POI POl
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
40 40
41 41
42 42
43 43
44 44

316
 Appendix B
Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores (continued)
(ect it adalat al lll
College Students
Subtest Standard Scores Composite Standard Scores
PO! POI
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score
100 101 100 = 100 100 101 100 100 100 101
101 101 102
101 103
102 104
102 105
106
107
108
109
110
111
112
118
114
115
116
117
118
11
120
121
122
123
124

PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD = Word Reading
NO = Numerical Operations MR = Math Reasoning RD = Reading _
RC = Reading Comprehension WE = Written Expression MA = Mathematics
LC = Listening Comprehension WL = Written Language 317
SP= Spelling
Appendix B

Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
POl POl
Score}WRD NO RC SP PD MR WE LC OE RD MA WL OL Total |Score

318
 Appendix B
- Table B.14. WIAT-II Subtest and Composite Standard Scores Predicted From
WAIS-III POI Scores (continued)
Adults
Subtest Standard Scores Composite Standard Scores
POl
Score |WRD NO RC SP PD MR WE LC OE RD MA WL OL Total tad
100. 101 101 100 ~=—-:101 101 100 101 100
101 101
101 102
102

160 [Ree Nee

POI. Presented in the order shown in
Note. Based on the correlations between the W IAT-II standard scores and the WAIS-II|
.39, .61, .58, .47, .57, an d .41, and the correlations
the table, the correlations for the nine WIAT-II subtests are 50, .60, .56,
for the five WIAT-II composites are .68, .60, .43, .51, and .64.

PD = Pseudoword Decoding OE = Oral Expression OL = Oral Language

WRD = Word Reading
NO = Numerical Operations MR = Math Reasoning RD = Reading
RC = Reading Comprehension WE = Written Expression MA = Mathematics
WL = Written Language 319
SP = Spelling LC = Listening Comprehension
Appendix B

Table B.15. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Required for Statistical
Significance With Prediction Based on WAIS-III Scores
Ages 16:0-19:11 (High School)
Subtests p FSIO vio PIO. vcl PO!
Word 05 7.46 7.81 Weais 8.29 UA
Reading 01 9.82 10.28 9.93 10.92 9.49
Numerical 05 8.14 8.30 8.70 8.51 9.02
Operations 01 10.71 10.93 tHEAS Ane20 11.87
Reading 05 8.50 8.87 8.41 9.26 8.18
Comprehension 01 me) 11.67 da). OY 12419 10.77
Spelling 05 9.46 9.60 9.71 9.77 9.54
01 12.45 12.64 12.79 12.87 25
Pseudoword .05 6.59 6.87 6.76 7.05 6.53
Decoding 01 8.68 9.04 8.90 9.28 8.60
Math 05 8.79 8.88 9.56 8.98 9:92
Reasoning 01 e577 11.69 (Pete, 11.82 13.06
Written 05 ow iene e25 qs eS)
Expression 01 14.50 14.64 14.81 14.80 14.67
Listening 105 ee 13.98 1Se72 14.34 13.70
Comprehension 01 18.05 18.40 18.05 18.88 18.04
Oral 05 11.66 11.68 11.97 11.78 12253)
Expression 01 ec" 15.38 Los 1S 5 16.49

Composites
Reading 05 5.61 6.18 5.65 6.73 5.42
01 7.38 8413 7.43 8.86 TKS
Mathematics 05 7.19 7.34 8.09 7.53 8.58
107 9.47 9.66 10.65 9.92 11.30
Written 105 8.50 8.68 8.88 8.89 8375
Language 01 (APS 11.42 11.69 eval ley?
Oral .05 (les 52 11.69 11.86 11.93
Language 01 14.89 Tei IY/ 15:39 15.61 Sea
Total .05 6.02 6.39 6.99 6.83 7.28
01 WAo2 8.42 9.20 8.99 9.59
Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

320
 Appendix B
Table B.15. Differences Between Predicted and Actual WIAT_II
Subtest and Composite Standard Scores Required
for Statistical Significance With Prediction Based on
WAIS-III Scores (continued)
College Students
Subtests p FSIO via PIO vcl POI
Word .05 6.21 6.46 6.30 6.51 6.30
Reading (017 8.17 8.50 8.30 8.57 8.29
Numerical 05 6.99 7.05 7.68 7.02 WXe}s)
Operations 01 9.20 9.28 10.11 9.24 10.06
Reading .05 6.32 6.52 6.71 6.64 6.70
Comprehension 01 B32 8.58 8.84 ish7/6) 8.82
rs .05 7.42 7.56 7.53 7.58 725i]
Spelling
01 9.76 9°95. 9.91 9.98 9.88
Pseudoword 05 6.17 6.36 6.24 6.37 6.17
Decoding 01 Ste be 8.37 8.21 8.38 8.12
Math els 8.77 8.90 9.35 8.83 ADI
Reasoning 01 Takeo ier AS 11.63 Wisp
Written .05 14.20 14.23 14.36 14.24 14.37
Expression 01 18.70 18.73 18.90 18.75 18.91
Listening :05, 10.93 ie HS 1asl5 ls) ais
Comprehension 01 14.39 14.65 14.68 14.80 14.65
Oral 05 14.79 14.80 14.97 14.79 15.00
Expression 01 19.47 19.48 19.70 19.46 iS )AS)

Composites
: LO 4.92 5.36 Sy) 5.46 5.05
Reading 01 6.47 7.05 6.78 7.19 6.65
f 05 6.56 6.71 AZ 6.64 7.48
Mathematics 01 8 63 8.84 9.77 8.74 9.85
Written .05 8.54 8.62 8.77 8.64 8.76
Language .01 11.24 11.34 AiO: (Sy loos
Oral .05 10.51 10.65 10.81 10.68 10.86
Language 01 13.83 14.02 14.24 14.06 14.30
505 5.09 5.42 5.76 5.48 5.87
Ue 01 6.70 7.14 7.58 724 7.72
Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

321
Appendix B

Table B.15. Differences Between Predicted and Actual WIAT-II

Subtest and Composite Standard Scores Required
for Statistical Significance With Prediction Based on
WAIS-III Scores (continued)
Adults
Subtests p FSIO via. PIO. vcl PO
Word 05 6.49 6.67 Wel’; 6.90 7.03
Reading 01 8.54 8.78 9.46 9.08 9.26
Numerical .05 7.40 Ti 2 eA 7.94 8.06
Operations 01 S75) 10.16 10.75 10.45 10.61
Reading 05 6.67 6.97 7.44 7.26 7.30
Comprehension 01 8.78 Gilg. 9.79 9.56 9.61
Spalling 05 UeiPh 7.96 8.09 8.13 7.81
01 ONS 10.48 10.65 10.70 10:29
Pseudoword .05 6.54 6.68 Peas) 6.71 P50
Decoding 01 8.61 8.79 9.99 8.84 9.94
Math .05 8.93 ml LOS 9.38 9.46
Reasoning .01 ilen6 12.07 ey) 12235 12.45
Written 05 14.33 14.42 14.62 14.53 14.57
Expression 01 18.86 18.98 19.24 (9213 ISLA
Listening 05 11.03 EAS 11.56 ihileeye 11.48
Comprehension .01 14.51 14.69 ieee 14.97 A5ela
Oral 05 14.84 14.88 ~ 15.03 14.93 TS05
Expression 01 iis 19.59 19.79 19.66 19.81

Composites

Reading 05 5.49 5.97 6.93 6.32 6.71

01 Wee 7.86 OA2 8.32 8.83
NMathantatte .05 748) iene! 7.63 7.39 7.49
01 8.94 9.40 10.04 OMS 9.86
Written 05 8.71 8.89 Si 9.05 8.97
Language .01 11.47 ZO 11.98 11291 11.80
Oral .05 10.56 10.70 10.97 10.88 10.92
Language .01 13.90 14.09 14.44 14.32 14.38
Total f05 5.47 6.01 6.62 6.35 6.46
01 7.20 7.91 8.71 8.36 8.51
Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

322
 Appendix B

Table B.16. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III FSIQ Scores
High School Percentage
Subtests 25 20 15 10 5 4 3 2 1
Word Reading a 9 11 14 18 19 20 22 25
Numerical Operations 7 8 10 13 17 18 KS) 21 23
Reading Comprehension ¥j 9 11 14 18 19 21 23 26
Spelling 8 S 382 4. FIB 819° Fiuuee@ee 26
Pseudoword Decoding 8 10 13 AS 20 21 23 BS 28
Math Reasoning 6 8 =) 11 15 16 17 18 21
Written Expression 8 0
Listening Comprehension 7
Oral Expression 9

Composites

Reading 11 ihe iW 18 19 7 24
Mathematics 2 a 14 15 16 18 20
Written Language pa ps mdpS 18 ite. 21 23 26
Oral Language 10 12 16 17 18 19 22
Total N!|O|N|]
alan! QO
SN] 8
|]O}]N 10 13 14 15 16 18

Note. Percentage of individuals whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

323
Appendix B

Table B.16. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III FSIQ Scores (continued)
College Students Percentage
Subtests 25 20 15 #410 5 4 3 2 1

Word Reading g 11 14 17 22. 23 25 Di 31

Numerical Operations 8 10 13 16 20 A 23 25 29
Reading Comprehension 8 10 13 16 20 22 23 25 AS)
Spelling i) 11 14 Wa 22 24 25 28 31
Pseudoword Decoding 9 it 14 AZ, 22 23 25 27 31
Math Reasoning 8 9 12 14 18 19 Di 23 26
Written Expression §) 11 14 17 De 24 IS, 28 32
Listening Comprehension 8 10 {zz 15 19 20 21 23 27
Oral Expression 9 11 14 17 22 24 26 28 32

Composites

Reading 10 2 5 i. 20 22 24 DT
Mathematics 9 11 14 18 19 20 22 25
Written Language 11 14 17 22 23 25 27 31
Oral Language 10 it 15. 43.— -20 22 24 27
Total 10d
|)
NCO)
SOO g 11 14 17 ike, 20 22 2S

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.
 Appendix B

Table B.16. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III FSIQ Scores (continued)

Adults Percentage
Subtests 25 20 15 10 5 4 3 2 1
Word Reading 8 9 12 14 19 20 21 23 26
Numerical Operations 6 fi 9 11 14 15 16 18 20
Reading Comprehension 7 8 10 13 16 a 18 20 23
Spelling 8 9 2 14 18 20 2A 23 26
Pseudoword Decoding % 9 11 14 18 19 20 22 25
Math Reasoning 6 8 10 12 15 16 Ali 19 22
Written Expression 8 10 ie 15 19 21 22 24 27
Listening Comprehension 7 9 11 13 13 18 19 21 24
Oral Expression ) 11 13 17 21 23 24 27 30

Composites
Reading 5 6 8 10 (2 18) 14 5 18
Mathematics 6 7 9 11 14 ie 16 18 20
Written Language 8 10 2 ifs 19 20 22 24 OT
Oral Language 8 ) 12 14 18 19 21 23 26
Total 5 ] 8 10 (Ke) 14 15 16 18

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

32
Appendix B

Table B.17. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VIQ Scores

High School Percentage

Subtests 25>—420, «15: #10 5 4 S 2 1

Word Reading 7 9 11 ii Ag. 18 20 21 24

Numerical Operations J 9 11 14 18 19 20 22 BS
Reading Comprehension 7 ¢) 11 ik 19 18 19 21 24
Spelling 8 10 2. IS 19 20 on 23 27
Pseudoword Decoding 8 10 12 iS 20 24 22 24 28
Math Reasoning i € 11 13 17 18 19 21 24
Written Expression 8 10 iS 16 20 21 2S PLS) 29
Listening Comprehension 6 7 $s) 11 14 AS 17 18 72)
Oral Expression 9 11 14 17 22 Ze 2S 27 31

Composites

Reading 8 10 12 16 17 18 19 22
Mathematics 8 10 13 16 (We 19 20 23
Written Language 9 12 14 19 20 21 23 26
Oral Language 8 10) 22. BIG. 17) SASreePOdRet29
5
Total Oo,
O}|HD}O};}N/]Q Site B49" 15 16 17 +20

Note. Percentage of individuals whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

2)26
 Appendix B

Table B.17. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VIQ Scores (continued)

College Students Percentage

Subtests 25 720 815 G10 5 4 3 2 1

Word Reading 11 13 16 21 22 24 26 30
Numerical Operations 11 13 16 21 23 24 26 30
Reading Comprehension 10 13 16 20 22 23 25 2g
Spelling 11 14 iW 22 23 25 27 31
Pseudoword Decoding as 14 iW 22 23 25 ay, 31
Math Reasoning 10 ile 15 19 20 22 24 27
Written Expression 11 14 18 23 24 26 28 32
Listening Comprehension 9 iZ 14 18 20 Zi 23 26
Oral Expression Ww 14 18 23
O}DMDI/O/WMDIO}/O};]Mso;o 24 26 US 38)

Composites

Reading 8 9 12 14 18 20 21 23 26
Mathematics 8 10 ie ike; IS) 20 22 24 Bi

Written Language a 11 14 17 22 23 25 OF, 31

Oral Language 8 10 12 5 20 21 22 24 28
Total “4 o al 14 18 19 21 22 25

was below their

Note. Percentage of examinees whose obtained achievement standard score
predicted-achievement score by the specified amount or more.

327
Appendix B

Table B.17. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VIQ Scores (continued)

Adults Percentage
Subtests 25 420 45 10 @ 4 Weel 1
Word Reading 8 10 WP 15) 19 2 22 24 Bai:
Numerical Operations 6 8 10 2 15 16 17 19 DH
Reading Comprehension 7 9 11 13 17 18 19 oF 24
Spelling 8 9g (2 14 18 20 21 23 26
Pseudoword Decoding 8 10 12 1G 9 20 2, 24 DY.
Math Reasoning 7 8 10 iz 16 184 18 20 723}
Written Expression 8 10 2 13 20 2A 23 25 28
Listening Comprehension 7 9 11 14 18 19 21 22 25
Oral Expression 9 11 14 17 22 23, 25 Di 31

Composites

Reading 5 7 8 10 13 14 15 17 19
Mathematics 6 8 10 2 NS, 16 (7 19 2
Written Language 8 10 (WW (en ils, 2 22 24 DF
Oral Language 8 10 12 15 WS 20 21 23 27
Total 5 Z 8 10 13 14 1S 16 18

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.
 Appendix B

Table B.18. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III PIQ Scores
High School Percentage
Subtests 25 20 15 10 5 4 3 2 1
Word Reading 2 11 14 17 22 23 25 27 31
Numerical Operations 8 10 12 15 19 20 22. 24 yf,
Reading Comprehension 9 11 14 ied 22 24 26 28 32
Spelling 9 11 13 16 21 22 24 26 30
Pseudoword Decoding 9 11 14 WA 22 24 25 28 31
Math Reasoning 7 9 11 13 17 18 20 21 24
Written Expression 3 11 13 i\G4 2 23 24 27 30
Listening Comprehension 9 11 13 16 21 23 24 26 30
Oral Expression 9 11 14 ‘WA 22 a) 25 27 Sil

Composites

Reading 11 13 17 21 23 DAT, 30
Mathematics 9 11 13 17 18 20 Zit 24
Written Language ~_oO (ee) 16 21 22 23 26 29
Oral Language 10 12 tf) 19 20 22 24 27
Total NW S,
]}O}@O}]N/011 13 i 18 We) 21 24

Note. Percentage of individuals whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

329
Appendix B

Table B.18. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III PIQ Scores (continued)

College Students Percentage

Subtests 25 20 #15 + °&410 5 4 3 2 1

Word Reading 9 WZ 15 18 23 25 27 29 33
Numerical Operations 8 10 13 16 21 22 23 26 29
Reading Comprehension 9 11 14 17 22 23 25 27 31
Spelling 10 12 nS 18 23 25 Zh 29 33
Pseudoword Decoding 10 12 15 18 23 25 Dif, 29 33
Math Reasoning 8 10 12 15 20 21 23 gus 28
Written Expression 9 11 14 18 23 24 26 28 Su
Listening Comprehension 9 ti 14 17 22 23 BS 27 30
Oral Expression 9 11 14 WE 23 24 26 28 32

Composites

Reading FS) 11 14 17 22 24 DS 28 31
Mathematics 8 10 12 15 ie) 20 22 24 27
Written Language 9 @11 914 « Oh 20306 24 seven vor
Oral Language 9 11 13 16 21 23 24 26 30
Total 8 10 13 1G oee2O 22 23 25 29

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

& 30
 Appendix B

Table B.18. Differences Between Predicted and Actual WIAT_II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III PIQ Scores (continued)
Adults Percentage
25.
Le @0 Subtests
#15. éf0 5 4 3 2 1
Word Reading 8 10 13 16 20 21 8) 25 29
Numerical Operations 7 9 11 14 18 19 21 23 26
Reading Comprehension 8 10 12 15 19 20 22 24 27
Spelling 9 11 13 16 21 22 24 26 30
Pseudoword Decoding 8 9 2 14 18 20 21 28 26
Math Reasoning 8 9 TZ 14 19 20 21 23 26
Written Expression 8 11 13 16 a) Ze 24 26 29
Listening Comprehension 8 10 12 15 19 20 22 24 27
Oral Expression 9 11 14 17 22 23 25 27 31

Composites

Reading 6 8 10 2 16 17 18 20 22
Mathematics 7 i) 11 14 18 19 21 23 26
Written Language g 11 s\e5} 16 21 22 24 26 30
Oral Language 8 10 13 16 20 22 23 25 29
Total 7, 9 11 13 i 18 20 21 24

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

331
Appendix B

Table B.19. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VCI Scores

High School Percentage

Subtests 25, #20, #15 - #10 5 4 3 2 1

Word Reading 2 11 14 18 ie) 20 22 ZS

Numerical Operations 10 12 15 tg 20 22 24 27
Reading Comprehension S, ili) 14 18 19 20 22 25
Spelling 10 is 15 20 2a 23 25 28
Pseudoword Decoding = S) =w 16 21 22 24 26 we
Math Reasoning 10 12 15 19 20 22 24 2a,
Written Expression 11 13 16 21 Ze. 24 26 30
Listening Comprehension 8 g 12 15 16 iz 19 21
Oral Expression S|
SSI)
(oo
Koo)
||
oo)
(de)
|
ep)
(de) 11 14 17 22 24 25 28 32

Composites

Reading g 11 13 (e/ 18 19 21 24
Mathematics 8) 12 14 19 20 21 23 26
Written Language —v (jo) 12 OS 20) 21 22 25 28
Oral Language 8 la! 13 17 18 19 21 24
Total teen
os)
SS
|
SS 8 10 is: 16 Ws 19 20 23

Note. Percentage of individuals whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.
 Appendix B

Table B.19. Differences Between Predicted and Actual WIAT_II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VCI Scores (continued)

College Students Percentage

Subtests 25 20 15 «10 5 4 3 2 1
Word Reading 9 11 14 17 22 23 25 27 S11
Numerical Operations 9 11 14 V7 22 24 25 28 32
Reading Comprehension 9 11 13 16 21 22 24 26 30
Spelling 9 11 14 tz 22 24 26 28 32
Pseudoword Decoding 9 11 14 17 22 24 25 28 B32
Math Reasoning 9 11 13 16 21 22 24 26 30
Written Expression g 12 14 18 23 24 26 29 33
Listening Comprehension 8 10 12 15 19 20 22 24 aT}
Oral Expression 10 12 15 18 PAS 25 ih ZOE ECS

Composites

Reading 8 10 |be LS 20 21 22 24 28
Mathematics S) 11 ne 16 21 22 24 26 30
Written Language 3) 11 14 17 23 24 26 28 32
Oral Language 8 10 ile 16 21 22 23 26 29
Total 8 10 12 15 20 21 PL) 24 28
_—
———_—————___ainaaniiiannnnEERIREREREERERREREERIEE
— EE aiaaaeme

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

333
Appendix B

Table B.19. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III VCI Scores (continued)

Adults Percentage
Subtests 25 120 715° “10 5 4 3 2 1

Word Reading 10 12 15 19 2A 22 24 eT &

Numerical Operations 8 10 13 16 17 18 20 23
Reading Comprehension
Spelling
Pseudoword Decoding a (iS) = wo

Math Reasoning
Written Expression
Listening Comprehension 18 19 20 22 2S
Oral Expression 1a
SSI)
SI
OO!
CON
SSI
|
OOn
COO 22 23 25 27 31

Composites

Reading x 9 11 14 LS 16 18 21
Mathematics 8 10 12 16 17 18 20 23
Written Language 10 22 5 20 2 22 24 28
Oral Language 10 12 ills) 19 20 21 23 27
Total 7
|o};oO};}N/O
© 9 11 14 AS 16 18 20

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

i)34
 Appendix B

Table B.20. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III POI Scores

High School Percentage

Subtests z= 6200695 = 5 4 3 2 1
Word Reading 9 11 14 18 23 24 26 28 62
Numerical Operations Wi 9 ili 14 18 19 20 22 25
Reading Comprehension 9 2 3) 18 23 25 26 PL) 83
Spelling 9 11 14 V7 22 23 25 DH; 31
Pseudoword Decoding 9 ie 14 18 2S 24 26 28 32
Math Reasoning 6 8 10 12 i 16 17 fs) 21
Written Expression 9 11 14 AZ Z2 2s 25 Dy (|
Listening Comprehension ) 11 13 Ue, 21 23 24 27 30
Oral Expression 7 dS) 18) 14 18 19 21 23 26

Composites

Reading 9 11 14 ey. 22 23 25 ei, 31

Mathematics 6 8 9 11 5 16 17 18 21
Written Language g 11 13 16 24 22 24 26 30
Oral Language 7 9 11 14 18 19 20 22 25
Total 7. 8 10 12 16 ez 13 eZ Oe 22

Note. Percentage of individuals whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.
Appendix B

Table B.20. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III POI Scores (continued)

College Students Percentage

Subtests 25 #20. HS “go 5 4 3 2 1
Word Reading 10 12 15 18 ZS 25 27 29 3)
Numerical Operations g 17 13 16 21 ae 24 26 30
Reading Comprehension ©) 11 14 17 22 24 25 28 32
Spelling 10 12 15 18 24 25 27 29 oo
Pseudoword Decoding 10 12 15 18 24 25 27 30 34
Math Reasoning 8 10 12 iS 20 21 ZS 25 28
Written Expression ¢
Listening Comprehension 3 11 14 17 22 23 25 27 31
Oral Expression 9

Composites

Reading 2)
Mathematics 8
Written Language 9g 12 14 ls e23 24 26 29 32
Oral Language 9
Total 9

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.
 Appendix B

Table B.20. Differences Between Predicted and Actual WIAT-II Subtest

and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample With
Prediction Based on WAIS-III POI Scores (continued)
Adults Percentage
25
LS 20 15 Subtests
10 5 4 3 2 1
Word Reading 9 11 13 17 21 DS) 24 27 30
Numerical Operations 8 10 12 15 20 21 22 24 28
Reading Comprehension 8 10 13 16 20 22 23 YAS Ze)
Spelling S| 11 14 i 23 24 26 28 32
Pseudoword Decoding 8 10 12 a5 20 21 22 24 28
Math Reasoning 8 10 13 16 20 21 23 25 28
Written Expression 9 | 14 17 22 23 DNS 27 31
Listening Comprehension 8 10 Is 16 20 22 23 25 29
Oral Expression 9 bt 14 17 22 24 25 28 32

Composites

Reading z 9 11 14 18 19 2 23 26
Mathematics 8 10 12 15 20 21 23 25 28
Written Language 9 11 14 clyé De) 24 25 28 31
Oral Language 9 11 is 16 21 22 24 26 30
Total 8 10 12 1 19 202 Zhe,

Note. Percentage of examinees whose obtained achievement standard score was below their
predicted-achievement score by the specified amount or more.

337
 APPENDIX C

Discrepancy Score Tables Based on

Simple-Difference Method

339
 S Sit Gv 061 Syl eZt O'S| v9t 091 Or! OSI OSI vel eZ SiS LO" Mowe
Sh Ol vl Ob Zell ALL Sek Vel “SOL ty tie) ec Opec cece SO" BupoM
Col eo tke Sel vel Och "PS “Sel Gul OCP cc ie cca) eo es LO" Kowey\
€'6 88 86 ZOL = 86 rl0)a OM) 88 86 €6 £6 StL {8c 2866 SO" je1auey
Lie 98k Gl) «OS 68 0e )«6C ke «© ble 6 S06 6 es OSI 0 Glee cr cemeticc me ie LO" pehelaq
ser* eb. .2yk Ol OS QOL G9L 9G, 99L 9 Lyi ery) Sy Sieuee 2peeeeer SO’ uoluBooey
Kioypny
LO"
cel
O81
09M
ool
eOc
=pakeleq
GO,
OSL
OS)
LZ)
L£2ZL
ole
SAIS
emo
GCL
SEL
6%
/ivl
liebe
Lol”
lensi,
SO"
eorcr
olen
celtic
\Scle
VIL
PIL
iaeero
SYie00)
~G¥L»
PEt
OSL
O9L
SSL
OS
Syl
vel
Por”
pehelaq
LO"
or
Ocemeny
Oper
see,
GoL=
wi
Pel
8th
JIL
OL
2OLe
ic
Lee
U0
Aoupny
SO’
eee
oc
ce
50
ee
PEL Olls Pel Ovi Sek Sek Ol CCL OTS CcheuS clemey co Uoce 10" Mowery
COLme oe cok 90 66 86 €8 €'6 88 €'6 86 cOk Ech 26 SO" ayelpeuity|
69 org O6GE QL OG LZ) GSE O9L “OOP 9790 GOlmnCoiee? perce LO” 9ye|Pewity|
SclL..Sé. vel rl vel GSEl OF Vel CPR SS ciMO chun) emo cl memincT SO" yensi,
coke cOk OL Sek OFL Gl OL OL) 820s SOU Recicleeee cic eee ecere 10° a}eIPewWw|
€6 gL €'8 €6 €8 8'8 €'8 8'8 gL €'8 €6 €'6 90 88 SO’ Aoupny
jeneq
sebvily
xepuj
68-S8
S-Gh
9-SS
69-S9
pl-0L
62-SL
78-08
2-02
62-SZ
PE-OE
pb-SE
LI-9L
G6I-BL
Appendix C

a ~ es dnoiy aby ueay eouedyiubis HI-SIWM

340
(Pompe
edUeIajJIG-2[durl¢) Aq easy
[EINSNeIS sUeITGIUSIS
pormbay
1oy sei09g
Puke [TI-SINM
xepuy sa109¢ seouetayIq
Ueemjog II-SIVM
OISA ajgel*1°9
 ajgey"Z"9
+ S9dUSIIFFIG
UsaMJag TI-SIVM
OIA $2109g
Puke [II-SIM
xapuy sai10dg parmbay
10J [esNsNeI¢g
souvIyTUsIS
Ag a8y suUaIajjJIG-a]du
(poppaWrtg)

xapu|
dnoig aby
[eAe7
II-SWM uray eoursiyiubis
seBvily I-91 G6I-Bl 2-02 62-SZ E-0E bb-SE S-Sh 9-SS 69-S9 pl-OL 6/-SL 8-08 68-S8
Kioypny SO’ L'6 Ol €6 86 88 €'8 £6 88 88 es 86 eg €'8 86
ayelpewiuy| 10° Orc cleeC
cic oes ORR Gck VSR SYR OL) Och OLL Ol eck
rensi, GO’ eerclmerien
Wyle “er eh” Sil" tel ok Se eerie Pek Fel GZt- CEL
9}eIPSWW| LO" econo, Gem G/L Gol #9 POL OSL ZZL O6L OG O6L 691 eZ:
ayelpewity| GO’ COL Feel” Bok cok. 86 £6 8'6 88 8'6 eG sO ZO £6 90l
Mowe, LO" eecer wea wow Fel Sel “CEL SZ OU Oh ct Stl vet gd Ort
Kioypny gO’ ce eer Sire Sear Gel PTL PvE 9OL SOL ck are Ott OFF.. .90. Sci
pekejeq LO" "eer "Gee WoL PvOL OSL OSL Ol Sot ‘OOL Got Sct Svl Ov! 669i
fensi\ SO" minor > my sucicin Som GEL Sol Oils @F ger. Ger pe Lv Sz Gzt
pekejaq 10° eeeBee Fob FRE 2 SSt OGL LHL ZZ oO8f oO8f OL yo!
Aoypny
uomubooey SO’ eee! PeresFRE vel GOL SL GOL OL OSt 99 Lert vet Sse.
pekejaq LO" ee eo ee CP LC SC‘C CS SCC—CtCOSCCGG SOS. COIS GL )6OC 66 cet
je49Ua5 SO’ Sor BOGE) etl. 86 86 ZOky €6 S'6Os ZOlL 86 90l 86 €6 8'6
Mowery LO" mEECien mCy eco Gat Sigh “PSl gel Sale Pel cl OF ct gel scl
Bupyom SO" eee eee ee| Se Sh 6@ctl6U6USZL.)hCUOIE
vil wet. vik ve vt Ec
Kiows, LO" emmGicE ce Ghmmtr
G Gh “SENS VoL GOL OSI 21 OS OGL OSE OOl

341
Appendix C
 691 ¥9l God O91 Gel FOL “Set 2 yO ye Oe! Deen cic OG nce LO’ Mowe
CCl Giclee oic). erie cre nCiT 4qisyk ih Gi aiglk ery Wig bel gO’ BuDOM
OTL OF OSE “OSE OPT OS Sir 0 re cis e n Ctr ent) nn 9 lr EE LO" Mowe
SHO eH a al CRO AWE ORME SHOE RIL SL ILA SL ia ae Ll SO’ je1aues
O61 L0¢c 90¢ 9S'c¢ Ole Geo ~—Sisd Gilce Gicc iG SOG tzSCummotccunOice LO’ pekejaq
vel ©St esl ae ey va aL HSN GHA. SL EGIL ILL Lgl eZs9}- SO’ uoubooeY
Aioypny
Sy eit yA Ae SCRE OE Gee Gell Cal Cet PRL GGL G04 sah LO" pehejeq
Gol ah A eal rah azide ~ Ciel eral ial ae ie Lidl ePsl vk G0" ensi
LO”
Gs)
TI,
0:9
Gicive
“Gels
Go
SZ
“FOL
FOL
9!
ohemetc:o
commnrer
Onl
pekejaq
|
bec
mn
ay
Ch
wel
Gul
Geb
el
“Ge.
Gal
Gik
Goal
cies
Sek
Aioypny
GO’
Oer
Oh
2
er
OSL OFT “GG “Gig Fite VS e PiS Cire Git Penn, Ch) es a LO” Mowe
ayeIPewWu|
SO’
Guy
Gel
hh
Gu
COL
Om
OL
Aol
Oy
AO
Ce
CR
ON
Al
sl Tah es A AG OL Bb yak eel £2 OG L0¢ vé6lL get LO" ayeIPewWU|
Sel Bel OSL OSE OSL FP ecici SiC ee ic PetCiC enn at, Denne SC) eat eZ G0’ jensi,
OrL Sch PEL OF ck Vrs PFvicle Oiriee ic en Gic een OiS eee Crt | eeenO|S aun LO” ayeIPewU|
90l 986 20, GOL GE COL GOL COL Ol BE 7A) Oaee Zk 46GOL gO’ Asoupny
[ene7
sebvily
xepu|
6/-SZ
8-08
68-S8
62-SZ
~pE-0E
br-SE
G-Sh
¥9-SG
69-S9
fl-0L
ZI-91
61-81
2-02
Appendix C

dnoiy aby ueay eouedy!uBIS II-SWM

Ag asy
sdUeITUSIS (poyjey|
FUaAIIJIG-2[duIs)

342
Jeonsnet 10;
poambeay SadUaIAIG Old
WaeMIeg III-SIVM Pue
$2109S xapu] sa1o0og
[II-SINM +
age “E"9
 Appendix C

Table €.4. Frequencies (Cumulative Percentages) of Differences

Between WAIS-III FSIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method)
WMS-IIl Percentage
Primary Indexes 25 20 15 10 5 4 3 2 1
Auditory Immediate 10 12 14 17 23 24 26 28 30
Visual Immediate 12 14 18 22 29 30 33 36 44
Immediate Memory 10 12 15 17 23 24 26 28 32
Auditory Delayed 9 12 14 17 23 25 26 30 33
Visual Delayed 12 14 17 21 26 27 30 33 44
Auditory Recog Delayed 10 12 16 20 25 26 28 30 35
General Memory 9 11 13 16 22 24 26 28 33
Working Memory 8 11 13 15 20 21 22 23 29

Table €.5. Frequencies (Cumulative Percentages) of Differences

Between WAIS-III VIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method)
WMS-III Percentage
Primary Indexes 25 20 15 10 5 4 3 2 1
Auditory Immediate 9 12 15 18 23 26 27 30 35
Visual Immediate 13 15 19 23 29 31 33 39 46
Immediate Memory 10 12 15 19 25 26 28 32 37
Auditory Delayed 9 12 15 18 25 26 29 31 36
Visual Delayed 12 14 18 21 28 30 33 36 45
Auditory Recog Delayed 11 13 16 19 25 28 29 32 37
General Memory 9 12 14 17 24 25 27 31 38
Working Memory 9 11 13 17 22 23 25 28 31

Table €.6. Frequencies (Cumulative Percentages) of Differences

Between WAIS-III PIQ Scores and WMS-III Index Scores:
All Ages (Simple-Difference Method)
WMS-III Percentage
Primary Indexes 25 20 15 10 5 4 3 2 1

Auditory Immediate 10 12 15 19 24 26 27 30 34
Visual Immediate 11 14 17 21 29 31 33 34 37
10 12 15 18 23 25 27 29 33
Immediate Memory
10 12 16 19 24 25 28 30 34
Auditory Delayed
10 13 16 20 27 28 30 31 38
Visual Delayed
11 14 16 20 26 27 29 32 36
Auditory Recog Delayed
10 12 14 17 23 25 27 29 31
General Memory
8 11 ike 15 20 22 23 25 27
Working Memory
obtained WMS-III Index scores
Note. Percentages for Tables C.4-C.6 represent the portions of the sample who
lower than their WAIS-III 1Q scores by the specified amount or more.

343
Appendix C

Table C.7. Differences Between WAIS-III Scores and Actual WIAT-II

Subtest and Composite Standard Scores Required for
Statistical Significance (Simple-Difference Method)
Ages 16:0-19:11 (High School)
Subtests
Word 10.18 9.75 10.18
Reading 01 10.95 11.61 13.41 12.84 13.41
Numerical 05 8.82 9.30 10.60 10.18 10.60
Operations 01 11.61 AQEZA 13.95 13.41 13.95
Reading 05 9.30 oes 11.00 10.60 11.00
Comprehension 01 12.24 12.84 14.48 (RES 14.48
Spelling 05 10.18 10.60 11.76 Ess a VETS
13.41 13.95 15.48 14.99 15.48
Pseudoword 7.78 8:32 O75 9.30 SS
Decoding 10.24 IOs 12.84 12.24 12.84
Math 05 9.30 HS 11.00 10.60 11.00
Reasoning 01 |Z 12.84 14.48 13.95 14.48
Written 05 eeG 22 13:15 12.82 1SalS
Expression 1S Sie 7-34 16.87 Nes
Listening 05 14.10 14.40 15.28 14.99 15.28
Comprehension 01 18.56 18.96 20.11 1978 20.11
Oral 05 12.47 12:82 13.79 13.47 (RERPAS,
Expression

Composites
Pending 05 6.57 7.20 8.82 8.32 8.82
01 8.65 9.48 11.61 10.95 11.61
Natharaatee 0)5 7.78 8.32 OMS 9:30 Sh iAs
01 10.24 10.95 12.84 12.24 12.84
Written 05 9.30 O75 11.00 10.60 11.00
Language 01 12.24 12.84 14.48 iSkke 14.48
Oral 05 11.76 Ze N2 1s 12.82 eed
Language 01 15.48 15.96 7-3) 16.87 A e/Ror

Total

Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

344
 Appendix C
Table €.7. Differences Between WAIS-III Scores and Actual WIAT_II
Subtest and Composite Standard Scores Required for
Statistical Significance (Simple-Difference Method) (continued)
College Students
Subtests p FSIQ via PIO. vcl POI
Word .05 7.20 7.78 B30 8.32 eS)
Reading 01 9.48 10.24 12.24 10.95 12.84
Numerical 05 7.78 Shey 9°75 8.82 10.18
Operations 01 10.24 10.95 12.84 iG 13.41
Reading .05 7.20 7.78 9.30 8.32 Sh 7)
Comprehension 01 9.48 10.24 12.24 10°95 12.84
Selling 05 8.32 8.82 10.18 930) 10.60
01 10.95 11.61 Ses 12.24 1BQs)
Pseudoword 05 7.20 7.78 OSG 8.32 SS)
Decoding .01 9.48 10.24 12.24 10.95 12.84
Math .05 9.30 9.75 11.00 10.18 id 39)
Reasoning 01 12.24 12.84 14.48 13.41 14.99
Written .05 14.70 14.99 15.83 15.28 16.10
Expression 01 19.35 19.73 20.84 20.11 21.20
Listening 05 11.39 11.76 12.82 (212 NESTS
Comprehension 01 14.99 15.48 16.87 15.96 73
Oral 05 15.28 15.56 ers 15.83 16.63
Expression 01 20.11 20.48 DENS 20.84 21.89

Composites
.05 5.88 6.57 8.32 7.20 8.82
Reading
.01 7.74 8.65 10.95 9.48 11.61
. .05 720) 7.78 9.30 8.32 oe5
pasruomatios 01 9.48 10.24 12.24 10:95 12.84
Written oD 11.900 10.18 11.389
Language 12.24 12.84 14.48 13.41 14.99
Oral .05 11.00 11.39 12.47 11.76 12.82
Language

Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

345
Appendix C

Table C.7. Differences Between WAIS-III Scores and Actual WIAT-II

Subtest and Composite Standard Scores Required for
Statistical Significance (Simple-Difference Method) (continued)
Adults
Subtests p FSIO. via PIO vcl POI
Word 05 7.20 Waihe: 9.30 8.32 HS
Reading 01 9.48 10.24 12.24 10.95 12.84
Numerical 105 7.78 8.32 9.75 8.82 10.18
Operations 01 10.24 10.95 12.84 11.61 13.41
Reading 105 7.20 7.78 9.30 8.32 O75
Comprehension 01 9.48 10.24 12.24 10.95 12.84
Spelling .05 8.32 8.82 10.18 Sis 10.60
01 10.95 11.61 13.41 12.24 13.95
Pseudoword .05 AY 718 9.30 8.32 O75
Decoding 01 9.48 10.24 12.24 10.95 12.84
Math 05 9.30 9.75 11.00 10.18 Ni leeshe)
Reasoning .01 12.24 12.84 14.48 13.41 14.99
Written 0s 14.70 14.99 15.83 15.28 16.10
Expression 01 12.35 1973 20.84 20a 21.20
Listening .05 iiess) (ileykss 12.82 1212 Sets
Comprehension 01 14.99 15.48 16.87 15.96 (SH.
Oral .05 15.28 15.56 16.37 15.83 16.63
Expression 01 Oil 20.48 ZAESS 20.84 21.89

Composites
Reading 05 5.88 6.57 8.32 720 8.82
01 ea, 8.65 10:95 9.48 11.61
Mathematics 05 7.20 7.78 9.30 8.32 S75
01 9.48 10.24 12.24 10.95 12.84
Written .05 9.30 a7) 11.00 10.18 ess
Language .01 2524. 12.84 14.48 13.41 14.99
Oral .05 11.00 11.39 12.47 11.76 12.82
Language 01 14.48 14.99 16.42 15.48 16.87
Total 05 5.88 6.57 8.32 7.20 8.82
01 774 8.65 10.95 9.48 ARGH
Note. For all age groups, data were derived from the WIAT-II age-based reliability tables.

346
 Appendix C

Table €.8. Differences Between WAIS-III FSIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample
High School Percentage
Subtests 25 20 15 10 5 4 3 2 1
Word Reading 8 10 12 1S 19 20 22 24 ay
Numerical Operations ) 9 11 14 18 19 20 22 25
Reading Comprehension 8 10 12 15 20 21 28 25 28
Spelling 8 10 13 16 20 21 23 25 23
Pseudoword Decoding 9 11 14 1 22 24 26 28 eZ.
Math Reasoning 6 8 10 12 16 AZ 18 19 22
Written Expression 9 a 14 ied 22 24 25 28 32
Listening Comprehension ii £ el 13 ey, 18 19 21 24
Oral Expression 10 12 16 19 25 26 28 31 85

Composites

Reading 7 9 11 14 18 19 20 22. 25

Mathematics 6 8 10 2 15 16 Wa 19 21

Written Language 8 10 is 15 20 21 pe 25 28

Oral Language 7 10 i 17 18 A9 2a 23

Total 6 7 8 10 13 14 15 17 19

below their
Note. Percentage of individuals whose actual achievement standard score was
FSIQ by the specified amount or more.

347
Appendix C

Table C.8. Differences Between WAIS-III FSIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained byVarious
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
College Students Percentage
Subtests 25 20) #15 G10 5 4 3 2 1

Word Reading 10 (2 316) S18 25 27 29 31 35

Numerical Operations 9 11 14 18 PES 24 26 28 a2

Reading Comprehension 9 12 14 18 3 24 26 29 33
Spelling 11 ite 16 20 26 28 30 38 37
Pseudoword Decoding 10 13 16 20 26 IZ} 29 32 36
Math Reasoning 8 10 13 15 20 al 23 25 28
Written Expression 11 is Ne) 20 27 28 30 33 38
Listening Comprehension 8 10 13 16 ai 22 24 26 29
Oral Expression 11 13 iW 21 27 28 31 33 38

Composites

Reading 9 11 13 16 21 22 24 26 30
Mathematics 8 10 12 15 19 20 22 24 2g,
Written Language 10 13 16 20 25 27 29 32 36
Oral Language 9 11 ie 16 21 23 24 27 30
Total 8 10 12 iKs 19 20 22 24 Diy,

Note. Percentage of examinees whose actual achievement standard score was below their
FSIO by the specified amount or more.
 Appendix C

Table €.8. Differences Between WAIS-III ESIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
Adults Percentage
Subtests Na NRSo = ol 10 5 4 3 2 1

Word Reading 10 13 16 20 21 23 26 29
Numerical Operations 8 3 12 15 16 17 19 24
Reading Comprehension 9 11 is iy 18 20 21 24
Spelling 84S BIG B20 21) Be 29
Pseudoword Decoding oO No (5 BIS | Birbeed20 D427
Math Reasoning 8 660° 81S "16° 17 woteee204mr23
Written Expression ce RAS 047) B22 BBS 27 80
Listening Comprehension BS Olt B14 Fic. s1Sme2terO3 erie
Oral Expression DD
OO:
OO.
OO
|SO
|St
malt 13 16 AS) 25 26 28 31 35

ou
Composites
Reading ‘i 8 10 ifs! 14 15 16 18
Mathematics 8 9 12 15 16 iv 19 21

Written Language =o = =iwo 16 21 23 24 26 30

Oral Language 10 13 15 20 21 23 25 28
Total 7 8
A;DIo;n)|n 10 13 14 1S Wy ie

Note. Percentage of examinees whose actual achievement standard score was below their
FSIQ by the specified amount or more.

349
Appendix C

Table C.9. Differences Between WAIS-III VIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained byVarious
Percentages of the WIAT-II/WAIS-III Linking Sample
High School Percentage
Subtests 25 020 #15 W410 5 4 3 2 1

Word Reading 8 9 |Z 14 19 20 21 23 26
Numerical Operations 8 10 2 15 19 20 22 24 27,
Reading Comprehension 8 9 i 14 18 19 21 23 26
Spelling 8 10 ile 16 21 22 24 26 29
Pseudoword Decoding fe) did 14 17 22 23 25 27 31
Math Reasoning 7 9 11 14 18 19 21 23 26
Written Expression 9 11 14 18 23 24 26 28 SW
Listening Comprehension 6 8 10 12 15 16 A 19 22
Oral Expression 10 13 16 20 26 27 29 eZ 37

Composites

Reading 7 8 10 13 0h 18 19 21 23
Mathematics 7 i) 11 14 18 19 20 22 25
Written Language 8 10 13 16 20 22 23 25 29
Oral Language 7 9 11 is eZ, 18 19 21 24
Total 6 i 9 11 tS 16 17 18 21

Note. Percentage of individuals whose actual achievement standard score was below their
VIO by the specified amount or more.
 Appendix C

Table €.9. Differences Between WAIS-III VIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
College Students Percentage
a
Subtests 25 020 45 ©10 5 4 3 2 1
a a ce
Word Reading 10 12 15 1) 24 25 27 30 34
Numerical Operations 10 12 15 19 24 26 28 30 35
Reading Comprehension 9 12 (is 18 23 25 26 29 33
Spelling 10 13 16 20 26 ay, XS) eZ 36
Pseudoword Decoding 10 13 16 19 25 27 29 oi 36
Math Reasoning 9 11 rls: 17 2 23 24 27 30
Written Expression 11 14 17 21 27 29 31 34 69
Listening Comprehension 8 10 13 16 20 21 23 25 29
Oral Expression 14 14 18 22 28 30 32 35 40

Composites

Reading 8 10 13 16 20 21 23 25 29
Mathematics ] 11 Ag 16 21 22 2A Owe oO
Written Language 10 cs 16 20 26 27 29 32 37
Oral Language 9 11 14 ils 22 23 25 27 31
Total 8 10 12 15 20 21 22 24 28

Note. Percentage of individuals whose actual achievement standard score was below their
VIO by the specified amount or more.

351
Appendix C

Table €.9. Differences Between WAIS-III VIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
Adults Percentage
SS SS FS A I ER PA SE ES ST SLE EEE AF ETE

Subtests 25 20 #15 ~ #10 5 4 3 2 1

Word Reading 9 11 iis 17 21 23 24 27 30

Numerical Operations 7 8 10 2 16 17 18 20 22
Reading Comprehension 7 g) 11 14 18 19 21 22 25
Spelling 8 10 ie 16 20 BS 23 25 28
Pseudoword Decoding 9 11 13 AZ 21 23 24 Dy, 30
Math Reasoning y, 9 et 13 17 18 20 21 24
Written Expression 9 11 14 17 22 2S 25 27 31
Listening Comprehension 8 10 12 15 20 21 22 24 28
Oral Expression 10 is 16 20 25 27 29 32 36

Composites

Reading 7 9 11 14 1S 16 flit 20
Mathematics 8 10 12 16 she 18 20 22
Written Language —" = a Ww 5= ~N 21 23 24 2g. 30
Oral Language 10 is 16 21 22 24 26 29
Total D)}o;o;inN|o® 7 9 11 13 14 is AE 19
a a I I RE EES

Note. Percentage of individuals whose actual achievement standard score was below their
VIQ by the specified amount or more.
 Appendix C

Table €.10. Differences Between WAIS-III PIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample
High School Percentage
Subtests 25 920 #415 10 5 4 3 2 1
a
Word Reading 10 ie 16 19 25 27 29 Sil 36
Numerical Operations 9 11 13 16 21 28 24 26 30
Reading Comprehension 11 sks) 17 21 27, 28 30 3S 38
Spelling 10 12 15 19 Va 25 B27, 30 836334
Pseudoword Decoding 16 is 16 20 26 28 30 33 ou
Math Reasoning 8 g 2 14 ine, 20 21 a} 26
Written Expression 10 13 16 19 25 26 28 31 35
Listening Comprehension 10 12 15 19 24 26 28 30 65
Oral Expression 10 se 16 19 25 27 29 31 36

Composites

Reading 10 ie 15 19) 925°) 26") 23s 35

Mathematics 8 ¢) 12 14 19 20 24 23 26
Written Language ; 12 15 18 23 25 27 29 33
Oral Language 5) 11 13 16 21 22 24 26 30
Total 8 s) 12 14 18 2071 23 26

Note. Percentage of individuals whose actual achievement standard score was below their
PIO by the specified amount or more.

353
Appendix C

Table C.10. Differences Between WAIS-III PIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
College Students Percentage
Subtests 25 20 15 10 5 4 3 2 1
Word Reading 12 14 18 Ze 29 30 33 36 41
Numerical Operations 9 S12 “15 118 423. 2ieel2Gi 2oterse
Reading Comprehension 10 1183 16 20 26 27 29 2 Boe
Spelling 12 14 18 22 US) 31 3S 36 41
Pseudoword Decoding 12 15 18 23 zg 31 33 36 41
Math Reasoning 9 11 14 Ae 22. 23 25 28 31
Written Expression Hal 14 ile: 21 2, 29 31 34 39
Listening Comprehension 10 16 16 Is 25 26 28 Su 3
Oral Expression 11 14 NY 21 27 ae 31 34 38

Composites

Reading 11 13 16 20 26 28 30 33 Yi)
Mathematics 9 11 13 16 21 22 24 26 30
Written Language 11 14 17 21 27 29 31 34 38
Oral Language 10 12 15 19 24 26 28 30 35
Total 9 2 15 18 23 25 26 29 33

Note. Percentage of individuals whose actual achievement standard score was below their
PIO by the specified amount or more.
 Appendix C

Table €.10. Differences Between WAIS-III PIQ Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
Adults Percentage
Subtests 25° @20) #155 F110 5 4 3 2 1
Word Reading 9 11 14 18 oe 24 26 28 Bz
Numerical Operations 8 10 13 15 20 21 23 25 28
Reading Comprehension 9 11 13 16 21 22 24 26 30
Spelling 10 12 15 19 24 26 20mereOm 34
Pseudoword Decoding 8 10 is 16 20 1 23 25 28
Math Reasoning 8 10 is 16 | 22 23 26 29
Written Expression 10 12 15 18 24 25 27 29 Ss
Listening Comprehension 2 11 13 16 21 22. 24 26 30
Oral Expression 11 is 16 20 26 29) MeCOMS2 Sa

Composites
Reading x rs) 11 13 17 18 19 21 24
Mathematics 8 10 12 15 208 21 23 sy. | PAS}
Written Language 10 12 15 is 24 26 28 30 34
Oral Language 9 AZ 14 18 2s 24 26 28 32
Total 8 9 12 14 197 20 Di 23 26

Note. Percentage of individuals whose actual achievement standard score was below their
PIQ by the specified amount or more.

355
Appendix C

Table €.11. Differences Between WAIS-III VCI Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample
High School Percentage
Subtests 25° 820 $15 810 5 4 3 2 1

Word Reading 8 10 s12 5 849 20 pa@dsewase7

Numerical Operations 9 11. ©19 896 821 ‘22noW2 Ae Baier a0
Reading Comprehension 8
Spelling ) 11 14 17 22 24 26 28 32
Pseudoword Decoding 10
Math Reasoning g
Written Expression 10
Listening Comprehension Hi 8 10 iz 16 17 18 20 oe
Oral Expression 11 13 17 20 26 28 30 Se 3u.

Composites

Reading Zz 9 11 14 18 19 20 22 25
Mathematics 8 10 is 16 20 22 23 25 DE,
Written Language 9 11 14 ies 22 23 25 27 31
Oral Language 7 9 i 14 18 19 20 22 25
Total 7 9 11 ile ile, 18 20 22 25

Note. Percentage of individuals whose actual achievement standard score was below their
VCI by the specified amount or more.

356
 Appendix C

Table C.11. Differences Between WAIS-III VCI Scores and Actual WIAT_II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
College Students Percentage
Subtests 25° 820 #15. ¢10 5 4 3 2 1
Word Reading 10 13 16 Re 25 27 29 31 36
Numerical Operations 11 13 17 20 26 28 30 ee Si,
Reading Comprehension 10 12 15 19 24 25 27 30 34
Spelling 17 13 17 2) DT 28 30 Bs 38
Pseudoword Decoding Lt 13 17 20 26 28 30 33 Si
Math Reasoning 10 12 15 19 24 26 28 30 34
Written Expression 11 14 17 22 28 30 32 £15) 40
Listening Comprehension rs) 19 13 16 21 22 24 26 30
Oral Expression 12 15 18 23 29 31 34 37 42

Composites

« Reading g 17 14 17 22 23 25 27 Sil
Mathematics 10 12 15 18 24 25 Dif 30 34
Written Language 11 14 17 21 20 29 31 33 38
Oral Language 10 12 15 18 23 25 Dy, 2) SS
Total S, 11 14 17 22 23 25 if Sil
SR RS a a SES PA AAS SSS ERSEES RTS TERS SES SP SR

Note. Percentage of individuals whose actual achievement standard score was below their
VCI by the specified amount or more.

357
Appendix C

Table €.11. Differences Between WAIS-III VCI Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained byVarious
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
Adults Percentage
Subtests 25 20 15 10 5 4 3 2 1

Word Reading £ 11 4 17 22 23 25 27 31

Numerical Operations 7 9 1 13 17 18 20 21 24
Reading Comprehension 8 , 2 14 18 19 21 23 26
Spelling g 11 13 16 21 22 24 26 29
Pseudoword Decoding 9 12 15 18 23 25 2] 29 33
Math Reasoning ei 9 1 14 18 19 20 22 25
Written Expression 9 11 AZ 22 23 25 2F 31
Listening Comprehension 8 10 12 15 19 20 22 24 oe |
Oral Expression 10 13 16 20 26 27 29 32 36

Composites

Reading 6 8 10 12 15 16 17 19 22
Mathematics 7 9 11 13 17 18 19 21 24
Written Language 9 11 14 17 22 23 25 Fa| 31
Oral Language 8 10 13 16 21 IP 23 26 29
Total 6 8 9 12 15 16 17 19 21

Note. Percentage of individuals whose actual achievement standard score was below their
VCI by the specified amount or more.
 Appendix C

Table ©.12. Differences Between WAIS-III POI Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample
High School Percentage
Subtests 26° 620 «4816 40 5 4 3 2 1
Word Reading 1 14 17 21 27 29 | 34 iS)
Numerical Operations 8 10 12 15 19 20 22 24 27
Reading Comprehension 11 14 18 22 28 30 32 85 40
Spelling 10 13 16 20 25 27 29 32 36
Pseudoword Decoding 11 14 17 21 28 29 Bi 34 39
Math Reasoning 7 8 10 12 16 17 18 20 23
Written Expression 10 ie 16 20 26 27 29 32 37
Listening Comprehension 10 12 15 19 25 26 28 31 35
Oral Expression 8 10 13 1s 20 21 28 25 28

Composites
Reading 10 13 16 20 26 HH) 29 32 36
Mathematics 6 8 10 12 17 16 18 (9 22
Written Language 10 12 15 19 24 26 28 30 34
Oral Language 8 10 12 15 19 20 22 24 27
Total 7 9 11 13 iW 18 19 21 24

Note. Percentage of individuals whose actual achievement standard score was below their
POI by the specified amount or more.

359
Appendix C

Table €.12. Differences Between WAIS-III POI Scores and Actual WIAT-II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
College Students Percentage
Subtests 25° 920 #15 910 5 4 3 2 1

Word Reading 12 5 (iis) 22 29 Si 33 36 41

Numerical Operations 10 12 VS 19 24 26 28 30 215)

Reading Comprehension ili ik il 20 PAI 28 30 33 38
Spelling 12 16 18 23, 29 31 34 37 42
Pseudoword Decoding |Z 15 ig 23 30 32 34 37 43
Math Reasoning g) 11 14 17 22 JES 25 28 31
Written Expression 11 14 iw 21 28 ue 32 So 39
Listening Comprehension 11 13 16 20 26 28 30 32 37
Oral Expression al 14 ly 21 2 29 31 34 38

Composites
Reading itl 14 17 21 27 29 3 34 39

Written Language WW 14 17 zi 28 BS 32 SS SiS)

Oral Language 10 13 16 (8) 25 26 28 31 So
Total 10 12 15 18 24 25 ay, 30 34

Note. Percentage of individuals whose actual achievement standard score was below their
PO! by the specified amount or more.
 Appendix C

Table €.12. Differences Between WAIS-III POI Scores and Actual WIAT_II
Subtest and Composite Standard Scores Obtained by Various
Percentages of the WIAT-II/WAIS-III Linking Sample (continued)
Adults Percentage
Subtests 25 20 15 10 5 4 3 2 1
aa
Word Reading 10 12 15 19 25 26 28 cil 30
Numerical Operations 9 11 14 17 22 28 25 27 Si
Reading Comprehension 9 12 14 18 23 25 26 29 es)
Spelling 11 14 (WA 21 27 29 981 Cia S
Pseudoword Decoding 9 11 14 Zz, 22 23 25 27 31
Math Reasoning 9 11 14 Ny ee 24 26 28 2
Written Expression 10 13 16 19 255 27 29 31 36
Listening Comprehension 9 12 14 18 Ze 24 26 28 Se
Oral Expression 11 13 17 20 27 28 30 Se 38

Composites

Reading 8 10 12 152ee 20) 824 23 25 28

Mathematics 9 11 14 17 PI 23 AS} Zi oii
Written Language 11 13 16 20 26 28 30 33 37
Oral Language 10 12 15 19 HE Ply hss RO) GY
Total 9 11 ic 16 21 22 2A eee oO

Note. Percentage of individuals whose actual achievement standard score was below their
POI by the specified amount or more.

361
 Bak - id ‘ss os I 7 7 4

7 :
subs, a

pine ee gery anaes GS

—_— > az 7
= oa = «
> ve | afl we :
oe = . a
: ‘ ~ Ri rei? guy. peers seh oa
4
orearane anwar
a 7 crn <—
 APPENDIX D

Frequencies of WAIS-—III 1Q Score

Discrepancies Based on Ability Level

363
 LLL
Lol ees L'8 LS LS es vt 02
LOL Lg eg 8Y L's es 0} ke
LOL Ov ev Be 0'€ ev S0 CC
bide Oe ge 6¢ 0'€ ee 00 €¢
LS Ove 6} vk 0'€ 6} 00 ve
O'V 0'e 6} 0'} 0'€ vil 00 GS?
0¢ 0'¢ 0'} G0 0'l yh 00 9¢
02 02 S'0 0'0 0} vt 00 LZé
0°? Ol S'0 00 0'} 0} 0°0 82
0} 0} S'0 00 “00 S'0 00 6¢
Ol Ol 00 0°0 00 0°0 00 0€
Ol O'l » 0'0 0°0 0°0 0'0 0'0 LS
Ol 0'l 00 0'0 0'0 0°0 00 ce
Ol 00 00 0'0 00 0°0 0'0 ce
00 0°0 0°0 0°0 00 00 0°0 vE
00 0°0 00 00 00 0°0 0°0 SE
00 00 00 00 00 0°0 00 9€
00 0'0 00 00 00 0°0 00 /BS,
00 00 0'0 00 00 0°0 0'0 8e
0°0 0'0 00 0'0 00 00 00 6€
00 00 0°0 0°0 00 00 00 Or<
peeds Asowa peeds pseds Aiowsy uoneziuebio Old-DlA Aouedaiosig
BHuissa901g Bupjiom BHuissad01d BHulssa001d Bupyiom jenydao1eg jo
—Aioway —uoljeziuebio —uolsuayaidwog —uoleziuebio —uolsuayaidwiog —uolsuayaidwog junowy
Bupjiom jenydad1eq Jeque, jenjdaoieq Jequa, Jequs,,
Appendix D

SdIXIPU]/SITBIS

364
5 62
Arpiqy:[427T DISA Aq
$a109§ xapu] SeduarazJIG
Usemjeg [II-SIVM
OJ pure sa~uenbely aANL[NUIND) (sase}UsdIeg
JO 3jqel“Lg
 ase sanyea aup J] ‘afdures ay} JO %Z Aq paure}qo Tr)
‘TeuoNdarIp uaaq Aay pey Se10os xepu] puke CO] Usamjaq saduarayfIp ot} ayeurxordde [jim Aap ‘omy Aq paplarp
U are Sa109§ xapu] pue OH] UIeMjaq SeduUerEIP 9U.L
Appendix D

36
sem II puke O]d SNUT OIA 0} pure OIA snuyU O]d 0} s1ayar syutod gz Jo aquarEyIp B a[durexa 10] ‘[euoToaIIp-UO
OL 09 09 09 0S ueIpe|\|
0°6 09
G9 19 19 L'9 61 gs
89 9°9
68 GL ol VL 19 ues|\
0h 0°8
0°00} 0°00} 0°00} 0°00} 0001 0
0°00} 0001
9°46 LS6 66 616 L246 L
0°86 0°26
£48 258 6°98 0°68 cov8 @
6°26 6°06
0°82 OL LYl cel erZ €
6°68 8°62
y'89 0°99 9°79 OL me) v
8°08 LOL
c £9 €'09 99S 8°6S 90S S
8 8Z 9°SS
69S bes Svs g'9S vV9V 9
89l Ges
LtS 8°87 GOV v9v BLE VE
LOZ Gly
vLYV LLY SGV Sev Lee 8
9°79 voV
Ler Le €0€ O'ee Ble 6
9°9S y8E
eve Sle L Le €°9¢ L Le 9°02 Ol
SSP
eee O'ee 6c occ vec jy/A\ bh
y'6€
ec 8°S¢ Gee co8k 9°02 8 rl ch
V9E
c 0c Sle i 78\' c8l e9l ra! €l
Ele
c 8k L6L 8'Sl bel vel 16 vi
CES
cSt L91 vrt bel Olt ol SI
ole
hel esi Oct LOL OL o9 Ot
c 6h
PLL 6 El Gibb 18 96 ee ZI
c Ol
bh O'bt 18 bl 18 ce 8
Sl
LOL 96 Lg L'9 8 Ve 6h
ck
peeds peeds Aiowey uoneziuebio Did-DlA Aouedsaiosiq
paeds Aioway jo
Bursseo0ig Buissa90ig Bupion jenydedieg
Buissa00ig Bupjiom
—uojsuayaidwog —uoeziuebio —uolsuayaidwiog —uolsusyaidwiog junowy
—Kiowew —uoneziuebio
Jequsp jenjdeo1eg Jequs, Jequs,
Bupyom jenydaaieq
SIXIPU] /S8TBIS
——————
(panuyuos) 5 62
Aiqy ‘12427 OISA Aq
sa10og xapu] SadUaTaFfIC
UseMjog III-SIVM
OI pue soIsueNba1y VANETNUIND) (sasvyUsd10g
JO ajgel*|{"g
 rE a a A RR SR RE RE LS
vol 6'cl cSt c Ol vcl 16 LY 02
6 Ol vcl O'el 88 S'6 9°8 ee ke
06 vol 9} [ofp G8 Jesh CG co
0°8 v'Ol 66 GL G9 v9 CC €?
OL 06 9°8 v9 Gs 0S a ve
0S GL 99 Ha) Sv vv vt Ge
0's OL LY LY Ov gE bh 9¢
SV G9 LY Lv 0'€ BC 9'0 Le
oh 0'9 0’ 6'€ Gc 61 €°0 8d
Ov 0'V Gc Oe “0¢ ae €0 6¢
Ov Ov Gc Gc 0'¢ El 0°0 0€
Ge Ge » al 6} 0¢ Lap 00 KS
Gc O'e vt 6} Gh 9°0 0°0 ce
Gk 0'¢ 80 vt Ol 90 00 os
O'l Gh 80 80 00 €0 0°0 ve
Ol Gh 80 80 0'0 €0 0°0 Ge
0} Gh 80 80 0°0 0°0 00 9€
0} Gk 80 9°0 00 0°0 00 LE
Ol
9°0
8€
00
0'0
80
0°0
OL
6€
00
0'0
€0
9°0
Ol 0°0 €0 €0 00 00 0'0 Op<
peeds Asowa| peeds peeds Asoway| uoneziuebio Dld-DlA Aouedaiosiq
Buissa901d
Burson
jo
jenydao10g
BHuissa90/1d
Buiysonn
Buissa90/1g
—Aioway
—uolsueyaidwog
—uoljeziuebic
—uolsuayaidwog
—uojeziuebic
junowy
—uolsueyaiduioy
jenydao1eq
jeqisa
jenjdao1eq
Jequa,
[equap
Burysion
Appendix D

Sdx9opU]/SIfBIS

366
Ayiqy ‘12427 OISd 68-08 Aq
Sa109¢ xapu] saouatajsiq CI pue
Useemjeg III-SIVM JO
sa—uenbaly aANe[NUIND) (saseyUsd1eg age] -Z"g
 ‘jeuoNdamp usaq Aay) pey sa109s xapuy pue O] Udemjaq saduaiayIp at ayeurxordde [jim Aaup ‘om} Aq paplalp are sanyea aup J] “eduues ay} Jo %z Aq peureyqo

367
Appendix D

SPM }I puke O]d SNUIUT OIA 0} pue OIA snuIW O]q 0} Stayer syutod gz Jo adUarayyIp & ofdurexe 10] ‘[eEUOTIeIIP-UOU are SaI0Ig Xapu] puke CJ UseMjaq SadUaIEHIp 9Y.L
EEE
0°8 0°8 0°8 OL 02 GL 09 UeIPS|\|
L'8 c8 8 SL gL ee 6S gs
ZO 86 c Ol €6 68 L6 OL ues|\|
0001 0°00} 0°00} 0'001 0°00} 0°00} 0001 0
0°86 0°26 9°S6 co ¥6 S°c6 v6 y'96 b
S26 0°S6 S°c6 6°06 9°68 LZ 16 928 6
9°78 9°98 0°¢8 G18 b6Z 208 gel €
992 8 eZZ 0°92 9CL G°9Z 6°69 v
ZL9 co99 is S°89 Leg LL9 209 S
cS9 Z09 G'S9 v'6S c 09 619 vVS 9
c 8s evs 9°19 Lvs ols 99S 9°8Y us
LYS Z0S 8°¢S bY €6v 00S 9 EV 8
© 6V eer 8 Ly Cov Sly OV LOE 6
8 >P 8°9E 6 EP L8€ €9E G'6e ce Ol
Cer os LE Sve v'6¢ vGE 82 Lh
8°8E 6°62 LSE eee 692 9°62 0've cl
8°9E yLé 0'CE €'6¢ 6'9¢ 9°12 Ske el
y'8¢ 6°S¢ le Sc vee L02 9°91 vi
6°S¢ 6c vS¢ coe v0¢ c8i Cel Sl
vst v6 Ge 9°61 681 9°91 OLE Ot
v9l v'St v'0¢ VL v9l 9'vl 88 Li
6 v1 vvl c8l Lvl 61 vl 9°9 8
vel vvl VL 6th 6 rk Ol 8S 61
peoeds Asoway peeds peeds Asoweaw uoneziuebio Old-DlA Aouedaiosig
Buissa00iq Bupion Bulssa90ig Bulssa90ig Bupiom jenydadieg jo
-Aiowey —uoleziuebio —uolsusyoidwog —uojez|ueB19O —uolsueyaidwiog —uolsusyeidwog yunowy
Bupiom jenjdesieq Jeqica jenydadieq jeqiusa Jequsp
SIXIPU]/SITVIS
(panuyju0d) Apiqy 1249 OISA 68-08 Aq
xopu] $3109¢ SaduarasTIG
UaeMjag III-SIVM
OJ pure safuanbaty aATeINUIND) (sadejUsd1ag
JO sige-Z"g
 EE
€9l 6cl eB 9°91 4! Sab OL 02
ol OLE o9l vl S'6 vOl 8S ke
9'El 86 vl Lol 88 o8 97 44
6 LL G8 Gch 601 Pe VL LE €¢
6'0l Cr oth G8 6S Gs 8°¢ ve
vol e'9 0'6 el es LS VC Go
€'8 67 L'8 L9 vv 6'€ HI 9¢
aie vv 89 LS Lv Le Gi} Le
19 Le og 91 i) 9°¢ LL 82
ov Le vs Lv - 626 Co Z0 6¢
ov 6k 6'V ge 6¢ (9)|! 9°0 0€
LS Vii ev Le 4 9°} v0 Le
ge vt ve Ve 0¢ vk €0 ce
be c| Oe 0'¢ Lt 80 0 €€
Ge Z0 6? HE|! JE! Z0 LO ve
ve Z0 0'¢ Gt vk 0 0°0 Ge
0'¢ Z0 9} GI vt v0 0°0 9€
Gk €0 bh LL vt c0 00 Le
vk 20 Z0 80 G0 c0 00 8€
cl 00 Z0 Z0 G0 c0 0'0 6€
Ol 0°0 9°0 Z0 SO c0 00 Ov<
pseds Aiowal pseds peeds Asowosy uoneziuebio DId-DIA Aouedaiosiq
Buissao0id Buryson BHuissa901q Buissa01d Buiyson jenjdad18q jo
—Aiows —uoljeziuebio —uolsuayaidwog —uoljeziuebio —uolsuayaidwog —uolsuayaidwog junowy
Burpiom jenjdadi8q |eqis, jenjdadieg Jequs, JBqisa
Appendix D

SdxXoIPU]/SITBIS
601-06 ATIqy :12427 OISd Aq
xapu] $3109§ SadUaJIIJIG
U20M)eg [[I-SIVM
QC] pue satIsuenba1y JANLINUIND) (SasejJUsd10g
Jo 3jqgel“¢E"g
 are sane ay} J] ‘aydures ayy Jo %z Aq poutego
‘TeUOTaIIp Useq Ast] peY Saiods XapU] PUL OJ UseMjaeq SedUaIEHIp at} areutxoidde jim Aay} ‘om Aq papiaip

369
U are sa10dg XapU] pue CD] Usemjaq seouaIayIp 9U.L
sem II pue O]d snuruL OIA 0} puke OIA snurur Ojd 0} srayar syutod gz Jo souaIayIp & a]durexa 10] ‘PUOTIaIIp-UO
Appendix D

0°8 0°01 06 08 0°8 OL UeIpe|\

0°6
BL 88 G8 BL 97 v9 gs
68
66 ott bbb bol 96 9°8 ueel|\|
lh
0°00! 0°00} 0°00! 0°00! 0°00! 0
0°00} 0°00}
€6 8°96 0°66 co96 9°96 IL
6°96 €°96
66 0'€6 9°68 268 8°88 4
6°26 9°68
0°88 8°S8 cv8 818 v8 iS
9°98 G18
6SZ co 18 y08 8B BSL 6 Vv
8°08
eZZ GZ 669 €69 89 S
Gel 8°69
eg OL LOL y'99 eg 619 9
6°69
€'6S 6'S9 8°¢9 8S v'8S 97S Z
89
O'S LL9 62S LSS 90S €8v 8
28S
6° 8Y 09S BLS L8v ely OEY 6
Les
Lov ees G'8Y cov Sev cle Ol
8 SV
vyOv OLY Sev 90V VLE VCE bE
Lvy
Lee Cov 0'0V 8'9E eee Gle cl
9°6E
c 0€ L6E 6'SE cle 6°62 9? el
OLE
Le ee 8°ce ele 1'9¢ c 02 vi
9°Ce
eve ce v 6¢ eve L'€¢ GL St
vy0€
661 6°82 1S? bbe 661 e'St 9}
ele
8°81 8°9¢ Cee v 6k 8k bok Zl
6°0¢
LO} 6c 8°02 bSt GSl 8°01 8t
vie
8 Lh ISL ele Z8i bel vel 88 61
peeds peeds Asowey uoneziuebi9 DId-OlA Aouedaiosiq
peeds Asowsw
Bulsseo0id Burpiom jenydeo10q jo
Buissa0id Bupjiom Bulssa00/1d

(panu1ju02)
junowy

(sa8ejuad1eg
—uolsusyaidwiod

sAneINUIN))
—uolsusyaidwog

sasuanbely
—Aioweaw —uoleziuebio —uolsusyaidwog —uojeziuebio

Sa9dUaIasJIC
Jeqisn jenydao1edq jequs, jequs,
Bunion jenjdeo10dq
ee
ee

GOT-06
Aypiqy
SdXdPU] /SITBIS

III-SIVM

Use0MjJog

2124297
OISA
a —

$a109§

3ajgel
reer

Aq
xapu]

“|E"g
pue

JO
QI
 ajgeL*y"g seUanbery BANeTNUIND) (seseyUadIeg
JO saduarayyIC
Uaemjag III-SIVM
OI puke xepuy Sa109§
Aq

370
AypIqy*feAe7 OISd I-O1T
61
SoxdpU]/SITBIS

[equea; [equa jenydaaieq

Appendix D

Jeqisp jenydaoiaq Bupjiom

junowy —uolsusysidwoy —uolsusyaidwog —uoneziuebic —uolsuayaidwoy —uoljeziuebio —Aiowsay
jo jenjdao190q Buryiom Buissa90iq Buissa90ig Bupiom Buissad0i1q
Asuedasosiq DIid-DIA uoleziuebico Asowa paeds peeds Asowa pseds
Ov< 0 S‘0 v0 G0 Gh el 0°0
6€ Z0 G0 v0 G0 GI el 0°0
8€ Z0 S'0 v0 G0 Gk cil 00
ZE Ol Z0 v0 S°0 6k el 60
9¢ On Z0 v0 S0 AKG ral 730
Ge ck Z0 v0 0} Le (Ei L?
ve cl Zt v0 Gh 6¢ b? Le
€€ ley? ACG v0 VC 9 9°¢ 4
ce at Le L? ve Ly 9'¢ 96
be 6°} ve L’? 6'E 9g VE» Ove
0€ 6} 9¢ Le 67 9°9 Be LS
6¢ CG 6€ Vie €9 GL 9S 9g
8¢ Le Lg Oe My Bl v9 0°9
Zé 6¢ VS ve €8 Z6 89 G8
9¢ vv €'9 8’ L6 Lt G8 €Ol
Gc 97 Bi 9S thc vcl Ot Gt
ve VS 88 9S 6th OE Z0h Let
€¢ 89 vcl WL 9'EL 9S} [Ey vt
oC 08 bel L'8 ol Gli Gh LL
Lé G8 vl8 86 8c v'0¢ cel vst
02 6 G9l 86 vile Vcd Stl 6°02
 ‘TeuoMoaIIp usaq Aatp pey saioos xapuy pue OI Usamjaq saduarayIp ayy ayeurtxodde [jm Aay) ‘omy Aq paplarp are sanjea au} J] -ajdures au Jo %z Aq paurejqo

371
Appendix D

sem 11 puke OIg snutur OIA 0} pue OIA snr Yd 0} Stayer syutod gz Jo aouarayIp B a[durexe 10y ‘feuONdaIIp-UOU are Sa109S XepU] puke O] UdEMjaq SedUaIEFIP OU.L
LD
0'0} 00 Och 00 06 06 0°8 UeIps||
68 9°8 v6 68 SL 28 OL gs
6 tL 60l Leb aA Z0l Ol 9°6 Ued|\|
0°00} 0°00} 0°00! 0°00} 0°00} 0°00! 0001 0
6°26 9°66 €°86 €°86 0001 ov6 8°96 |
6°16 9°06 c v6 6°26 6°16 o 16 €'68 C
€98 ¢ 48 9°98 8°88 2 06 Ses c€8 €
9°18 662 v8 018 18 0°08 VOL v
LOL GOL 62 LGL 9°GZ cel 8°02 S
LZ9 y'89 Lvl OL 629 6°99 Lv9 )
L'79 v'c9 LOZ 6°29 bL9 9°19 6°8S Z
c £9 €°09 6°S9 oS9 els o9S 8S 8
LLS LoS 19 62S v'0S oes v8v 6
LoS v'0S L'6S evs L8v LGV SEV Ol
cov €6e O'vS L6y 9EV eGy 0'8E bh
voy €'6€ 90S S vv Cov L6E VCE ol
OlP €°0€ VLV OLY 8'9E v'8E LLle el
e9e S'6¢ 9°oV c 8e 6SE 6'0E 8c vi
6ce 8'7C v Ov €'9€ Loe c 62 9°0¢ Sl
coLE 8 e BE LLE 6'9¢ ele v0¢ Ot
vile L'0¢ VLE c 62 ES 82 c8i ZI
SC 6 Lh c 8c gle vst 6°¢¢ QE 8
CCC SLi 8'S¢ 8'7¢ O'S S'6l oth 61
peeds Auowoy pseds psoeds Aiowey uoieziuebio Old-DlA Aouedaiosiq
Buisseo0iq Bupon Bulssa00idg Buisseo0iq Buryion jenjdac19g jo
—Aioway —uoj}jeziuebio —uoisuayaidwog —uoljeziuebio —uoisusysidwog —uolsusyaidwog junowy
Bunion jenjdacieg [equa jenjdao1e8q Jeqian jeqian
SdX9PU]/SITBIS
(panuyjuo9) GT
Ariqy 312497 OISA T-O1T Aq
xapu] $9109 SaduUaIaTIG
Uaemjag [II-SIVM
OJ pue sasuanbery aAVe[NUIND) (sadejUao1ag
Jo age “y"q
 0'9¢ LSh 9°82 8°CC 0'¢¢ vst cv 02
oS¢ cvl O'Le Z8i 681 oth 80 ke
8°CC c Ol 8'¢d L8t O'S} S'6 S'6 46
681 v6 ole 8'Sl 9h Z8 6L €¢
8h ML 8 Zi LS O'LL L8 8S ve
GOL ee OZLI O'?l Z8 fe ov Ge
O'Sl €9 vst otk 6L og Ge 9¢
cvl 9 6'cl c th €9 8g Gc le
OL €9 80 16 Gg ov Lay 82
Z8 LY S'6 61 OE JEN Da 6¢
6L May) L8 99 VC ee Ly 0€
€9 6E - 6Z 0's vc 6¢ cl Le
LY ve GL 0's 9} Le ol cf
Lav, Ve bas 9'V 91 Let ol €€
6€ 9 0S 9” 91 ok ar ve
Ve
Ge
0's
Gi!
cl
9}
tS;
Ve
9€
cl
ol
9
6¢
LES
9}
Ve ol LNs: 6? 9 Al! cl LE
91 80 LS; Go? 9} cl 80 8e
9} 80 ee Gg? 9} 80 80 6€
9} 00 Ge? jit 9} 80 v0 Or<
peeds Asowa paeds peeds Asowa uoieziuebio Did-DIA Aosuedaiosig
Buissa901g Burson Bulsss001g Buissas0i1g Buryiom jenjdad10q jo
—Aiows —uoneziuebio —uolsueyaidwoy —uoljeziuebiC —uolsuayaidwiog —uolsueyaidwioy junowy
a Bupjiom jenjdao190q jequan jenjdao18q [equan Jeqsan
SdX9PU]/SaTeIS
5
4 O21 < OISA *19497 ATqy Aq

372
= SdIOIS x9puy pue Ol III-SIVM U9x9M jog SIDUVII{fIq Jo (sasejus019g aArNemMuiny) sa1ouonbelj "a'd ajgel
 ‘TeuoTIaIp useq Asay] pey sa10os xapUT pue O] UeeMieq saduarayip ayy ayewrxodde qm Asay} ‘omy Aq paplalp are sanyea ay} J] ‘ajdures ayy Jo %z Aq pauteygo

373
Appendix D

SPM II pue O]d SNUIW OIA 0} pue OIA SNUIW Og 0} SiayaI syutod gz Jo aduaIayfIp & a[durexe 1Oj ‘[eUONIeIP-UOU aIe sar0S XepU] pue O] UseMjeq sedudIOFIpP 9YL
SS SSS SS SSS SSS
OL 06 O'El 00! O'lL 06 06 ueIps|/|
66 9°8 c Ol 66 06 G8 6L ds
LEE Olt 971 Vol cl 6°01 Ol ued
0001 0°00} 0001 0°00} 0°00} 0001 0001 0
9°46 6°96 6°26 2°96 6°96 8°86 126 L
9°06 0°68 A) ¢ 68 ¢ 88 € 16 8°88 C
8°68 0°S8 L28 8'e8 v8 0'€8 0'€8 S
v'L8 CLL cov8 0°82 L¢8 9 LL GGL v
ma 4°) cel 18 Lvl L418 StL Vibe G
Z8L S89 V8Z 0'eZ LOZ 6°89 ol9 9
Oe 7 v'19 LYl 0°89 6°99 V9 AKA’) Z
€'69 LSS BL Seg 0'€9 86S 86S 8
8'e9 y'0S OZ 0°95 9°09 LOS Ges 6
6SS 8°87 coL9 VS 6S LLv 87 Ol
0°¢S 6 VV 069 LLv ols Lov GLY LL
9°6Y Cev vy9S OY LSv LGE€ 9°8E cl
6 vb OLE 01S Ly 60v 9CE VCE el
c Ov v'SE 0'6V 6'9€ OLE Z0€ 0°62 vi
9°8E LOE 8 rY 9't OLE 662 Svc Sl
yVSE L'6¢ Sly ole co9e Ble Lee QI
ce 912 ele €'0€ ee L'9¢ 661 ZI
6°62 Z6h OVE 022 662 voc BL 8I
L'6¢ 68h S'6¢ L'9¢ oS¢ VL o9l 61
peeds Asowesy pseds poeds Aioweay uoneziuebio Did-DlA Asuedaiosig
Buissa20i1g Bupyiom Bulssal0id Bulsseo0i1d Bupiom jenjdasieg jo
—Mioway —uoeziuebio —uolsuayaidwog —uoneziuebio —uolsusyaidwog —uolsusyaidwog junowy
Bupiom jenjdasieq jeqiop, jenjdasieg Jeqispn jequsp
/S8TBIS
SIXIPU]
(panuyuos) =< 0ZT
Apqy *[eA9T OISA Aq
Usemjag
(Sadeyusd1eg
[[I-SIVM
OJ
sefuanbeiy
seduarajsIq
Jo
$3109¢
aATeyNUIND)
xepu]
pue
“gg
ajqe]
 ‘papnyoxa a1aM
Q Jo satouedarosiq ‘0SZ‘T = N 9104M “TW UM suostreduros 107 1d990x9 (GP‘z = N ‘ajdures UONezIpiepue}s []J-STV¥M Woy eyep UO paseq are san[eA ‘aJONy
a a a a

Sc-ve-
8S ev 6S ev 8'¢ 6°¢ 0'}
oS ve BY re 4 Le 60
Lv a ov ve 6} CC Z0
ve Ove ge L¢ egal 0% S0
Ove Le Se Go vk Gh G0
CC 0c « Oe L?é a! or €'0
b?é vt Le 8h eh LL c0
vil cl ve Gh cl 60 c0
vt 0} 8k sill 0} Z0 c0
ck 60 iE Ee 9°0 S0 20
0} Z0 9°} 60 G0 v0 0
60 Z0 cl Z0 S'0 c0 10
Z0 Z0 0} Z0 S'0 c0 LO
S°0 9°0 80 S0 S0 0 0
v0 c0 G0 v0 10 0 L'0
v0 c0 v0 €0 0 c0 LO 6e-
v0 c0 €0 €0 L'0 20 00 S$S9| Op_—10

junowy
peeds Auoway peeds peeds Aiowayy uoneziuebio Did-DlA
Aouedaiosig
Buissas0ig Bupjiom Buisse001g Buisse90/g Burpjion jenjdaoi9q

jo
Q —Aiowea; —uoleziuebio —uolsuayaidwog —uoleziuebio —uolsuayaidwog —uolsuaysidwog
4 Bupiom jenjdadieq Jeqis, jenydacieg jeqis, JEqis,
Ss
5 SdXIPU]/SITeIS

374
UIem)jeg
OI
$3109
II-SIVM
SalduLdaJIs
Salsuanbaly
pue
Xapu]
JO
[eUONIaIIG
3jqel
“g*g
=
 (panuyuod a
<= a a o = s =)

Buisses0ig
—Aioway
Bupyom
peeds
€9 ML 0°8 88 66 60L8 velv'Sl89 G6 O'l28°C?8'e?C82CLE0'CE9°SE
lh
xapuy sa1odg

—uojeziueBiC
jenjdesieg
Bupion
Aiowey
LY cS 8S OL LL v8 96 601ora!ctlvSl6 G6}LedLv¢eleS'8¢VCE
Lt
OJ pue
Ueemjeg [II-SIVM

—uojsueyaidwoa
Bulsseo0ig
G9 GL
Jeqisa peeds c8 S'6 SO}Gtkcelcvlc GLl8'6l61¢8'E?L'9¢L8¢9'0€8°c£0'SE
9k

SdXIPU]/SITeIS
safuedasosiq

—uoeziuebio
Bujsses0ig
jenjdaoi0g
peeds
es 0°9 Z9 L'8 c6 LOLOblvelSrl9 0'8t66 Ble6?0°92c 6'0€CVE
8¢

—uolsusyaidwog
[euoNsIaaGg

Bupiom
Aiowey
Jequsp,,
Le 0'V oF 0°9 OL 0°8 S'6 8'0l8 GelO'St99}SLLvie9°ce9°92el2oc’
bk
sajauanbeiy

—uolsusysidwog
uopeziuebio
jenjdesi9g
jeqisa
8'e ev c L029°?92L182VLE
L's 09 69 v8 S'6 Z0bSthO'elvt v9l6h
Jo

Old-DlA
Gh icaG7? ce LY LS eg sZ 68 G0lora!ctlv9l61 ee6'¥2ole6°08
age]"g"g

Aouedaiosig
jyunowy
jo
€e-Co—ko— 0¢-6}-8l-Z\-ie) Gl-vl-el cl—3 (0) 6- 8- Z- 9-

Jo= ‘papnyoxe
294M
UM
10} alam
sanye,
UoNnezIprepue}s
‘a]dures
Woy
paseq
sa1iouedarosiq
suostreduioo
[]J-STVM
ejep
‘azony
Q}daoxa
“0SZ‘T
N
‘TIM
GP'z
NV
UO
are

375
 “‘papnyoxe alam
0 Jo sajouedasosiq ‘0SZ'T = N ey ‘TAM UM suostedur0s 104 ydaoxa (GPr'z = N ‘a[dures uoneziprepueys []]-SIVM Woy eJep UO paseq are sone, ‘a/0N
ae
i,
co9h 9°12 ZL8k 061 iS)! 9'El cl
¢ 02
v6l Gee 6 0¢ L'be 9°81 0°91 bh
oC
vile o9¢ L'€¢ VEC Vie c 8k OL
Ove
8'E? 6 Le Lvc 1'S¢ 8'ec S02 6
8°92
9°9¢ 0'0€ 0°82 , © 8¢ G'9¢ L€¢ 8
€'62
8°6¢ 8°Ce 8'0€ c 0€ G'6¢ o9¢ Ls
GLE
L Ye OSE Gee 6°ce oe G'6¢ 9
Sve
LYE 0'8E c 9E 8°SE VSE 9°CE S
OLE
8°6E BLE 6 0V 8°8E Ov SBE 6SE Vv
coey Sl Sev 8 lL 6 vv GLY 6'6E €
LSP 0'SY 9°9v 6 vv co8V GOV Vev C
c 8Y Sev c 6V LLv AS) S'6v Gly L
oa = =. =y = cmp — 0
c 60 9°81 87 687 VSP 8ST 0'6Y L—
ely 6 SY cov L9V VLV 6 Ev 6 vv C—
EL Sly Sov 6cV c 6€ 8°6E oly e—
oly S'6E S'6E 9'6€ VLE Sle VLE y-
OLE O'SE 0°8E LOE LVE Les See G-
Asoway pseds peads Aioway uoneziuebio Dld-DIA Aouedaiosiq
pseds
Buryiom Buissad01d Buissad0/d Burson jenydao18q jo
Buissed0jd
—uoljeziuebio —uolsuaysidwog —uoljeziuebio —uolisuaysidwog —uolsusyaidwog qunowYy
= —Miowsw
Bunion jenydadi9d Jequs, jenjdad10q Jequan Jequs,
ia
Py
as)
SdXIPU]/SaTeIS

376
= (panu1juor) SI10IG Xapu] puke OI III-SIVM UIem)0g sapuedassiq jeuonseig jo saiuanbaiy *9"g 3jqgel
 (panuyjuo9) <= Ba
= =Fo f= cs =)

Buissao0ig
—-Mioway
Bunion
psaeds
9°8l8st6vlVelalt!90106 v8 VL OL 6S es oP ov 6€ 0 Gc
xapuy sa10dg

—uo1eziuebi9
jenjdao10q
Aioway
Suppo
cvlVelvclZ6 6 eZ el 0°9 LS oP cv ge ce L¢ Aird6} Gc}
OT pue
Ueemjeg III-SIVM

—uoisusysidwiog
Buissad0id
96hLl c9lLYlvelLZ
Jequa, peeds 80h6 v8 OL og cS 9” cv ve LG e?
bh
satouedasosiq

—uoljeziuebio
Buissa90ig
jenydao10q
peeds
8'9lvt cv vclGb ome)o6 8 ee G9 es VY 0'V ve 0'e Ve CG

SdXIPU]/SaTVIS
—uolsusyaidwog
[euoTdaaIq

Asowesy
Bupion
Jequs,
9°919'[Link]€'Ol9°8 VL og gg 8'V cv ee L€ ve CC 8} oh
salsuanba1y

—uo!susysidwiog
uoneziuebio
jenydad18q
jequsp
srlGelc 00 €'6 0°8 oL 8S vs CoV LY L'€ Le 0c 8h Gh batt
th
Jo

Old-OlA
8 66 Z8 BL G9 9S Sy LY ve Ss 9°¢ L? [Lieee bab Z0 9°0
lt
age] “g"g

“pepnypoxe
Asuedaiosig
junowy
jo el vi Sl Ol ZI 8 6h 0¢ ké cc x4 ve GS? 92 Le 82 62
S°0
vl
VC
6}
0€
Ol
0} Pk yi)
L? ica
ar
Le
S'0
Ol
80
cap oak
ct
60
sl
ce
v0
Z0
alam

Jo
e404
YIM
10j
UoNezIpiepue}s
‘a[]duies
elep
WO
pase
sanye,
salouedainsiq
‘TIM
Nsuostreduios
[]J-STVM
‘A707
Q}daoxe
‘0SZ‘I
N
OGP‘z
UO
are
=

377
 ‘papnyjoxe alam
0 Jo satouedarosiq ‘0S7'T = N 204M ‘TAM IM suostreduros 103 1daoxe 0Gr‘z = N ‘e[|dues uoNezIprepue}s []I-SIVM Woy eJep UO pase are sanje, ‘a0
a
v0 LO G0 €'0 v0 m0) 00 aJOW 10 OV
S‘0 c0 9°0 v0 v0 10 LO 6€
9°0 c0 9°0 S0 v0 0 0 8€
80 c0 9°0 G0 v0 c0 c0 LE
0} €0 Z0 G0 v0 c0 0 9€
OL €0 60 Z0 v0 €'0 0 ce
ck v0 Ol 80 9°0 v0 c0 ve
vk Z0 eh 60 9°0 G0 v0 e€
pseds Asoway pseds peeds Aiowsy uoneziuebio Old-DlA Asuedaiosig
Bulsses0ig Bupiom Buissas01g Buissel0i1g Bupon jenjdao1eqg jo
—Aioway —uoljeziuebio —uolsuaysidwog —uoljeziuebio —uolsuayeidwog —uolsuayaidwog junowy
Appendix D

Bunion jenydaoseq exe PLYAN jenjdaoi8g 1eqisn jeqisn

SdX9PU]/SITBIS

378
(panu1juod) $9109 xapu] pue OI [[J-SIVM UseM)eg SafPULdatDSIG [eUOTIeIIG JO SaIsuUaNbaly -9"g ajgqey
 aigqel*/"g feuonsenqg aapemumy safuenbary
410j []I-SIM
Xapuy 9109¢ SEdUdIAIIC
pny —wuy] pny —wuw| SIA WU] pny -j9q pny -jeq
= —-Wweyww]
ww) -way
3
——
-wayuey284 -jeq
soussayigSIA WL] pny 1eq SIA 19 190 9ey SIA 19G UAH Wa= JOM WW= OM We SIA 19d
Ob- OL 00 00 00 9'0 00 £0 v'0 90
6e- vb 00 00 0'0 8'0 00 s‘0 90 £0
8e- vt 00 0'0 00 OL 00 9'0 £0 vk
Le rh 00 00 00 OL 00 60 bh vk
9e- ol 00 00 00 el 00 OL Zk rh
Se- 0% 00 00 me) vk Lo Zk vk Lt
ve Ze zo 00 z'0 gL LO pL gL Ze
ee- 92 zo 00 v0 sz 10 8'L el sz
cee Le zo 0'0 s‘0 sz LO (ord 'L ee
Le- Le z0 00 90 Gz me) v2 ez ge
oe- ze zo z0 £0 ve Lo 92 sz ge
6c- 9 z0 z0 ome BE m0) ve oe oF
8z- ge re) z'0 a Le ae) o'r Be o's
Vir o's zo z0 come ZY LO vy ZY a's
9z- 6S zo z0 rd zs me) o's Sy LL
Se- €9 z0 z0 ze oe) 10 z'9 a's ee
ve- Be zo v'0 Ved rl 0 rl Zz9 98
rs €'6 z0 $0 6z 88 LO ze SL v6
2e- 86 90 O'L ge 06 z0 6'8 28 gO
Le ral 60 gL ran 96 v0 86 6 ell
oz- ZL ol ro 6Y Pur s'0 Lb vi LOL ZL
61 vel OL sl v9 lL 6 80 Lek ZL Zh
8l- orl el 92 SL ZL Zh Bel Sel est
Le L'6L roam 92 96 cme vk o's Ly EOL
OL e'6L ez ze 90 L'SL Lk ash O'9L esl
Sit 8'6L ve or O'LL 9S oz e'6L vel O'6L
vile o'ez ay or OeL 8'6L ES s'0z e'6L Z'02
ee Lee vs VS rr £02 ee Lez v2 92
rae zve 99 $8 EEE ele SY Lve Bez Lye
bl 022 GL $8 L6L S92 zs Z'92 2'S72 €'92
OL- lz 06 vO Elz 9°92 eZ $82 wkd L6z
6- Z'8z Sz SSL eaord vl OO z'0€ Loe rile
8- Lee VSL Z9L Sr Z2e kee zze sze Lze
& zee Zl ZLL ras zze SOL ove ese 6'SE
i Sve 922 z92 bee Zve 86h Lge re oze
S- Sly zse 92s “se 9'0r ve 6 z'6e 0'6e Lee
Be Sly Sz o2z rly 60r 262 Sly Zev ely
se ber 6Se z6e zer Slr ve 6 Lpy o'Sr Ler
Ze 8'er Sse Lor vor 69r z0r Lor OLr Lov
[= 9'6r 6'8e Z'0r S'6r o'er had ver eer 6'6r
) _ _ _ - - _ a -- os

379
Appendix D
 0) 0 ) 0 0 0 0 L Ov
0)
0 0 0 L 0 (0) 0 L 6E
L
L ) 0 L 0 0 (0) L se
L
L L 0 L ) 0 0) L Ze
L
i L 0) i 0) 0 0 I 9€
L
L L 0 iL 0) 0 0) zZ Se
L
L L 0 L 0) 0 9) zZ ve
rAd
zZ IL 0 Zz ) (0) 0 z ee
zZ
rd a ) zZ (0) 0 () € Ze
ra
Zz zZ 0 zZ 0) 0 0 € Le
€
y zZ Zz 0 zZ L 0) 0 € oe
re € ) € IL 0 0 € 62
v
S € € ) € L 0 0 v ez
S v v 0 v I 0) 0 v le
9 v v 00 v zZ 0 0 S 92
v S 00 S zZ 0) 00 9 Gz
8
8 9 9 00 9 zZ ) z0 9 ve
6 I Z 00 b € v0 z0 8 €z
6 8 8 LO 6 ‘ € s‘0 zo 6 raed
LL 8 6 z0 6 v OL 90 OL 1?
LL 6 OL v0 OL v OL 60 LL 0z
ex = LL LL sO Zt S OL OL el 6L
vL el el 80 Zt 9 es ol €L gL
Sl ve vk Zh el 8 92 el SL fa}
9k St St vik St OL 92 GL 6L 9L
cae 9L oy ay St 9'0L Ze e~2 6L St
6 gt 6L o7 roe OLE oF ve 02 tL
(Ye 6L 0z Le 0z Oe or ov €z el
€z lz zz ee ka vv bs vs ve Zt
Gz ve Gz cy 2 LLt S38 99 v2 LL
92 ze 92 zs G92 L6L 38 GL le OL
62 gle S82 el 9°92 ele vol 06 ez 6
Le LOE Z0€ OO vle sitar GSt cel 8z 8
Lee Sze ze Lob eA Sve ZO LSL ee yy
6SE ese Ove remey Zee ZLe Gay “gt zee 9
oze ve L9€ 86l Zve Lee z92 922 sve S
L8e 06 Z6 62 9°0r Lge Gg92 ase Sly v
ely Zev Sit z 6c 60r vy o-22 ve Slr €

S9dIUdIIIJIG
Lev ost ag 6 ve SLY Zev Z6E 6SE Ler zZ

(panu1juod)
Sse eer L

sapuanbaly
L'9r Z Or 69r vor L'Or

saAnelnunyD
Lov OLY

jeuOHselg
Appendix D

19d SIA Weil 440M WAIN 440M Wei Ue 19d SIA 99H 19d 190 SIA leq phy Wi] SIA SIUdISHIG

[TI-SINM
-|9q 99y —weyy uay = —Wway) wwy = Way Wu] -l2q pny -|9q pny —WUW] SIA —ww] pny —wlw| pny
eT

3ajgel
X9pU]
3109S

"f"g

380
10J
 APPENDIX E

WAIS-ill Digit Span Backward

Scaled-Score Equivalents
of Raw Scores
 yl-lL vI-Cl pI-el pl-ClL vi-el vt rk - - ~ _ — _ 6L
ZI-LL = ok rat Zh rAl el el rh vk vl rk rh rl gL
OL im im im im ral ra el el el el el el Zt
6 Ol Ol -- — im LL ral All rAl ral oe rAl QL
_ 6 6 Ol Ol os —- ‘t LL LI im LL im SI
8 8 _ 6 _ Ol Ol OL Ol ~ _ Ol OL rk
Z — 8 8 6 6 6 6 6 OL OL 6 6 El
~ Z Z l 8 g ) 8 8 6 6 — _ ral
9 9 9 -- l l - _ 8 g g 8 LL
— — _ 9 9 9 l l Z Z l Z l OL
s S S g S S 9 9 9 9 9 9 9 6
v -- _ o- — g S S S _ ~ — 8
— p p v v v v - — ~ g S S Z
€ € € € € € € v v v v v v 9
Z Z Z Z Z Z — € € c € € € S
L | | L | | Z Z Z Z Z Z Z v
0 0 0 0 0 0 | | | | | | 1-0 €
— — _ — — = 0 0 0 0 0 0 — Z
— — — — — — — — —_— — — — — L
Appendix E

68-S8 8-08 62-SL vlZ-0L 69-S9 9-SS ~S-Sh tr-SE EOE 62-S2 2-02 G6I-8l LI-9l 2109S
ee See es ETT
ee S:C:*«YES
aby
dnoiy

382
$a109S Mey JO S}UaTeAINbY 3109$-payessg premyseg ueds 131g III-SIVM “L"9 3/geL
 APPENDIX F

Inclusion and Exclusion Criteria for

Participation in Special Group Studies

General Exclusion Criteria

Potential participants were excluded from the studies if they met any of the
following criteria:
a vision or hearing impairment that precluded valid assessment,
previous completion of the WMS-R or the WAIS-R within 8 weeks of
testing,
knowledge of English insufficient to ensure valid test results,
evidence of receptive or expressive aphasia of a severity that would
interfere with testing,
concurrent substance or alcohol abuse (except for Alcohol-Related
Disorders groups),
disruptive behavior or insufficient compliance to ensure valid testing,
upper extremity disability that would significantly affect motor
performance during testing, or
a history of psychotic disorder(s) (except for the Neuropsychiatric
Disorders Group).

Neurological Disorders
Alzheimer’s Disease
Participants with mild symptoms were included in the study if they met the
following criteria:
NINCDS-ADRDA diagnostic criteria for probable Alzheimer's disease,
outpatient residential status,
age of 60-80 years,
a score of 18 or less on the Beck Depression Inventory (BDI; Beck & Steer,
&
1987) or ascore less than 15 on the Geriatric Depression Scale (Sheikh
Yesavage, 1986; Yesavage et al., 1983), and
383
Appendix F

* ascore of >95 on the Dementia Rating Scale (Mattis, 1988) or a score

of 18-23 on the Mini-Mental State Test (Folstein, Folstein, & McHugh,
T97a)2

Potential participants were excluded from the study if they met any of the
General Exclusion Criteria.

Huntington’s Disease
Participation criteria included
a family history of Huntington’s disease;
evidence of decline from a previous level of social or occupational
functioning;
presence of an upper-extremity motor disorder, typically manifested by
eye-movement abnormalities, involuntary choreiform movements, or
impaired initiation of voluntary movement;
presence of a cognitive abnormality, typically manifested by impair-
ments of initiation, executive control, attention, and learning;
confirmation of caudate atrophy via magnetic resonance imaging
(when available);

age of 18 years or more; and

diagnosis confirmed by presence of an expanded trinucleotide CAG
repeat in chromosome 4 (namely, repeat number greater than 40) and
Stage I, II, or II of Huntington's disease (total functioning capacity
between 3 and 13, inclusive of the Unified Huntington’s Disease Rating
Scale (Kieburtz, Penney, Como, Ranch, & Shoulson, 1996).
Participants who otherwise would have met the criteria for the study were
excluded if they met any of the General Exclusion Criteria or any of the fol-
lowing criteria:
a history of serious head injury as determined by (a) loss of conscious-
ness (LOC) for more than 20 minutes or (b) neurological or behavioral
sequel of trauma to the brain, or
current major depressive episode.

384
 Appendix F

Parkinson’s Disease
All participants were clinically diagnosed by neurologists and were included
in the study on the basis of the following criteria:
diagnosis by two independent neurologists based on two or more
cardinal signs of Parkinson's disease (i.e., rigidity, bradykinesia, resting
tremors, clinical responsiveness to levodopa as judged by attending
neurologist);
cognitive impairment demonstrated by a score of 90-125 on the DRS
(Mattis, 1988); and
a score of less than 18 on the BDI (Beck & Steer, 1987) or a score less
than 21 on the Geriatric Depression Scale (Sheikh & Yesavage, 1986;
Yesavage et al., 1983), that is, scores that are below the cutoff value for
mild depression or dysthymia.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

Traumatic Brain Injury

Participants in this study met the following inclusion criteria:
an initial score lower than 13 on the Glasgow Coma Scale (Jennett &
Bond, 1975; Teasdale & Jennett, 1974),
age of 16-65 years,
occurrence of injury 6-18 months prior to testing,
loss of consciousness for at least 60 minutes, and
primary language of English.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria or if any of the following conditions existed:
continuing posttraumatic amnesia;
past or present medical history significant for symptoms of deficits .
that could significantly affect test scores (e.g., previous traumatic brain
injury, cervical quadriparesis, history of learning disability, current
substance abuse); or
of
severe ataxia, tremors, or motor impairment precluding construction
subtest of the
3 x 3 configurations of blocks (as in the Block Design
WAIS-III).

385
Appendix F

Multiple Sclerosis
Participants in this study met the following inclusion criteria:
¢ probable or definite MS (C. Poser, Poser, & Paty, 1984; S. Poser at al.,
1986),
¢ age of 21-55 years, and
¢ ascore <8 on the Kurtzke disability scale (Kurtzke, 1951, 1983).
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

Temporal Lobe Epilepsy

Participation criteria for this study included
e evidence of lobectomy postoperative clinical improvement in seizure
disorder,
¢ age of 16-60 years,
¢ an FSIQ score greater than 70,
¢ surgery 1-24 months prior to testing, and
e right or left speech dominance.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

386
 Appendix F

Alcohol-Related Disorders
Chronic Alcohol Abuse
Participants in this study met the following inclusion criteria:
DSM-IV criteria for alcohol abuse,
3 weeks of sobriety prior to testing,
a VIQ score of at least 80,
intact attention and language,
inpatient residential status,
a score less than 18 on the BDI (Beck & Steer, 1987),

age of 35-55 years,

from 9 to 16 years of education,
absence of seizures and no intake of seizure medications, and
a history of at least 2 years of alcohol abuse.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

Korsakoff’s Syndrome
Participation criteria for this study included
age of 53-74 years;
previous diagnosis of Korsakoff’s syndrome by neurological and neuro-
psychological testing and, in some cases, by radiological confirmation;
attention span within normal limits; and
no history of visual—perceptual or language disorders.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

387
Appendix F

Neuropsychiatric Disorders: Schizophrenia

Participants with schizophrenia met the following inclusion criteria:
¢ full DSM-IV diagnostic criteria for schizophrenia of any subtype,
¢ outpatient status,

e age of 18-65 years, and

¢ clinical stability.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

Psychoeducational and
Developmental Disorders
Mental Retardation
All participants for this WAIS-III study were recruited from private or public
facilities according to the following inclusion criteria:
¢ for the group with mild mental retardation, an FSIQ score of 55-70 and
VIQ and PIQ scores <70 on a standardized intelligence test other than
the WAIS-III; for the group with moderate mental retardation, an FSIQ
score of 35-55 and VIQ and PIQ scores <54 on a standardized intelli-
gence test other than the WAIS-III;
* scores on the Vineland Adaptive Behavior Scales (Sparrow, Balla, &
Cicchetti, 1984) consistent with the disorder, that is, a score of 55-65 for
the group with mild mental retardation and a score <55 for the group
with moderate mental retardation;
¢ age of onset earlier than 18 years; and
* no known acquired brain damage.
Participants were excluded from the sample if they met any of the General
Exclusion Criteria.

388
 Appendix F

Attention-Deficit/Hyperactivity Disorder
The participants in these studies met the following inclusion criteria:
e diagnosis of ADHD according to the DSM-IV criteria,
e an FSIQ score >90,

¢ adiscrepancy of <15 points between ability and achievement scores,

e age of 16-24 years, and
¢ normal vision and hearing.
Participants were excluded from this study if they met any of the General
Exclusion Criteria or any of the following criteria:
* concurrent psychopathology such as depressive disorders, anxiety
disorders, or conduct disorder; or
¢ aprevious diagnosis of a neurological disorder, such as epilepsy, brain
tumor, or head injury.

Learning Disabilities
For both the WAIS-III and WMS-III studies, participants met the following
criteria:
* aprevious DSM-IV diagnosis of a learning disability, with documented
supporting data, including specific test scores;
¢ an FSIQ score >90 on tests other than WAIS-III;
and
¢ a discrepancy >15 between measures of ability and achievement;
¢ age of 16—24 years.
of the General
Participants were excluded from this study if they met any
Exclusion Criteria or any of the followi ng criteria :
disorders, anxiety
¢ concurrent psychopathology such as depressive
disorders, or conduct disorders; or
such as epilepsy, brain
oa previous diagnosis of a neurological disorder,
tumor, or head injury.

389
Appendix F

Deaf and Hearing Impaired

Participants for this study were recruited according to the following criteria:
¢ no disabilities or impairments other than deafness or hearing
impairment,
¢« American Sign Language (ASL) considered by participant as primary
language, and
¢ age of 18-65 years.
Participants were excluded if they met any of the General Exclusion
Criteria.

390
 APPENDIX G

Examiners, Reviewers, and

Participating Clinics and
Organizations

Examiners
Elizabeth Abraham, MS Katherine Bell, MA
Catherine Acuff, PhD Betsy Benson, PhD
Michelle Adams, MA Gary D. Berger, MEd
Shirley A. Albertson Owens, PhD Karen I. Berland, PsyD
Sandra Alexander, MA Pelagie Besson, PhD
Lucy Allen, MA Ruth Bewley, PhD
Ronald O. Allen, EdD Julian Biller, EdS
Linda A. Allen-Clay, MS Thomas W. Bishop, MA
Diana Allensworth, MA Karen Blackwell, MA
Amy Amarello Sanford, PsyD Jonathon W. Blaine, BA
Cynthia Andrews, MEd Deborah G. Blair, PsyD
Kimberly Anthony, MS Eadye Bollinger, MA
Patricia Antonelli, MEd Mark A. Bolton, MA
Trinidad Arguelles, MS Tamara Y. Boney, MS
Kara Arman, BS Ray Booth, PhD
Jeffrey Armstrong, EdS Mary Borders, MA
Lois Armstrong, PhD Barbara Bordner, BA
Brenda Arrington, EdS Robin Boren, PhD
Michelle Austin, MEd Monica K. Borinstein, MS
Leslie Baker, MEd Tom Bottenfield, MA
Elvyn Barrable, MA H. Marie Boultinghouse, EdD
Mary Barrows, EdS Sara Bourque, EdD
Michael Basso, PhD Jan Boyle, EdS
Patricia Bates, MS Kathryn Bradford, MS
Robert Bauste, MS Mary Brant, PhD
Bonnie Nash Bawel, EdS Karen Brewer, PhD
Trish Beach Thomas Brewer, BS
Letitia Bean, MA Renee Briggs, PhD
Elaine Beckwith, MS Barbara Brinson, MA
Roberta G. Beeler, MEd Sekai Broaden, MEd
Appendix G

Charles Broadfield, PhD Tim Crimmins, M.A.

Yvonne Brooks, EdD Leslie L. Crossman, PhD
Jane Brown, MS Anne S. Culp, EdS
Mary Brown Marcia B. Cunningham, MEd
Thomas Brown, PhD Virginia L. Curulla, PhD
Joanne Browne, PhD Patricia Kaiser Cutulle, BA
Kenneth M. Browner, BA Wendy E. Cwinar, MEd
Jerome Bruns, PhD Diane L. D’Agostino, MA
Corby Bubp, BA Gail Dahl, BS
Robert Buckner, EdD Gary M. Daily, MA
Don Cabell, PhD Christina Darby, BA
Glenn E. Cahn, PhD Helen Darks, EdD
Thomas C. Caldwell, MA June Fox Davis, EdS
Doris Callands, MEd Melonee Davis, EdS
Maximo J. Callao, PhD Stephanie Day, MS
Carol Ann Calney, MEd Debbie C. De Berry, EdS
Lucille A. Cardella, MEd Jody H. De La Pena, BS
Kenneth Carpinelli, MS Maria D. De La Sierra, MA
Marta Carrasco, MS Judith R. Defeo, BS
Della Carter, MS Milton J. Dehn, EdD
Gloria G. Casanave, PsyD Christa Dell, MA
Audrey L. Cercelle, EdS D. Denard
Barbara D. Chaplik, PhD Doina Denes, MA
Cheryl Charis Graves, MS Lynn Dennis, MA
Madhu Chaturvedi, MA Jeffrey Dersh, MA
Kathy Chauncey, MA Michelle D. DiGiovanni, MA
Lisa Childs, MS Joseph G. DiRaddio, MA
Paul G. Chrustowski, MA Nancy Dodge, PhD
Toni Cicerello, MA Paul Donecker, MS
Richard A. Clark, MA Becca Dotson, MA
Paul G. Clements, PhD Alan Dryden
Richard B. Cluff, PhD Sharon Durkin, MA
Michelle Coffman, EdS Pamela Eckard, EdD
Anat Cohen Rosman, PhD Steve Eckert, MA
Marija Colic-Turcinov, MA Jan Eckman, MA
James Collier Debra Eddy, EdS
John Consalvi, Jr., MA Oliver W. Edwards, EdD
Mary Lynn Cooper, MA Marjy Ehmer, PhD
Amy Cosby, BS Martha Eichenlaub, MEd
Barbara Couvadelli, MA Cynthia L. Eland, MA
Leon D. Cox, MA Judy E. Elkins, MS
Anita Craft, PhD Bonnie Ellefsen
Jason Craggs, BA Lisa Elliott, MS

392
 Appendix G

Marge E. Everhart, MA Gail Greenberg, MA

Laura M. Fairfax, PhD Karen Greep, MEd
Donna Fantozzi, MEd Deborah Grisham-Blair, PsyD
Deborah Farrell Coleman, BS Deborah Gussak, MA
Elias Fernandez, MA William A. Haas, MEd
Randy Fingerhut, MS Christine E. Hack, MEd
MarkW. Finkelstein, MA Catherine J. Hadden, EdS
Sandra Firth, BS Kristin Hagy, MA
Debbie Fishman, BA James Hale, MA
Ruth Fletes-Fonseca, BA Nancy G. Hale, BA
Ruth Fodness, MA Jill Hall, MA
Ann L. Foreman, MA, MEd DavidW. Hamilton, MA
Laura Guthermuth Foster, PhD Joanne Hamilton, BS
Mary Lou Francis, MA Debra K. Hamm, MA
Nicholas Fratto, EdD Cheryl T. Hammond, MA
Theresa Frazer, MS Holly Hancock, MA
Lesa A. Frazier, MA Lucita Hanlin, MS
Michael J. Furhman, PhD Michael Hans, PhD
Stacey L. Gabriel, EdS Kathleen Hanson, MA
Eugenio J. Galindro, MA Sherry L. Harden, MA
Kathryn Garrett, PhD David Hardy, PhD
Haley Gaskell, MA Cynthia M. Harpenau, BA
Karen Gavin, MS Bea Harris, PhD
Paula Gebauer, MA Josette Harris, PhD
Ellen A. Gertz, EdD Janice N. Harrison, PhD
Kathleen Gilbert, MA Teresa M. Hart, PhD
Alberta Gilinsky, PhD Teresa Hatfield, MS
Melisha Gilreath, EdS Edwina J. Hawes, PhD
Nancy Gimbert Fritze, MA Tamara L. Hazelton, MA
Christine Girard Amy Heefner, MS
Margaret Jo L. Glaser, PhD E. Lynn Heeren, PhD
Deanne R. Goben, MS Bertha Henderson, MS
Robert Godsall Brenda A. Henderson, PhD
Jim Gold, PhD Kara M. Hendry, MA
Greg Golden, MS Dianne P. Hengst, MA
Heather Goldman, EdS William Henry MC
Jacob Goldstein, MS Ann Hershberg, MS
Claudia Goleburn, MS Wilson Hess, PhD
Ana Gomez, PsyD Carol Hickam
George Gonzalez, MA John Robert S. Higgins, EdD
Mercedes Graf, EdD Karen Hike, MA
Janice J. Graham, MEd Fiona Hill, MA
Dee Ann Grant, BS Thomas E Hill, MA

393
Appendix G

Beverly Hime, MA Robin Knoblach, PhD

Cindy Hogue, MEd Richard D. Koehn, MA
Lynn Hohrmann, MS Ronald Komers, MA
Jill Hoilien, MS Brenda D. Kosaka, PhD
Frederick Holley, BA Christine E. Kostrubala, MA
Patricia D. Hollinger, MS Marlene Krupa
Margaret A. Hooks, MA John A. Kupoinski, EdS
Carole Hooven, MEd ‘Robert A. Kutner, PsyD
Loretta Houck, MA Laura Lacritz, PhD
Linder G. Howze, MS Karen Ladd, PhD
David G. Hull, MA MatthewV. LaGrange, PhD
Lori Nikkel Hurtik, MA Robert J. Lamparello, BA
Margaret A. Hutmacher, MEd Rick J. LaMura, EdS
Chris Huzinec Dee Langley, MEd
Robert Huzinec, MA Rayna P. Larson, MEd
Virginia Iannone, PhD Sue Larson, PhD
Nancy A. M. Ingwell, PhD Atlas Laster, Jr., PhD
Maureen Innes, EdS Janice E Lawrence, EdD
S. Mohammed Iqbal, PhD Harry H. Lawson, PhD
Patricia A. Isopo, MEd Carla Lee
Ryan D. Jaarsma, MA Michael Leland, PsyD
Calvin C. Jackson, MA Susan Leonard, MA
Clare Jacobs Carol J. Lepera, BS
Estelle Jasnoff, EdD Michael H. Levine, EdS
Jean Jellema, MA Brian Levitt
Norma O. Jenkins, EdS David Libon
Tom Jenkins, EdS Maria Deinzer Lifrak, PhD
Wendy Jerred, MA Joe A. Lipetzky, PsyD
Carrie H. Johnson, MA Myra Little, MS
Renee Johnson-Shelley, EdS Gregory Littlejohn, MEd
Shirley Machocky Jones, MA Jerry Livesay, PhD
Devonna K. Jonsson, BS Catherine B. Lochner
Timothy J. Jovick, PhD Constance Locraft, PhD
Giselle Juneau, BA Damond J. Logsdon, MA
Lloyd J. Kallial, PhD Shane Lopez, MS
Sharon S. Kaufman, MA Andrea V. Lorkowski, MS
Sandra Kazor, MS Stephen L. Loughhead, PhD
Laurence Merrill Kelly, EdD Jodi L. Lowther, MA
Mary Ann Kelly, MEd Dolores R. Ludwig, MA
Julie Kibler-Karl, MA Karen Luque, PsyD
Brenda K. Kilpatrick, MA John P. Lutchko, EdS
Edward Kittinger,Jr., MA Marlene Lyman, EdS
Jeffery S. Kixmiller, PhD Robert Lynch, MA

394
 Appendix G

Phyllis MacCortney, PhD Roy E Morgan, MA

Richard Mace, PhD Linda Morris, MA
Linda Mack A. K. Morrison, MEd
Michael Maclean, MA Toby Motycka, PsyD
Mary J. Macys, PhD Dan Mungas, PhD
Juliette M. Madigan, MEd Sandra P. Munoz, MS
Sandra D. Mahoney, PhD Alycia L. Muto, EdS
Lori Manade, MA Donna Nallett
Gail H. Reichman Mancini, MA Howard Nathan, PhD
Bill Maniago, MS Myra Nathan, PhD
Mariano Maqueda, MS Nancy L. Naveaux, MS
Edward Marshall, MA Deborah Nemit, MEd
Beth Anne B. Martin, PhD Charles Nguyen, MA
James M. Martin, EdD Deanne Nolte, MA
Julie Martin, MA Michael Nomikos, MA
Patricia Martin-Carr, MEd Cindy Nordlund, MA
Janet Martin-Day, EdS Mark D. Nordlund, MA
Sherri Matkovich, EdS Michael Norris, PhD
Elaine S. Max, EdS Sally O’Connor, MEd
C. Jill McClanahan, EdS Margaret O’Grady, PhD
Shelly McCoy, MEd Stephen L. O’Keefe, PhD
Robert B. McCue I, PsyD Deborah O’Meara, MA
Lyn McDonald, MA Cynthia A. Olson, MS
Patricia McGarrey, PhD Bruce J. Oppenheimer, MA
Karen McGee, MA Peter Orlando, MEd
Patricia McGinty, PhD Emily G. Osgood, MS
Anne P. McGloin, MA Terry P. Overton, EdD
Michael McGrath, PhD Linda Page, MA
Paige Davis McGuire, MA Kimberly Palko
Kathleen McKean, MS Gunda Jacobson Palmer, MEd
Sandra McKinnis, MA Anthony Paolitto
Thomas Meidinger, MS Sylvia V. Parga, MEd
Brad E Meier, MS Maxine Parvin, MA
Lewis H. Meltzer, MEd Denise Peloquin, MS
Carol A. Micalizzi, MEd Yolanda Perez, MA
Nancy L. Michael, EdS Deborah M. Perry, MS
Dimaris E. Michalek, EdD Rosario C. Pesce, PhD
Gail A. Mills Bigham, MA Robert Peterson, MEd
Susan Mitchell, MEd Rosemary Peterson, MS
Carol Mongar, MA Christina Petofi-Casal, MA
Cristal T. Moore, MA Marie R. Petrie, MS
Kathryn L. Moore, MS Warren Phillips, PhD
Luz-Martha Moore, EdS Paulette G. Pilsner, MEd

395
Appendix G

Lorine E. Pitter, MEd Robin Satchell, EdS

Tina D. Ploof, MA Fausta M. Satterlee, MA
Norman Pomerantz, EdD William H. Savage, MA
E. Jeanne Pound, EdS Jeanne Schillaci, MA
Kathleen Powers, MA Linda Schmechel, PhD
Phillip C. Pratt, EdS Reva Schwartz, MEd
Paula Precht, MEd ’ Elizabeth Ann Scott, EdS
Anne L. Price, MS John M. Sebben, BA
Mitch Prinstein, MS Bryan Senn, BA
Tanya Quille, PhD Scott Senn, BS
Nancy D. Ramirez, MA Emanuel Shapiro, MA
Julia M. Ramos-Grenier, PhD Michael E Shaughnessy, PhD
Darcie Randleman, EdS Judy Shaw, BA
Dorothy M. Rasener, MA Patricia T. Shea, MS
Lisa M. Raufeisen, MS Glenn P. Shell, MEd
Judith A. Reaven, PhD Julie Shifley, MA
Karen Reese, MA Cleatta Jackson Shumate, MA
Joanne Regina, PhD John M. Siebel, PhD
Maryellen H. Reid, EdS Ruby Simmons, MA
Laura Rencher, MS James Simonds, PhD
Adriana Restrepo, EdS Shirley Simpson, MS
Jean Reynolds, EdS Preston Sims, PhD
Dorothy Rhodes, MEd Darshan Singh, EdD
Diana M. Richardson, MA Michael E. Siyufy, MS
Robert Riedel, PhD Heidi Smaltz, MA
Jonathan E Rightmyer, PhD Barbara Smith, MA
Cheryl Robbins, EdS Billie Smithson, MA
Larry Roberts, MS Tara Sommers, MA
Marva Roberts, MA Linda A. Soucek, PsyD
Laurie K. Robinson, BA Susan Sperry, MA
Nancy Robinson, MA Kenneth Stanton, MA
Carolyn Rodriguez, PhD Pamela A. Stein, PhD
Jose Rodriguez Naomi Steinberg, PhD
Ann Romer, EdS Roy D. Steinberg, PhD
Lane Roosa, EdD Barbara R. Sterin, EdS
Audrey A. Rosenberg, PhD Diane Sterling, MA
Vera Rosenhand, PhD Sara Stevenson, MA
Maryann R. Roth, MEd Randall J. Stiles, PhD
Emily Rummel, MA Eloise C. Stoehr, MS
Tamara L. Russell, EdS Robert J. Stoever, EdS
Mark Rutledge-Gorman, PsyD Michael L. Stranathon, MA
G. Nohl Sandall, PhD Gary K. Sturgill, PhD
Dave Sanford, PhD Mary B. Summerville, PhD

396
 Appendix G

Christine Svetina, MA Shirley M. Warford, MEd

Margaret R. Swailes, MS Valarie Warmflash, PhD
Charles W. Szasz, EdS Barry A. Wartenberg, MA
Joseph Szyszko Fiona E Weekes, MA
Scott Talbert, BS Fredric J. Weiner, EdD
James M. Talone, PhD Sally A. Weisman, MS
Florence Tam, PsyD Patricia Weiss, EdD
Elizabeth Tamborella, EdD Cecily Weistein, MA
Claire Tarte, PhD Sylvia Weisz, MEd
Teresa Taylor, MA Patricia J. Wellman, MEd
Nan Taylor-Balser, PhD Kim R. Welsh, EdS
David C. Terjanian, MEd Marc U. Wenzel, MA
Terry L. Thatcher, MS Jim G. White, EdS
Elizabeth Thomas, MS Kristin Wiens, MA
Judy Thompson, MA Martin J. Wiese, PhD
Kathi Y. Thompson, MS Barbara Wilkerson
Gary D. Tolman, MS Beatrice B. Wilkins, MEd
Konnie Torbahn, MA Greg Williams
Carolyn Truesdale, MA Judy Wolfram, MA
Connie Tucker, MS Dennis Wood, PhD
Gretchen Tucker, MA Walter M. Wood, EdD
Alan G. Tuft, PhD John L. Woodard, PhD
Helen M. Tulsky, MA John Woodland, MA
Jennifer Turner, MEd William H. Worrall, MA
Julieanne Turnley, EdS Gloria B. Wuhl, PhD
M. Eron Tworetzky, MA Tim Wynkoop, PhD
Jan M. Ueckert, MEd Terri B. Yerman, MA
Hiromi Unno, MA Jennifer L. Yount, MA
Karen J. VanHandel, MS Margaret Zabel, EdS
Dale J. Veith, MS
Sister Sue Verbiscus SC
David Vesel, MA
Nicole Vincent, MS
Elizabeth A. Vosper, MEd
Brian Wagner, MS
Sharon Wakefield-Brown, MEd
Marie Walbridge, PhD
Robin Walker, BA
Robert Walkow, MEd
Louise Walsh, MA
Amy Sanford Walters, PhD
Robert Walters, PsyD
Victoria Ware, EdS

397
Appendix G

Validity Studies Cooperators

Bradley N. Axelrod, PhD Jeffery S. Kixmiller, PhD
William B. Barr, PhD Kathleen Knee, PhD
James T. Becker, PhD Christine E. Kostrubala, MA
Robert Bornstein, PhD Maria Deinzer Lifrak, PhD
Rosemary Bowler, PhD - Thomas Lozinski, PhD
Jeffrey Braden, PhD Donald Marion MD
Thomas Brown , PhD Dan Mungas, PhD
Robert Buchanan MD Anthony Paolo, PhD
Meryl Butters, PhD Marie R. Petrie MS
Gordon Chelune, PhD Joseph Ryan, PhD
C. Munro Cullum, PhD Donald H. Saklofske, PhD
Michael J. Furhman, PhD David Salmon, PhD
Jim Gold, PhD Vicki Schwean, PhD
Robert Heaton, PhD Esther Strauss, PhD
David O. Herman, PhD Michael Whetstone, PhD
Elizabeth Heron, PhD John L. Woodard, PhD

Reviewers and Consultants

Jennifer Abe-Kim, PhD Candace M. Fleming, PhD
Kenneth Adams, PhD Michael D. Franzen, PhD
Andres Barona, PhD Craig Frisby, PhD
Russell M. Bauer, PhD Lucy Frontera, PhD
Linas A. Bieliauskas, PhD Ollie Gibbs EdD
Lawrence M. Binder, PhD David Goh, PhD
Robert Bornstein, PhD Jim Gold, PhD
Jeffrey Braden, PhD Richard Gorsuch, PhD
Patricia Brazil, PhD Patti L. Harrison, PhD
Nelson Butters, PhD Robert Heaton, PhD
Meryl Butters, PhD Janet Helms, PhD
John B. Carroll, PhD David Herman, PhD
Gordon J. Chelune, PhD Elizabeth Heron, PhD
Cindy Cimino, PhD George W. Hynd EdD
Raymond M. Costello, PhD Robert Ivnik, PhD
Munro Cullum, PhD Sharon Johnson, PhD
Malcolm Cummings, PhD Edith Kaplan, PhD
Dean Dellis, PhD Alan Kaufman, PhD
Barry Dewlan, PhD Jeff King, PhD
Maureen Drews, PhD Essie Knuckle, PhD
Ruben J. Echemendia, PhD Glenn Larrabee, PhD
Eugene Emory, PhD Asenath LaRue, PhD
Ian Evans, PhD Muriel D. Lezak, PhD
 Appendix G

David W. Loring, PhD Joseph J. Ryan, PhD

Hector Machabanski, PhD Jonathon Sandoval, PhD
Charlie Magruder, PhD Jerome Sattler, PhD
David Martino, PhD Abigail Sivan, PhD
Neil Massoth, PhD John Slate, PhD
Joseph D. Matarazzo, PhD Jean Spruill, PhD
Kevin Miller, PhD Herbert H. Stenson, PhD
Charles Morton, PhD Martha Storandt, PhD
Hector E Myers, PhD Lisa Ann Suzuki, PhD
Tom Oakland, PhD Pat Tanner-Halverson, PhD
Sumie Okazaki, PhD John E Taylor, PhD
Esteban Olmeda, PhD Pat Thompson, PhD
Elijio Padilla, PhD Alexander I. Tréster, PhD
Anthony Paolo, PhD Joseph Trimble, PhD
Marcel Ponton, PhD Susana Urbina, PhD
Anthony Puente , PhD Rodney Vanderploeg, PhD
Chris Randolph, PhD Irla Lee Zimmerman, PhD
Gale H. Roid, PhD

Participating Schools and Organizations

B’Nai Tikvah, Deerfield, IL Metro Health Care Campaign, San
Cleveland Clinic Florida, Antonio, TX
Ft. Lauderdale, FL St. John’s Regional Hospital,
Cleveland Clinic Foundation, Rehabilitation Administration,
Cleveland, OH Springfield, MO
Commander's House Senior Citizen The Research General Post Fund,
Center, San Antonio, TX Pittsburgh, PA
Comprehensive Neuropsychological University of California San Diego,
Services, Albany, NY Department of Neurosciences, La
Emory University, Atlanta, GA Jolla, CA
Evanston Township High School, University of Kansas Medical Center,
Evanston, IL Kansas City, KS
Goodwill Industries of San Antonio, TX University of Saskatchewan, Saskatoon,
Jersey Shore Medical Center Behavioral SK, Canada
Health Services, Neptune, NJ University of Texas Southwestern
Long Island Jewish Medical Center, Medical Center at Dallas, Dallas, TX
Research Department, Glen Oaks, NY University of Wisconsin Madison,
Lutheran High School, San Antonio, TX Department of Educational
MacArthur High School Band, San Psychology, Madison, WI
Antonio, TX Wilford Hall Medical Center, San
Maryland Psychiatric Research Center, Antonio, TX
Baltimore, MD

399
Sic | Vive
Ww 0 xoH

ibaa athe
nev, Lived

‘ or. i lis%i

t é rr) ia |
: _ HIM) nrg
on H iA f = hatte

te } mT) F ; irrwalt
-
-

itsidoO eel
i olevsd() sing
“P sirefyltl messiat
SoS vohirontl mae Cit Vetta) olga
=Sayry Sete? i 7 AT otew hynny
us tit) ome sot
iV vs , reha “aquearrdinaé
fl peestsanaks, 14 a! i . rs ant(ibirie® abst
3 “ . aii Silo Hsia
=

‘cele Seema Ue viogdig2,yausalgin

we otileSL oeae ean bia. lane! ‘fe Li bobs i

i ol
buaitalet }
wigul vevlid sz Pa a
~—s heel! reget
: LO TEN Atygcurigr'}
7
in
heft ospubl lenaaynaitiaoanees
Pweliet
Nwo ny eye f
Metori
 REFERENCES

Ackerman, P. T., Dykman, R. A., & Peters, J. E. (1976). Hierarchical factor patterns on
the WISC as related to areas of learning deficit. Perceptual and Motor Skills, 42,
583-615.
Adams, J. H., Graham, D. I., Murray, L. S., & Scott, G. (1982). Diffuse anoxal injury due
to nonmissle head injury in humans: An analysis of 45 cases. Annals of Neurology,
12, 557-563.
Akaike, H. (1987). Factor analysis and AIC. Psychkometrika, 52(3), 317-332.
Alexander, G. E., Prohovnik, I., Stern, Y., & Mayeux, R. (1994). WAIS-R subtest profile
and cortical perfusion in Alzheimer’s disease. Brain and Cognition, 24, 24-43.
Allen, M. J., & Yen, W. M. (1979). Introduction to measurement theory. Monterey, CA:
Brooks/Cole.
Almkvist, O., Backman, L., Basun, H., & Wahlund, L. O. (1993). Patterns of neuro-
psychological performance in Alzheimer’s disease and vascular dementia. Cortex,
29, 661-673.
American Association on Mental Retardation. (1992). Mental retardation: Definitions,
classification, and systems of supports (9th ed.). Washington, DC: Author.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
Anastasi, A., & Urbina, S. (1997). Psychological testing (7th ed.). Upper Saddle River,
NJ: Prentice Hall.
Andreasen, N. C., & Carpenter, W. T., Jr. (1993). Diagnosis and classification of schizo-
phrenia. Schizophrenia Bulletin, 19(2), 199-214.
Angoff, W. H. (1984). Scales, norms, and equivalent scores. Princeton, NJ: Educational
Testing Service.
Annegers, J. E, Grabow, J. D., Kurland, L. T., & Laws, E. R. (1980). The incidence,
causes, and secular trends of head trauma in Olmstead County, Minnesota,
1935-1974. Neurology, 30, 912-919.
Arcia, E., & Gualtieri, C. T. (1994). Neurobehavioural performance of adults with
closed-head injury, adults with attention deficit, and controls. Brain Injury, 8(5),
395-404.
analysis of
Arnau, R. C., & Thompson, B. (2000). Second-order confirmatory factor
the WAIS-III. Psychological Assessment, 7(3), 237-246.
, A. R. (1991). Some cyto-
Arnold, S. E., Hyman, B. T., Van Hoesen, G. W,, & Damasio
Archives of
architectural abnormalities of the entorhinal cortex in schizophrenia.
General Psychiatry, 48, 625-632.
Moss, H., & Van Thiel, D. H.
Atria, A. M., Tarter, R. E., Kabene, M. A., Laird, S. B.,
cs. Alcoholism:
(1991). The role of cirrhosis in memory functioning of alcoholi
Clinical and Experimental Research, 15 (6), 932-937.
scatter analysis. Journal of
Atkinson, L. (1992) Mental retardation and WAIS-R
Intellectual Disability Research, 36, 443-448.
ed system and its control processes.
Atkinson, R. C., & Shiffrin, R. M. (1968). A propos
Spence (Eds.), The psycho logy of learning and motivation:
In K. W. Spence & J. T.
New York: Academic Press.
Advances in research and theory (Vol. 2, pp. 82-90).
iated Scale of Intelligence and
Axelrod, B. N. (2002). Validity of the Wechsler Abbrev Assessment, 9(1),
intell ectual functi oning.
other very short forms of estimating
1f-23.

401
References

Baddeley, A. D. (1986). Working memory. Oxford, England: Oxford University Press.

Baddeley, A. D. (1992). Is working memory working? The fifteenth Bartlett lecture. The
Quarterly Journal of Experimental Psychology, 441(1), 3-29.
Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. H. Bower (Ed.), The
psychology of learning and motivation: Advances in research and theory (Vol. 8,
pp. 47-90). San Diego, CA: Academic Press.
Bannatyne, A. (1968). Diagnosing learning disabilities and writing remedial prescrip-
tions. Journal of Learning Disabilities, 1(4), 242-249.
Bannatyne, A. (1974). Diagnosis: A note on recategorization of the WISC scaled ©
scores. Journal of Learning Disabilities, 7(5), 272-273.
Barkley, R. A., Anastopoulos, A. D., Guevremont, D. C., & Fletcher, K. E. (1991).
Adolescents with ADHD: Patterns of behavioral adjustment, academic function-
ing, and treatment utilization. Journal of the American Academy of Child and
Adolescent Psychiatry, 30(5), 752-761.
Barona, A., Reynolds, C. R., & Chastain, R. (1984). A demographically based index of
pre-morbid intelligence for the WAIS-R. Journal of Consulting and Clinical
Psychology, 52(5), 885-887.
Basso, M. R., Harrington, K., Matson, M., & Lowery, N. (2000). Sex differences on the
WMS-III: Findings concerning Verbal Paired Associates and Faces. Clinical
Neuropsychologist, 14(2), 231-235.
Baving, L., Laucht, M., & Schmidt, M. H. (1999). Atypical frontal brain activation in
ADHD: Preschool and elementary school boys and girls. Journal of the American
Academy of Child and Adolescent Psychiatry, 38(11), 1363-1371.
Bawden, H. N., Knights, R. M., & Winogron, H. W. (1985). Speeded performance
following head injury in children. Journal of Clinical and Experimental
Neuropsychology, 7(1), 39-54.
Beardsworth, E. D., & Zaidel, D. W. (1994). Memory for faces in epileptic children
before and after brain surgery. Journal of Clinical and Experimental
Neuropsychology, 16(4), 589-596.
Beck, A. T., & Steer, R. A. (1987). Beck Depression Inventory. San Antonio, TX: The
Psychological Corporation.
Bentler, P. M. (1980). Multivariate analysis with latent variables: Causal modeling.
Annual Review of Psychology, 31, 419-456.
Bentler, P. M., & Wu, E. J. C. (1993). EQS/Windows user’s guide (Version 4). Los Angeles:
BMDP Statistical Software.
Benton, A. L., Eslinger, P. J., & Damasio, A. R. (1981). Normative observations on
neuropsychological test performances in old age. Journal of Clinical Neuro-
psychology, 3(1), 33-42.
Benton, A. L., & Hamsher, K. deS. (1994). Multilingual Aphasia Examination (3rd
ed.).
Iowa City: AJA Associates.
Benton, A. L., Hamsher, K. deS., Varney, N. R., & Spreen, O. (1983). Judgment of Line
Orientation, Form H. New York: Oxford University Press.
Berardi, A., Haxby, J. V., Grady, C. L., & Rapoport, S. I. (1991). Asymmetries of
brain
glucose metabolism and memory in the healthy elderly. Developmental
Neuropsychology, 7(1), 87-97.
Berg, E. A. (1948). A simple objective technique for measuring flexibility in
thinking.
Journal of General Psychology, 39, 15-22.

402
 References

Berk, R. A. (1984). Screening and diagnosis of children with learning disabilities.

Springfield, IL: Charles C. Thomas.
Biederman, J., Faraone, S. V., Spencer, T., Wilens, T., Norman, D., Lapey, K. A., et al.
(1993). Patterns of psychiatric comorbidity, cognition, and psychosocial function-
ing in adults with attention deficit hyperactivity disorder. American Journal of
Psychiatry, 150(12), 1792-1798.
Bigler, E. D., Johnson, S. C., Anderson, C. V., Blatter, D. D., Gale, S. D., Russo, A. A., et
al. (1996). Traumatic brain injury and memory: The role of hippocampal atrophy.
Neuropsychology, 10(3), 333-342.
Blair, J. R., & Spreen, O. (1989). Predicting premorbid IQ: A revision of the National
Adult Reading Test. Clinical Neuropsychologist, 3(2), 129-136.
Blyler, C. R., Gold, J. M., Iannone, V. N., Buchanan, R. W. (2000). Short form of the
WAIS-III for use with patients with schizophrenia. Schizophrenia Research, 46,
209-215.
Bohman, M., Cloninger, R., Sigvardsson, S., & von Knorring, A. L. (1987). The genetics
of alcoholisms and related disorders. Journal of Psychiatric Research, 21(4),
447-452.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Bollen, K. A., & Long, J. S. (1993). Introduction. In K. A. Bollen & J. S. Long (Eds.),
Testing structural equation models (pp. 1-9). Newbury Park, CA: Sage.
Bondi, M. W., Monsch, A. U., Galasko, D., Butters, N., Salmon, D. P, & Delis, D. C.
(1994). Preclinical cognitive markers of dementia of the Alzheimer type.
Neuropsychology, 8(3), 374-384.
Bornstein, R. A., & Chelune, G. J. (1988). Factor structure of the Wechsler Memory
Scale—Revised. The Clinical Neuropsychologist, 2(2), 107-115.
cies in
Bornstein, R. A., Chelune, G. J., & Prifitera, A. (1989). 1Q-memory discrepan
samples. Journal of Consultin g and Clinical Psycholog y, 1(3),
normal and clinical
203-206.
stability of
Bowden, S. C., Whelan, G., Long, C. M., & Clifford, C. C. (1995). Temporal
heteroge neous sample of alcohol dependen t clients.
the WAIS-R and WMS-R in a
The Clinical Neuropsy chologis t, 9(2), 194-197.
etric profile on the
Braden, J. P. (1990). Do deaf persons have a characteristic psychom
Psychoe ducatio nal Assessme nt, 8,
Wechsler Performance scales? Journal of
518-526.
and hard-of-hearing people:
Braden, J. P. (1992). Intellectual assessment of deaf
School Psychology Review, 21(1),
A quantitative and qualitative research synthesis.
82-94. p-
difference versus regression discre
Braden, J. P, & Weiss, L. (1988). Effects of simple 2. |
School Psychology, 26, 133-14
ancy methods: An empirical study. Journal of Butters (Eds.),
ors. In L. R. Squire & N.
Brandt, J. (1992). Detecting amnesia’s impost
5). New York: Guilford Press.
Neuropsychology of memory (2nd ed., pp. 156-16
FW. (1993) . The demen tia of Huntington s disease. In R. W.
Brandt, J. A., & Bylsma,
psychology of Alzheimer's disease and
Parks, R. E Zec, & R. S. Wilson (Eds.), Neuro
Oxford University Press.
other dementias (pp. 265-282). New York:
Brandt, J., & Butters, N. (1986). The neuro psychology of Huntingtons disease. Trends
in Neuroscience, 9, 118-120.

403
References

Brennen, R. L. (1983). Elements of generalizability theory. lowa City, IA: American

College Testing Program.
Brinkman, S. D., & Braun, P. (1984). Classification of dementia patients by a WAIS
profile related to central cholinergic deficiencies. Journal of Clinical Neuro-
psychology, 6(4), 393-400.
Brody, N. (1992). Intelligence (2nd ed.). San Diego, CA: Academic Press.
Brooks, D. N. (1976). Wechsler Memory Scale performance and its relationship to
brain damage after severe closed head injury. Journal of Neurology, Neurosurgery,
and Psychiatry, 39(6), 593-601.
Brown, R. G., & Marsden, C. D. (1986). Visuospatial function in Parkinson's disease.
Brain, 109, 987-1002.
Brown, T. E. (1996). Brown Attention—Deficit Disorder Scales. San Antonio, TX: The
Psychological Corporation.
Burke, H. R. (1985). Raven’s Progressive Matrices (1938): More on norms, reliability,
and validity. Journal of Clinical Psychology, 41(2), 231-235.
Burke, H. R., & Bingham, W. C. (1969). Raven’s Progressive Matrices: More on
construct validity. Journal of Psychology, 72, 247-251.
Burton, D. B., Mittenberg, W., & Burton, C. A. (1993). Confirmatory factor analysis of
the Wechsler Memory Scale—Revised standardization sample. Archives of Clinical
Neuropsychology, 8(6), 467-475.
Butters, N. (1986). The clinical aspects of memory disorders. In T. Incagnoli, G.
Goldstein, & C. J. Golden (Eds.), Clinical application of neuropsychological test
batteries (pp. 361-382). New York: Plenum Press.
Butters, N., & Cermak, L. S. (1980). Alcoholic Korsakoff’s syndrome. New York:
Academic Press. :
Butters, N., Salmon, D. P, Cullum, C. M., Cairns, P, Tréster, A. I., Jacobs, D., et al.
(1988). Differentiation of amnesic and demented patients with the Wechsler
Memory Scale—Revised. The Clinical Neuropsychologist, 2(2), 133-148.
Butters, N., Sax, D., Montgomery, K., & Tarlow, S. (1978). Comparison of the
neuropsychological deficits associated with early and advanced Huntington's
disease. Archives of Neurology, 35(9), 585-589.
Butters, N., Wolfe, J., Martone, M., Granholm, E., & Cermak, L. S. (1985). Memory
disorders associated with Huntington's disease: Verbal recall, verbal recognition
and procedural memory. Neuropsychologia, 23(6), 729-743.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Capruso, D. X., & Levin, H. S. (1992). Cognitive impairment following closed head
injury. Neurologic Clinics, 10(4), 879-893.
Carlesimo, G. A., & Caltagirone, C. (1995). Components in the visual processing of
known and unknown faces. Journal of Clinical and Experimental Neuropsychology,
17(5), 691-705.
Carlson, R. A., Khoo, B. H., Yaure, R. G., & Schneider, W. (1990). Acquisition
of a
problem-solving skill: Levels of organization and use of working memory.
Journal
of Experimental Psychology: General, 119(2), 193-214.
Carroll, J. B. (1993). Human cognitive abilities: A survey offactor-analytic
studies. New
York: Cambridge University Press.

404
 References

Carroll, J. B. (1997). The three-stratum theory of cognitive abilities. In D. P. Flanagan,

J. L. Genshaft, & P. L. Harrison (Eds.), Contemporary intellectual assessment:
Theories, tests, and issues (pp. 122-130). New York: Guilford Press.
Castellanos, E X., Giedd, J. N., Marsh, W. L., Hamburger, S. D., Vaituzis, A. C.,
Dickstein, D. P, et al. (1996). Quantitative brain magnetic resonance imaging in
attention-deficit hyperactivity disorder. Archives of General Psychiatry, 53(7),
607-616.
Castellanos, FE X., Giedd, J. N., Eckburg, P, Marsh, W. L., Vaituzis, A. C., Kaysen, D. et
al. (1994). Quantitative morphology of the caudate nucleus in attention deficit
hyperactivity disorder. American Journal of Psychiatry, 151(12), 1791-1796.
Cattell, R. B. (1943). The measurement of adult intelligence. Psychological Bulletin,
40(3), 153-193.
Cattell, R. B. (1963). Theory of fluid and crystallized intelligence: A critical experi-
ment. Journal of Educational Psychology, 54(1), 1-22.
Cattell, R. B., & Horn, J. L. (1978). A check on the theory of fluid and crystallized
intelligence with description of new subtest designs. Journal of Educational
Measurement, 15(3), 139-164.
Chelune, G. J., & Bornstein, R. A. (1988). WMS-R patterns among patients with
unilateral brain lesions. The Clinical Neuropsychologist, 2(2), 121-132.
Chelune, G. J., Bornstein, R. A., & Prifitera, A. (1990). The Wechsler Memory Scale—
Revised: Current status and applications. In J. Rosen, P. McReynolds, & G. J.
Chelune (Eds.), Advances in psychological assessment (Vol. 7, pp. 66-95). New York:
Plenum Press.
Chelune, G. J., Naugle, R. I., Liiders, H., Sedlak, J., & Awad, I. A. (1993). Individual
change after epilepsy surgery: Practice effects and base-rate information.
Neuropsychology, 7(1), 41-52.
Cohen, J. (1952a). A factor-analytically based rationale for the Wechsler—Bellevue.
Journal of Consulting Psychology, 16, 272-277.
Cohen, J. (1952b). Factors underlying Wechsler-Bellevue performance of three
neuropsychiatric groups. Journal of Abnormal and School Psychology, 47, 359-364.
Cohen, J. (1957a). A factor-analytically based rationale for the Wechsler Adult
Intelligence Scale. Journal of Consulting Psychology, 21(6), 451-457.
Cohen, J. (1957b). The factorial structure of the WAIS between early adulthood and
old age. Journal of Consulting Psychology, 21(4), 283-290.
Cohen, M. (1997). Children’s Memory Scale. San Antonio, TX: The Psychological
Corporation.
Conrad, A. J., Abebe, T., Austin, R., Forsythe, S., & Scheibel, A. B. (1991). Hippo-
on.
campal pyramidal cell disarray in schizophrenia as a bilateral phenomen
Archives of General Psychiatry, 48, 413-417. he
verbal learning.
Cooper, E. H., & Pantle, A. J. (1967). The total-time hypothesis in
Psychological Bulletin, 68(4), 221-234. J a
E. V. (1991). Cognitive
Cooper, J. A., Sagar, H. J., Jordan, N., Harvey, N. S., & Sullivan,
ship to motor
impairment in early, untreated Parkinson's disease and its relation
disability. Brain, 114, 2095-2122.
Rosenshein, K. (1981).
Cordoni, B. K., O’Donnell, J. P, Ramaniah, N. V., Kurtz, J., &
ing disabled young adults.
Wechsler adult intelligence score patterns for learn
Journal of Learning Disabilities, 14(7), 404-407.

405
References

Corina, D. P, Richards, T. L., Serafini, S., Richards, A. L., Steury, K., Abbott, R. D., et al
(2001). MRI auditory language differences between dyslexic and able reading
children. Neuroreport, 12(6), 1195-1201.
Corsi, P. M. (1972). Human memory and the medial temporal region of the brain.
Unpublished doctoral dissertation, McGill University, Montreal, Canada.
Costa, L. D. (1975). The relation of visuospatial dysfunction to digit span performance
in patients with cerebral lesions. Cortex, 11, 31-36.
Craft, N. P, & Kronenberger, E. J. (1979). Comparability ofWISC-R and WAIS IQ
scores in educable mentally handicapped adolescents. Psychology in the Schools,
16(4), 502-504.
Crawford, J. R. (1992). Current and premorbid intelligence measures in neuropsycho-
logical assessment. In J. R. Crawford, D. M. Parker, & W. W. McKinlay (Eds.), A
handbook of neuropsychological assessment (pp. 21-49). Hillsdale, NJ: Erlbaum.
Crocker, L., & Algina, J. (1986). Introduction to classical and modern test theory. Fort
Worth, TX: Harcourt Brace Jovanovich College.
Cronbach, L. J., Gleser, G. C., Nanda, H., & Rajaratnam, N. (1972). The dependability
of behavioral measurements: Theory of generalizability for scores and profiles. New
York: Wiley.
Cronbach, L. J., Rajaratnam, N., & Gleser, G. C. (1963). Theory of generalizability: A
liberalization of reliability theory. British Journal of Statistical Psychology, 16(2),
137-163.
Cutting, J. (1978). Patterns of performance in amnesic subjects. Journal of Neurology,
Neurosurgery, and Psychiatry, 41(3), 278-282.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory
and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450-466.
Daum, I., Schugens, M. M., Spieker, S., Poser, U., Schonle, P W., & Birbaumer, N.
(1995). Memory and skill acquisition in Parkinson's disease and frontal lobe
dysfunction. Cortex, 31, 413-432.
Davila, M. D., Shear, P. K., Lane, B., Sullivan, E. V., & Pfefferbaum, A. (1994).
Mammillary body and cerebellar shrinkage in chronic alcoholics: An MRI and
neuropsychological study. Neuropsychology, 8(3), 433-444.
Davis, FE B. (1959). Interpretation of differences among averages and individual test
scores. Journal of Educational Psychology, 50(4), 162-170.
de Jong, P. F, & Das—Smaal, E. A. (1995). Attention and intelligence: The validity of the
Star Counting Test. Journal of Educational Psychology, 87(1), 80-92.
de Jonge, P, & de Jong, P. E (1996). Working memory, intelligence and reading ability
in children. Personality and Individual Differences, 21(6), 1007-1020.
Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California Verbal Learning
Test—Adult Version. San Antonio, TX: The Psychological Corporation.
Delis, D. C., Massman, P.J., Butters, N., Salmon, D. P, Cermak, L. S., & Kramer,
J. H.
(1991). Profiles of demented and amnesic patients on the California Verbal
Learning Test: Implications for the assessment of memory disorders. Psychological
Assessment: A Journal of Consulting and Clinical Psychology, 3(1), 19-26.
Denckla, M. B. (1993). The child with developmental disabilities grown up:
Adult
residua of childhood disorders. Behavioral Neurology, 11(1), 105-125.
Denckla, M. B. (1996). A theory and model of executive function. In G.
R. Lyon &N. A.
Krasnegor (Eds.), Attention, memory, and executive function (pp. 263-278).
Baltimore: Paul. H. Brookes.

406
 References

Desai, M. M. (1955). The relationship of the Wechsler—Bellevue Verbal scale and the
Progressive Matrices Test. Journal of Consulting Psychology, 19(1), 60.
Developmentally Disabled Assistance and Bill of Rights Act of 1975, 42 U.S.C. §6001.
Dichter, M. A. (1994). The epilepsies and convulsive disorders. In K. J. Isselbacher, E.
Braunwald, J. D. Wilson, J. B. Martin, A. S. Fauci, & D. L. Kasper (Eds.), Harrison's
principles of internal medicine (13th ed., Vol. 2, pp. 2223-2233). New York:
McGraw-Hill.
Dickerson, EF B., Ringel, N. B., & Boronow, J. J. (1991). Neuropsychological deficits in
chronic schizophrenics: Relationship with symptoms and behavior. Journal of
Nervous and Mental Disease, 179(12), 744-749.
Dikmen, S. S., Machamer, J. E., Winn, H. R., & Temkin, N. R. (1995). Neuropsycho-
logical outcome at 1-year post head injury. Newropsychology, 9(1), 80-90.
Dikmen, S., Reitan, R. M., & Temkin, N. R. (1983). Neuropsychological recovery in
head injury. Archives of Neurology, 40, 333-338.
Donders, J., Tulsky, D. S., & Zhu, J. (2001). Criterion validity of new WAIS-III subtest
scores after traumatic brain injury. Journal of the International Neuropsychological
Society, 7(7), 892-898.
Doss, R. C., Chelune, G. J., & Naugle, R. I. (2000). Comparability of the expanded
WMS-III standardization protocol to the published WMS-III among right and left
temporal lobectomy patients. Clinical Neuropsychologist, 14(4), 468-473.
Du, A. T., Schuuf, N., Amend, D., Laasko, M. P, Hsu, Y. Y., Jagust, W. J., et al. (2001).
Magnetic resonance imaging of the entorhinal cortex and hippocampus in mild
cognitive impairment and Alzheimer's disease. Journal of Neurology, Neurosurgery
and Psychiatry, 71(4), 431-432.
Dudek, E J. (1979). The continuing misinterpretation of the standard error of
measurement. Psychological Bulletin, 86(2), 335-337.
Eckardt, M. J., Rohrbaugh, J. W., Stapleton, J. M., Davis, E. Z., Martin, P. R., &
Weingartner, H. J. (1996). Attention-related brain potential and cognition in
alcoholism-associated organic brain disorders. Biological Psychiatry, 39, 143-146.
Education for All Handicapped Children Act of 1975, 20 U.S.C. $1400 et seq.
Ellenberg, L., Rosenbaum, G., Goldman, M. S., & Whitman, R. D. (1980). Recover-
ability of psychological functioning following alcohol abuse: Lateralization effects.
Journal of Consulting and Clinical Psychology 48(4), 503-510.
Ellis, R. J., & Oscar-Berman, M. (1989). Alcoholism, aging, and functional cerebral
asymmetries. Psychological Bulletin, 106(1), 128-147.
Scale—Revised
Elwood, R. W. (1991). Factor structure of the Wechsler Memory
(WMS-R) in a clinical sample: A methodological reappraisal. The Clinical
Neuropsychologist, 5(4), 329-337.
& Tulsky, D. S. (2000).
Fisher, D. C., Ledbetter, M. E, Cohen, N. J., Marmor, D.,
patients. Applied
WAIS-III and WMS-III profiles of mildly to severely brain-injured
Neuropsychology, 7(3), 126-132.
to detect and charac-
Fisher, J. S. (1988). Using the Wechsler Memory Scale—Revised
Neuropsychologist, 2(2),
terize memory deficits in multiple sclerosis. The Clinical
149-172.

407
References

Flanagan, D. P, & McGrew, K. S. (1997). A cross-battery approach to assessing and

interpreting cognitive abilities: Narrowing the gap between practice and cognitive
science. In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contemporary
intellectual assessment: Theories, tests, and issues (pp. 314-325). New York:
Guilford Press.
Flynn, J. R. (1984). The mean IQ of Americans: Massive gains 1932 to 1978.
Psychological Bulletin, 95(1), 29-51.
Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure.
Psychological Bulletin, 101(2), 171-191.
Folstein, M. E, Folstein, S. E., & McHugh, P. R. (1975). “Mini-Mental State”: A practical
method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12, 189-198.
Frank, Y., & Pavlakis, S. G. (2001). Brain imaging in neurobehavioral disorders.
Pediatric Neurology, 25(4), 278-287.
Franzen, M. D. (1989). Reliability and validity in neuropsychological assessment. New
York: Plenum Press.
Fry, A. FE, & Hale, S. (1996). Processing speed, working memory, and fluid intelligence:
Evidence for a developmental cascade. Psychological Science, 7(4), 237-241.
Fuld, P. A., Katzman, R., Davies, P, & Terry, R. D. (1982). Intrusions as a sign of
Alzheimer’s dementia: Chemical and pathological verification. Annals of
Neurology, 11(2), 155-159.
Gathercole, S. E. (1994). Neuropsychology and working memory: A review.
Neuropsychology, 8(4), 494-505.
Georgiewa, P,, Rzanny, R., Gaser, C., Gerhard, U. J., Vieweg, U., Freesmeyer, D., et al.
(2002). Phonological processing in dyslexic children: A study combining functional
imaging and event related potentials. Neuroscience Letters, 18(1), 5-8.
Gfeller, J. D., & Rankin, E. J. (1991). The WAIS-R profile as a cognitive marker of
Alzheimer’s disease: A misguided venture? Journal of Clinical and Experimental
Neuropsychology, 13(4), 629-636.
Giedd, J. N., Blumenthal, J., Molloy, E., & Castellanos, E X. (2001). Brain imaging of
attention deficit/hyperactivity disorder. Annals of the New York Academy Science,
931, 33-49.
Glenn, S. W., & Parsons, O. A. (1990). The role of time in neuropsychological perfor-
mance: Investigation and application in an alcoholic population. The Clinical
Psychologist, 4(4), 344-354.
Glenn, S. W., Parsons, O. A., Sinha, R., & Stevens, L. (1988). The effects of repeated
withdrawals from alcohol on the memory of male and female alcoholics. Alcohol
and Alcoholism, 23(5), 337-342.
Glutting, J. J., McDermott, P. A., & Stanley, J. C. (1987). Resolving differences among
methods of establishing confidence limits for test scores. Educational and
Psychological Measurement, 47(3), 607-614.
Gold, J. M., Carpenter, C., Randolph, C., Goldberg, T. E., & Weinberger, D. R. (1997).
Auditory working memory and Wisconsin Card Sorting Test performance in
schizophrenia. Archives of General Psychiatry, 54, 159-165.
Gold, J. M., Hermann, B. P,, Randolph, C., Wyler, A. R., Goldberg, J. E, & Weinberger,
D. R. (1994). Schizophrenia and temporal lobe epilepsy: A neuropsychological
analysis. Archives of General Psychiatry, 54, 265-272.

408
 References

Goldberg, T. E., Weinberger, D. R., Berman, K. E, Pliskin, N. H., & Podd,

M. H. (1987).
Further evidence for dementia of the prefrontal type in schizophrenia? A con-
trolled study of teaching the Wisconsin Card Sorting Test. Archives of General
Psychiatry, 44, 1008-1014.
Goldman, R. S., Axelrod, B. N., Giordani, B. J., Foster, N., & Berent, S. (1992).
Longitudinal sensitivity of the Fuld Cholinergic Profile to Alzheimer’s disease.
Journal of Clinical and Experimental Neuropsychology, 14(4), 566-574.
Goldstein, E C., & Levin, H. S. (1990). Epidemiology of traumatic brain injury:
Incidence, clinical characteristics, and risk factors. In E. D. Bigler (Ed.), Traumatic
brain injury: Mechanisms of damage, assessment, intervention, and outcome (pp.
51-67). Austin, TX: Pro-Ed.
Goldstein, F C., Levin, H. S., Presley, R. M., Searcy, J., Colohan, A. R. T., Eisenberg,
H. M., et al. (1994). Neurobehavioral consequences of closed head injury in older
adults. Journal of Neurology, Neurosurgery, and Psychiatry, 57(8), 961-966.
Goldstein, E C., McKendall, R. R., & Haut, M. W. (1992). Gist recall in multiple
sclerosis. Archives of Neurology, 49(10), 1060-1064.
Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Erlbaum.
Gorsuch, R. L. (1996). Number of exploratory factors: A clarification of the eigen-
values > | criterion. San Diego, CA: Fuller Theological Seminary, Graduate School
of Psychology.
Grant, D. A., & Berg, E. A. (1948). A behavioral analysis of the degree of reinforcement
and ease of shifting to new responses in a Weigl-type card-sorting problem.
Journal of Experimental Psychology, 38, 404-411.
Grigsby, J., Ayarbe, S. D., Kravcisin, N., & Busenbark, D. (1994). Working memory
impairment among persons with chronic progressive multiple sclerosis. Journal of
Neurology, 241, 125-131.
Grober, E., & Sliwinski, M. (1991). Development and validation of a model estimating
premorbid verbal intelligence in the elderly. Journal of Clinical and Experimental
Neuropsychology, 13(6), 933-949.
Guilford, J. P. (1954). Psychometric methods (2nd ed.). New York: McGraw-Hill.
Guilford, J. P, & Fruchter, B. (1978). Fundamental statistics in psychology and educa-
tion (6th ed.). New York: McGraw-Hill.
Gur, R. C., Jaggi, J. L., Ragland, J. D., Resnick, S. M., Shtasel, D., Muenz, L., et al.
(1993). Effects of memory processing on regional brain activation: Cerebral blood
flow in normal subjects. International Journal of Neuroscience, 72, 31-44.
Gur, R. E., Mozley, P. D., Resnick, S. M., Shtasel, D., Kohn, M., Zimmerman, Re et all
(1991). Magnetic resonance imaging in schizophrenia: I. Volumetric analysis of
brain and cerebrospinal fluid. Archives of General Psychiatry, 48, 407-412. ei
Gustafsson, J. E. (1984). A unifying model for the structure of intellectual abilities.
Intelligence, 8, 179-203. Kasi
Hall, J. C. (1957). Correlation of a modified form of Raven's Progressive Matrices
(1938) with the Wechsler Adult Intelligence Scale. Journal of Consulting
Psychology, 21(1), 23-26.
Hambleton, R. K. (1993). Principles and selected applications of item response
pp. 147-200).
theory. In R. L. Linn (Ed.), Educational measurement (3rd ed.,
Phoenix, AZ: Oryx Press.

409
References

Hambleton, R. K. (1994). Guidelines for adapting educational and psychological tests:

A progress report. European Journal of Psychological Assessment, 10(3), 229-244.
Harrison, P. L. (1990). Mental retardation, adaptive behavior assessment, and gifted-
ness. In A. S. Kaufman (Ed.), Assessing adolescent and adult intelligence (pp.
533-585). Boston: Allyn & Bacon.
Hart, R. P, Kwentus, J. A., Wade, J. B., & Hamer, R. M. (1987). Digit symbol perfor-
mance in mild dementia and depression. Journal of Consulting and Clinical
Psychology, 55(2), 236-238.
Haslam, C., Batchelor, J., Fearnside, M. R., Haslam, S. A., Hawkins, S., & Kenway, E.
(1994). Post-coma disturbance and post-traumatic amnesia as nonlinear predic-
tors of cognitive outcome following severe closed head injury: Findings from the
Westmead Head Injury Project. Brain Injury, 8(6), 519-528.
Hauser, S. L. (1994). Multiple sclerosis and other demyelinating diseases. In K. J.
Isselbacher, E. Braunwald, J. D. Wilson, J. B. Martin, A. S. Fauci, & D. L. Kasper
(Eds.), Harrison's principles of internal medicine (13th ed., Vol. 2, pp. 2287-2295).
New York: McGraw-Hill.
Haut, M. W,, Weber, A. M., Demarest, D., Keefover, R. W., & Rankin, E. D. (1996).
Controlling for constructional dysfunction with the visual reproduction subtest of
the Wechsler Memory Scale—Revised in Alzheimer’s disease. The Clinical
Neuropsychologist, 10(3), 309-312.
Haut, M. W., Weber, A. M., Wilhelm, K. L., Keefover, R. W., & Rankin, E. D. (1994). The
visual reproduction subtest as a measure of visual perceptual/constructional
functioning in dementia of the Alzheimer’s type. The Clinical Neuropsychologist,
8(2), 187-192.
Hawkins, K. A. (1999). Memory deficits in patients with schizophrenia: Preliminary
data from the Wechsler Memory Scale-Third Edition support earlier findings.
Journal of Psychiatry and Neuroscience, 24(4), 341-347.
Hawkins, K. A., & Tulsky, D. A. (2001). The influence of IQ stratification on
WAIS-III/WMS-III FSIQ—General Memory Index discrepancy base-rates in the
standardization sample. Journal of the International Neuropsychological Society,
7(7), 875-880.
Heaton, R. K., Baade, L. E., & Johnson, K. L. (1978). Neuropsychological test results
associated with psychiatric disorders in adults. Psychological Bulletin, 85(1),
141-162.
Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., & Curtiss, G. (1993). Wisconsin
Card Sorting Test manual—Revised and expanded. Odessa, FL: Psychological
Assessment Resources.
Heaton, R. K., & Crowley, T. J. (1981). Effects of psychiatric disorders and their
somatic treatments on neuropsychological test results. In S. B. Filskov & T. J. Boll
(Eds.), Handbook of clinical neuropsychology (pp. 481-525). New York: Wiley.
Heaton, R. K., Nelson, L. M., Thompson, D. S., Burks, J. S., & Franklin, G. M. (1985).
Neuropsychological findings in relapsing-remitting and chronic-progressive
multiple sclerosis. Journal of Consulting and Clinical Psychology, 53(1), 103-110.
Heaton, R., Paulsen, J. S., McAdams, L. A., Kuck, J., Zisook, S., Braff, D., et al. (1994).
Neuropsychological deficits in schizophrenics: Relationship to age, chronicity, and
dementia. Archives of General Psychiatry, 51, 469-476.

410
 References

Heilbronner, R. L. (1992). The search for a “pure” visual memory test: Pursuit of per-
fection? Clinical Neuropsychologist, 6(1), 105-112.
Hoehn, M. M., & Yahr, M. D. (1967). Parkinsonism: Onset, progression, and mortality.
Neurology 17(5), 427-442.
Hoff, A. L., Riordan, H., O’Donnell, D. W,, Morris, L., & DeLisi, L. E. (1992).
Neuropsychological functioning of first-episode schizophreniform patients.
American Journal of Psychiatry, 149(7), 898-903.
Holdnack, J. A., Ledbetter, M. FE, & Cohen, M. (1996, August). Performance of children
with ADD on the Children’s Memory Scale. Presented at the 75th Annual American
Psychological Association Convention, Toronto, Ontario.
Holdnack, J. A., Moberg, P. J., Arnold, S. E., Gur, R. C., & Gur, R. E. (1995). Speed of
processing and verbal learning deficits in adults diagnosed with attention deficit
disorder. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 8(4),
282-292.
Holland, P. W., & Thayer, D. T. (1988). Differential item performance and the Mantel-
Haenszel procedure. In H. Wainer & H. I. Braun (Eds.), Test validity (pp. 129-145).
Hillsdale, NJ: Erlbaum.
Iverson, G. L. (2001). Interpreting change on the WAIS-III/WMS-III in clinical
samples. Archives of Clinical Neuropsychology, 16, 183-191.
Ivnik, R. J., Malec, J. E, Smith, G. E., Tangalos, E. G., Petersen, R. C., Kokmen, E., et al.
(1992). Mayo’s older Americans normative studies: WAIS-R norms for ages 56-97.
The Clinical Neuropsychologist, 6(Suppl.), 1-30.
Jacobs, D. M., Sano, M., Dooneief, G., Marder, K., Bell, K. L., & Stern,Y. (1995).
Neuropsychological detection and characterization of preclinical Alzheimer's
disease. Neurology, 45, 957-962.
Jacobs, D., Tréster, A. I., Butters, N., Salmon, D. P, & Cermak, L. S. (1990). Intrusion
errors on the visual reproduction test of the Wechsler Memory Scale and the
Wechsler Memory Scale—Revised: An analysis of demented and amnesic patients.
The Clinical Neuropsychologist, 4(2), 177-191.
Jennett, B., & Bond, M. (1975). Assessment of outcome after severe brain damage.
Lancet, 1, 480-484.
Jones, B., & Parsons, O. A. (1971). Impaired abstracting ability in chronic alcoholics.
Archives of General Psychiatry, 24, 71-75.
Jéreskog, K. G. (1993). Testing structural equation models. In K. A. Bollen & J. S. Long
(Eds.), Testing structural equation models (pp. 294-316). Newbury Park, CA: Sage.
Jéreskog, K. G., & S6rbom, D. (1993). LISREL 8: User's reference guide. Chicago:
Scientific Software International.
cogni-
Jurden, E H. (1995). Individual differences in working memory and complex
tion. Journal of Educational Psychology, 87, 93-102.
In T. J. Boll
Kaplan, E. (1988). A process approach to neuropsychological assessment.
(Eds.), Clinical neuropsy chology and brain function: Research, mea-
& B. K. Bryant
DC: American Psycholo gical
surement, and practice (pp. 129-167). Washington,
Association.
WAIS-R as a Neuropsychological
Kaplan, E., Fein, D., Morris, R., & Delis, D. C. (1991).
Instrument. San Antonio, TX: The Psychological Corporation.
Boston Naming Test (2nd ed.).
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). The
Philadelphia: Lea & Febiger.

411
References

Kataria, S., Hall, C. W., Wong, M. M., & Keys, G. FE (1992). Learning styles of LD and
NLD ADHD children. Journal of Clinical Psychology, 48(3), 371-378.
Katz, L., Goldstein, G., Rudisin, S., & Bailey, D. (1993). A neuropsychological approach
to the Bannatyne recategorization of the Wechsler intelligence scales in adults
with learning disabilities. Journal of Learning Disabilities, 26(1), 65-72.
Kaufman, A. S. (1975). Factor analysis of WISC-R at 11 age levels between 65 and 163
years. Journal of Consulting and Clinical Psychology, 43(2), 135-147.
Kaufman, A. S. (1979). Intelligent testing with the WISC-R. New York: Wiley.
Kaufman, A. S. (1990). Assessing adolescent and adult intelligence. Boston: Allyn &
Bacon. :
Kaufman, A. S. (1994). Intelligent testing with the WISC-III. New York: Wiley.
Kaufman, A. S., Reynolds, C. R., & McLean, J. E. (1989). Age and WAIS-R intelligence
in a national sample of adults in the 20- to 74-year age range: A cross-sectional
analysis with educational level controlled. Intelligence, 13(3), 235-253.
Kaufman, A. S., & Van Hagen, J. (1977). Investigation of the WISC-R for use with
retarded children: Correlation with the 1972 Stanford-Binet and comparison of
WISC and WISC-R profiles. Psychology in the Schools, 14(1), 10-14.
Kender, J. B, Greenwood, S., & Conard, E. (1985). WAIS-R performance patterns of
565 incarcerated adults characterized as underachieving readers and adequate
readers. Journal of Learning Disabilities, 18(7), 379-383.
Kieburtz, K., Penney, J. B., Como, P,, Ranch, N., & Shoulson, I. (1996). Unified
Huntingont’s Disease Rating Scale: Reliability and consistency. Movement
Disorders, 11(2), 136-142.
Killgore, W. D. S., & DellaPietra, L. (2000a). Item response biases on the Logical
Memory Delayed Recognition subtest of the Wechsler Memory Scale-Ill.
Psychological Reports, 86(3, Pt.1), 851-857.
Killgore, W. D. S., & DellaPietra, L. (2000b). Using the WMS-III to detect malingering:
Empirical validation of the rarely missed index (RMI). Journal of Clinical and
Experimental Neuroopsychology, 22(6), 761-771.
Kopelman, M. D. (1994). Working memory in the amnesic syndrome and degenera-
tive dementia. Special section: Working memory. Neuropsychology, 8(4), 555-562.
Kostrubala, C., & Braden, J. P. (1997, April). Testing with the WAIS-III in individuals
who are deaf. Paper presented at the annual convention of the National
Association of School Psychologists, Orange County, CA
Kraus, J. E, Black, M. A., Hessol, N., Ley, PR, Rokaw, W,, Sullivan, C., et al. (1984). The
incidence of acute brain injury and serious impairment in a defined population.
American Journal of Epidemiology, 119(2), 186-201.
Krupp, L. B., Sliwinski, M., Masur, D. M., Friedberg, E, & Coyle, P. K. (1994). Cognitive
functioning and depression in patients with chronic fatigue syndrome and multi-
ple sclerosis. Archives of Neurology, 51(7), 705-710.
Kurtzke, J. F (1951). A new scale for evaluating disability in multiple sclerosis. Journal
of the American Academy of Neurology, 5, 580-583.
Kurtzke, J. F (1983). Rating neurologic impairment in multiple sclerosis: An expanded
disability status scale (EDSS). Neurology, 33, 1444-1452.
Kyllonen, P. C. (1987). Theory-based cognitive assessment. In J. Zeidner (Ed.), Human
productivity enhancement: Organizations, personnel, and decision making. (Vol. 2,
pp. 338-381). New York: Praeger.

412
 References

Kyllonen, P. C., & Christal, R. E. (1987). Cognitive modeling of learning abilities: A

status report of LAMP. In R. Dillon & J. W. Pelligrino (Eds.), Testing: Theoretical
and applied issues. New York: Freeman.
Kyllonen, P. C., & Christal, R. E. (1990). Reasoning ability is (little more than) working-
memory capacity?! Intelligence, 14, 389-433.
Kyllonen, P. C., & Stephens, D. L. (1990). Cognitive abilities as determinants of success
in acquiring logic skill. Learning and Individual Differences, 2(2), 129-160.
Lafayette Instrument Company. (1989). Grooved Pegboard. Lafayette, IN: Author.
Leckliter, I. N., Matarazzo, J. D., & Silverstein, A. B. (1986). A literature review of factor
analytic studies of the WAIS-R. Journal of Clinical Psychology, 42, 332-342,
Ledbetter, M. E, & Prifitera, A. (1993, November). Education-adjusted percent
retention scores based on the Wechsler Memory Scale—Revised standardization
sample. Paper presented at the 13th annual meeting of the National Academy of
Neuropsychology, Phoenix, AZ.
Levin, H. S., Goldstein, F C., High, W. M., Jr., & Eisenberg. H. M. (1988).
Disproportionately severe memory deficit in relation to normal intellectual func-
tioning after closed head injury. Journal of Neurology, Neurosurgery, and Psychiatry
51(10), 1294-1301.
Levine, B., & Iscoe, I. A. (1954). A comparison of Raven's Progressive Matrices (1938)
with a short form of the Wechsler—Bellevue. Journal of Consulting Psychology,
18) 10.
Levine, E. S. (1974). Psychological tests and practices with the deaf: A survey of the
state of the art. Volta Review, 76, 298-319.
Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York: Oxford
University Press.
Lineweaver, T. T., & Chelune, C. J. (2000, November). Use of the WAIS-III and WMS-III
in the context of serial assessment and meaningful change. Paper presented at the
meeting of the National Academy of Neuropsychology, Orlando, PL.
Litvan, I., Grafman, J., Vendrell, P, & Martinez, J. M. (1988). Slowed information
processing in multiple sclerosis. Archives of Neurology, 45, 281-285.
Loge, D. V., Staton, R. D., & Beatty, W. W. (1990). Performance of children with ADHD
on tests sensitive to frontal lobe dysfunction. Journal of the American Academy of
Child and Adolescent Psychiatry, 29(4), 540-545.
Logie, R. H. (1995). Viswo-spatial working memory. Hove, East Sussex, UK: Erlbaum.
Logie, R. H. (1996). The seven ages of working memory. In J. T. E. Richardson, R. W.
Engle, L. Hasher, R. H. Logie, E. R. Stoltzfus, & R. T. Zacks (Eds.), Working memory
and human cognition (pp. 31-65). New York: Oxford University Press.
Loring, D. W. (1989). The Wechsler Memory Scale—Revised, or the Wechsler Memory
Scale—Revisited? The Clinical Neuropsychologist, 3(1), 59-69.
of
Macartney-Filgate, M. S., & Vriezen, E. R. (1988). Intercorrelation of clinical tests
verbal memory. Archives of Clinical Neuropsychology, 3, 121-126.
Magnusson, D. (1967). Test theory. Reading, MA: Addison-Wesley.
Mahurin, R. K, Feher, E. PB, Nance, M. L., Levy, J. K., & Pirozzolo, EF J. (1993). Cognition
Wilson
in Parkinson's disease and related disorders. In R. W. Parks, R. FE Zec, & R. S.
(Eds.), Neuropsychology of Alzheimer’s disease and other dementias (pp. 308-349).
New York: Oxford University Press.

413
References

Maller, S. J., & Braden, J. P. (1992). The construct and criterion—related validity of the
WISC-III with deaf adolescents. Journal of Psychoeducational Assessment
[WISC-III Monograph], 105-113.
Malloy, P, Noel, N., Rogers, S., Longabaugh, R., & Beattie, M. (1989). Risk factors for
neuropsychological impairment in alcoholics: Antisocial personality, age, years of
drinking, and gender. Journal of Studies on Alcohol, 50(5), 422-426.
Malloy, P. E, & Richardson, E. D. (1994). Assessment of frontal lobe functions. Special
issue: The frontal lobes and neuropsychiatric illness. The Journal of Neuro-
psychiatry and Clinical Neurosciences, 6(4), 399-410.
Marsh, H. W,, Balla, J. R., & McDonald, R. P. (1988). Goodness-of-fit indexes in confir-
matory factor analysis: The effect of sample size. Psychological Bulletin, 103(3),
391-410.
Martin, J. B., & Gusella, J. E (1986). Huntington’s disease: Pathogenesis and manage-
ment. New England Journal of Medicine, 315(20), 1267-1276.
Massman, P. J., Delis, D. C., Butters, N., Dupont, R. M., & Gillin, J. C. (1992). The
subcortical dysfunction hypothesis of memory deficits in depression:
Neuropsychological validation in a subgroup of patients. Journal of Clinical and
Experimental Neuropsychology, 14(5), 687-706.
Massman, P. J., Delis, D. C., Butters, N., Levin, B. E., & Salmon, D. P. (1990). Are all
subcortical dementias alike? Verbal learning and memory in Parkinson’s and
Huntington’s disease patients. Journal of Clinical and Experimental
Neuropsychology, 12(5), 729-744.
Masur, D. M., Sliwinski, M., Lipton, R. B., Blau, A. D., & Crystal, H. A. (1994).
Neuropsychological prediction of dementia and the absence of dementia in
healthy elderly persons. Neurology, 44, 1427-1432.
Matarazzo, J. D. (1972). Wechsler’s measurement and appraisal of adult intelligence
(5th ed.). Baltimore: Williams & Wilkins. ;
Matarazzo, J. D. (1990). Psychological assessment versus psychological testing:
Validation from Binet to the school, clinic, and courtroom. American Psychologist,
45(9), 999-1017.
Matarazzo, J. D., Daniel, M. H., Prifitera, A., & Herman, D. O. (1988). Inter-subtest
scatter in the WAIS-R standardization sample. Journal of Clinical Psychology, 44,
940-950.
Matarazzo, J. D., & Herman, D. O. (1985). Clinical uses of the WAIS-R: Base rates of
differences between VIQ and PIQ in the WAIS-R standardization sample. In B. B.
Wolman (Ed.), Handbook of intelligence: Theories, measurements, and applications
(pp. 899-932). New York: Wiley.
Mataro, M., Garcia-Sanchez, C., Junque, C., Estevez-Gonzalez, A., & Pujol, J. (1997).
Magnetic resonance imaging measurement of the caudate nucleus in adolescents
with attention-deficit hyperactivity disorder and its relationship with neuropsy-
chological and behavioral measures. Archives of Neurology, 54(8), 963-968.
Mattis, S. (1988). Dementia Rating Scale. Odessa, FL: Psychological Assessment
Resources.
Mattson, A. J., & Levin, H. S. (1990). Frontal lobe dysfunction following closed head
injury: A review of the literature. Journal of Nervous and Mental Disease, 178(5),
282-291.

414
 References

Maurelli, M., Marchioni, E., Cerratano, R., Bosone, D., Bergmaschi, R., Citterio, A., et
al. (1992). Neuropsychological assessment in MS: Clinical, neurophysiological and
neuroradiological relationships. Acta Neurologica Scandinavica 86(2), 124-128.
McCurry, S. M., Fitz, A. G., & Teri, L. (1994). Comparison of age-extended norms for
the Wechsler Adult Intelligence Scale—Revised in patients with Alzheimer’s dis-
ease. Psychological Assessment, 6(3), 231-235.
McIntosh, G. C. (1992). Neurological conceptualization of epilepsy. In T. L. Bennett
(Ed.), The neuropsychology of epilepsy: Critical issues in neuropsychology (pp.
17-37). New York: Plenum Press.
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., & Stadlan, E. M.
(1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA
work group under the auspices of Department of Health and Human Services Task
Force on Alzheimer’s Disease. Neurology, 34, 939-944.
McLaurin, W. A., & Farrar, W. E. (1973). Validities of the Progressive Matrices Tests
against IQ and grade point average. Psychological Reports, 32, 803-806.
McMillan, T. M., Powell, G. E., Janota, I., & Polkey, C. E. (1987). Relationships between
neuropathology and cognitive functioning in temporal lobectomy patients.
Journal of Neurology, Neurosurgery, and Psychiatry, 50(2), 167-176.
McQuaid, M. E, & Alovisetti, M. (1981). School psychological services for
hearing-impaired children in the New York and New England area. American
Annals of the Deaf, 126, 37-42.
Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on
our capacity for processing information. Psychological Review, 63, 81-97.
Millis, S. R., Malina, A. C., Bowers, D. A., & Ricker, J. H. (1999). Confirmatory factor
analysis of the Wechsler Memory Scale-III. Journal of Clinical and Experimental
Neuropsychology, 21(1), 87-93.
Milner, B. (1968). Visual recognition and recall after right temporal-lobe excision in
man. Neuropsychologia, 6, 191-209.
Milner, B. (1975). Psychological aspects of focal epilepsy and its neurosurgical man-
agement. Advances in Neurology, 8, 299-321.
Morris, J. C., Heyman, A., Mohs, R. C., Hughes, J. P., van Belle, G., Fillenbaum, G., et
al. (1989). The consortium to establish a registry for Alzheimer’s disease (CERAD):
Part I. Clinical and neuropsychological assessment of Alzheimer’s disease.
Neurology, 39(9), 1159-1165.
Morris, J. C., Storandt, M., Miller, J. B, McKeel, D. W., Price, J. L., Rubin, E. H., et al.
(2001). Mild cognitive impairment represents early-stage Alzheimer disease.
Archives of Neurology, 58(10), 1705-1706.
of the Rey
Mungas, D. (1983). Differential clinical sensitivity of specific parameters
Learning Test. Journal of Consulti ng and Clinical Psycholog y,
Auditory—Verbal
-51(6), 848-855.
and mentally
Naglieri, J. A. (1980). WISC-R subtest patterns for learning disabled
retarded children. Perceptual and Motor Skills, 51, 605-606.
I. A. (1993). Detection of
Naugle, R. I, Chelune, G. J., Cheek, R., Liiders, H., & Awad,
my using the
changes in material-specific memory following temporal lobecto
Clinical Neurops ycholog y, 8(5),
Wechsler Memory Scale—Revised. Archives of
381-395.

415
 References

Naugle, R. I., Chelune, G. J., Schuster, J., Liiders, H. O., & Comair, Y. (1994).
Recognition memory for words and faces before and after temporal lobectomy.
Assessment, 1(4), 373-381.
Neisser, U., Boodoo, G., Bouchard, T. J., Jr., Boykin, A. W,, Brody, N., Ceci, S. J., et al.
(1996). Intelligence: Knowns and unknowns. American Psychologist, 51(2), 77-101.
Nelson, H. E. (1982). National Adult Reading Test (NART): Test manual. Windsor:
NFER-Nelson.
Nettelbeck, T., & Rabbitt, P M. A. (1992). Aging,.cognitive performance, and mental
speed. Intelligence, 16(2), 189-205.
Newcombe, E, de Haan, E. H. E, Ross, J., & Young, A. W. (1989). Face processing, later-
ality and contrast sensitivity. Newropsychologia, 27(4), 523-538.
Newell, A. (1973). Productions systems: Models of control structures. In W. G. Chase
(Ed.), Visual information processing (pp. 463-526). New York: Academic Press.
Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ:
Prentice-Hall.
Nicolson, R. I., Fawcett, A. J., & Dean, P. (2001). Developmental dyslexia: The cerebel-
lar deficit hypothesis. Trends in Neurosciences, 24(9), 508-511.
Nunnally, J. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.
O’Mahony, J. E, & Doherty, B. (1993). Patterns of intellectual performance among
recently abstinent alcohol abusers on WAIS-R and WMS-R subtests. Archives of
Clinical Neuropsychology, 8, 373-380.
O’Mahony, J. E, & Doherty, B. (1996). Intellectual impairment among recently absti-
nent alcohol abusers. British Journal of Clinical Psychology, 35, 77-83.
Ormrod, J. E., & Lewis, M. A. (1985). Comparison of memory skills in learning dis-
abled, low-reading, and nondisabled adolescents. Perceptual and Motor Skills,
61(1), 191-195.
Orsini, A., Fragassi, N. A., Chiacchio, L., Falanga, A. M., Cocchiaro, C., & Grossi, D.
(1987). Verbal and spatial memory span in patients with extrapyramidal diseases.
Perceptual and Motor Skills, 65, 555-558.
Oscar-Berman, M., Clancy, J. P., & Weber, D. A. (1993). Discrepancies between IQ and
memory scores in alcoholism and aging. The Clinical Neuropsychologist, 7(3),
281-296.
Palmer, K., Wang, H. X., Backman, L., Winblad, B., & Fratiglioni, L. (2002). Differential
evolution of cognitive impairment in nondemented older persons: Results from
the Kungsholmen project. American Journal of Psychiatry, 159(3), 436-442.
Parsons, O. A., & Farr, S. P. (1981). The neuropsychology of alcohol and drug use. In
S. B. Filskov & T. J. Boll (Eds.), Handbook of clinical neuropsychology (pp. 320-365).
New York: Wiley.
Parsons, O. A., & Prigatano, G. P. (1978). Methodological considerations in clinical
neuropsychological research. Journal of Consulting and Clinical Psychology, 46(4),
608-619.
Paulsen, J. S., Heaton, R. K., Sadek, J. R., Perry, W., Delis, D. C., Braff, D., et al. (1995).
The nature of learning and memory impairments in schizophrenia. Journal of the
International Neuropsychological Society, 1(1), 88-99.
Payne, R. W., & Jones, H. G. (1957). Statistics for the investigation of individual cases.
Journal of Clinical Psychology, 13, 115-121.

416
 References

Pennington, B. F, Bennetto, L., McAleer, O., & Roberts, R. J. (1996). Executive func-
tions and working memory. In G. R. Lyon & N. A. Krasnegor (Eds.), Attention,
memory, and executive function (pp. 327-348). Baltimore: Paul H. Brookes.
Pennington, B. P., & Ozonoff, S. (1996). Executive functions and developmental
psychopathology. Journal of Child Psychology and Psychiatry, 37(1), 51-87.
Peterson, R. C., Doody, R., Kurz, A., Mohs, R. C., Morris, J. C., Rabins, P. V, et al.
(2001). Current concepts in mild cognitive impairment. Archives of Neurology,
58(12), 1985-1992.
Pickles, D. G. (1966). The Wechsler Performance Scale and its relationship to speech
and educational response in deaf slow-learning children. Teacher of the Deaf, 64,
382-393.
Pilgrim, B. M., Meyers, J. E., Bayless, J., & Whetstone, M. M. (1999). Validity of the
Ward seven-subtest WAIS-III short form in a neuropsychological population.
Applied Neuropsychology, 6(4), 243-246.
Pirozzolo, E J., Hansch, E. C., Mortimer, J. A., Webster, D. D., & Kuskowski, M. A.
(1982). Dementia in Parkinson's disease: A neuropsychological analysis. Brain and
Cognition, 1, 71-83.
Portin, R., Muuriaisniemi, M. L., Joukamaa, M., Saarijarvi, S., Helenius, H., &
Salokangas, R. K. (2001). Cognitive impairment and the 10-year survival prob-
ability of a normal 62-year-old population. Scandinavian Journal of Psychology,
42(4), 359-366.
Poser, C. M., Poser, S., & Paty, D. W. (1984). A revised numerical scoring system for
multiple sclerosis. In C. M. Poser (Ed.), The diagnosis of multiple sclerosis (pp.
234-241). New York: Thieme-Stratton.
Poser, S., Poser, W., Schlaf, G., Firnhaber, W., Lauer, K., Wolter, M., et al. (1986).
Prognostic indicators in multiple sclerosis. Acta Neurologica Scandinavica, 74,
387-392.
Powell, D. H. (1994). Profiles in cognitive aging. Cambridge, MA: Harvard University
Press.
Powell, D. H., Kaplan, E. E, Whitla, D., Weintraub, S., Catlin, R., & Funkenstein, H. H.
(1993). MicroCog: Assessment of Cognitive Functioning. San Antonio, TX: The
Psychological Corporation.
Price, L. R., Tulsky, D., Millis, S., & Weiss, L. (in press). Redefining the factor structure
of the Wechsler Memory Scale-III: Confirmatory factor analysis with cross-
validation. Journal of Clinical and Experimental Neuropsychology.
Prifitera, A., & Dersh, J. (1992). Base rates of the WISC-III diagnostic subtest patterns
among normal, learning-disabled, and ADHD samples. Journal of Psycho-
educational Assessment (WISC-III Monograph], 43-55.
Prifitera, A., & Ledbetter, M. E (1992, November). Normative delayed recall rates
- based on the Wechsler Memory Scale—Revised standardization sample. Paper
presented at the 12th annual meeting of the National Academy of Neuro-
psychology, Pittsburgh, PA.
presented
Prigatano, G. P. (1974, May). Memory deficit in head injured patients. Paper
the Southwes tern Psycholo gical Associati on, El Paso, TX.
at the meeting of
g test for brain
Prigatano, G. P. (1977). The Wechsler Memory Scale is a poor screenin
dysfunction. Journal of Clinical Psychology, 33(3), 772-777.

Al7
References

Prigatano, G. P. (1978). Wechsler Memory Scale: A selective review of the literature

[Special Monograph Supplement]. Journal of Clinical Psychology, 34(4), 816-832.
The Psychological Corporation. (1992). Wechsler Individual Achievement Test. San
Antonio, TX: Author.
The Psychological Corporation. (1999). The Wechsler Abbreviated Scale of Intelligence
Manual. San Antonio, TX: Author.
The Psychological Corporation. (2001). Wechsler Test of Adult Reading. San Antonio,
TX: Author.
The Psychological Corporation. (2002a). Wechsler Individual Achievement Tester
Edition Examiner’s Manual. San Antonio, TX: Author.
The Psychological Corporation. (2002b). Wechsler Individual Achievement Test-2nd
Edition Scoring and Normative Supplement for Grades PreK—12. San Antonio,
TX: Author.
The Psychological Corporation. (2002c). Wechsler Individual Achievement Test-2nd
Edition Supplement for College Students and Adults. San Antonio, TX: Author.
Pugh, K. R., Mencl,W. E., Jenner, A. R., Katz, L., Frost, S. J., Lee, J. R., et al. (2001).
Neurobiological studies of reading and reading disability. Journal of
Communication Disorders, 34(6), 479-492.
Pugh, K. R., Mencl, W. E., Shaywitz, B. A., Shaywitz, S. E., Fulbright, R. K., Constable,
R. T., et al. (2000). The angular gyrus in developmental dyslexia: Task-specific
differences in functional connectivity within posterior cortex. Psychological
Science, 11(1), 51-56.
Putnam, S. H., Ricker, J. H., Ross, S. R., & Kurtz, J. E. (1999). Considering premorbid
functioning: Beyond cognition to a conceptualization of personality in postinjury
functioning. In J. J. Sweet (Ed.), Forensic neuropsychology: Fundamentals and
Practice (pp. 39-81). Lisse, The Netherlands: Swets & Zeitlinger.
Quadfasel, A. E, & Pruyser, P. W. (1955). Cognitive deficit in patients with psycho-
motor epilepsy. Epilepsia, 4, 80-90.
Quinlan, D. M., & Brown, T. E. (1997). Assessment of working memory impairment in
adults with ADHD. Manuscript in preparation.
Randolph, C., Mohr, E., & Chase, T. N. (1993). Assessment of intellectual function in
dementing disorders: Validity of WAIS-R short forms for patients with Alzheimer’s ,

Huntington's, and Parkinson's disease. Journal of Clinical and Experimental

Neuropsychology, 15(5), 743-753.
Rao, S. M., Hammeke, T. A., McQuillen, M. P, Khatri, B. O., & Lloyd, D. (1984).
Memory disturbance in chronic progressive multiple sclerosis. Archives of
Neurology, 41, 625-631.
Raven, J. C. (1976). Standard Progressive Matrices. Oxford, England: Oxford
Psychologists Press.
Reed, R. J., Grant, I., & Rourke, S. B. (1992). Long-term abstinent alcoholics have
normal memory. Alcoholism: Clinical and Experimental Research, 16(4), 677-683.
Reid, D. B., & Kelly, M. P. (1993). Wechsler Memory Scale—Revised in closed head
injury. Journal of Clinical Psychology, 49(2), 245-254.
Reitan, R. M., & Wolfson, D. (1993). Halstead-Reitan Neuropsychological Battery.
Tucson, AZ: Neuropsychology Press.

418
 References

Rey, A. (1941). Lexamen psychologique dans les cas d’encéphalopathie traumatique.

Archives de Psychologie, 28, 286-340.
Rey, A. (1959). Sollicitation de la mémoire de fixation par des mots et des objets pre-
sentés simultanément. Archives de Psychologie, 37, 126-139.
Reynolds, C. R. (1985). Critical measurement issues in learning disabilities. The
Journal of Special Education, 18(4), 451-475.
Reynolds, C. R. (1990). Conceptual and technical problems in learning disability
diagnosis. In C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook of psychological
and educational assessment of children: Intelligence and achievement (pp.
571-592). New York: Guilford Press.
Rhodes, S. S., & Jasinski, D. R. (1990). Learning disabilities in alcohol-dependent
adults: A preliminary study. Journal of Learning Disabilities, 23(9), 551-556.
Richards, T. L., Dager, S. R., Corina, D., Serafini, S., Heide, A. C., Steury, K., et al.
(1999). Dyslexic children have abnormal brain lactate response to reading-related
language tasks. American Journal of Neuroradiology, 20(8), 1393-1398.
Richardson, J. T. E. (1996). Evolving concepts of working memory. In J. T. E.
Richardson, R. W. Engle, L. Hasher, R. H. Logie, E. R. Stoltzfus, & R. T. Zacks (Eds.),
Working memory and human cognition (pp. 3-30). New York: Oxford University
Press.
Richardson, S. A., & Koller, H. (1985). Epidemiology. In A. M. Clarke, A. B. D. Clarke, &
J. M. Berg (Eds.), Mental deficiency: The changing outlook (4th ed., pp. 27-52).
London: Methuen.
Ritchie, K., Artero, S., & Touchon, J. (2001). Classification criteria for mild cognitive
impairment: A population-based validation study. Neurology, 56(1), 37-42.
Rizzo, L., Danion, J. M., Van Der Linden, M., Grange, D., & Rohmer, J. G. (1996).
Impairment of memory for spatial context in schizophrenia. Neuropsychology,
10(3), 376-384.
Roid, G. H., Prifitera, A., & Ledbetter, M. F. (1988). Confirmatory analysis of the factor
structure of the Wechsler Memory Scale—Revised. The Clinical Neuropsychologist,
2(2), 116-120.
Rosenberg, H. M., Ventura, S. J., Maurer, J. D., Heuser, R. L., & Freedman, M. A. (1996).
Births and deaths: United States, 1995. Monthly Vital Statistics Report, 45(3, Suppl.
2). (Preliminary Data from the Centers for Disease Control and Prevention/
National Center for Health Statistics).
Roth, D. L., Conboy, T. J., Reeder, K. P., & Boll, T. J. (1990). Confirmatory factor analysis
of the Wechsler Memory Scale—Revised in a sample of head-injured patients.
Journal of Clinical and Experimental Neuropsychology, 12(4), 834-842.
s of alcoholism. In
Rourke, S. B., & Loberg, T. (1996). The neurobehavioral correlate
IL. Grant & K. M. Adams (Eds.). Neuropsy chologic al assessme nt of neuropsychiatric
‘disorders (2nd ed., pp. 423-485). New York: Oxford University Press.
, S. G, Simmons, A. etal.
Rubia, K., Overmeyer, S., Taylor, E., Brammer, M., Williams
during higher-
(1999). Hypofrontality in attention deficit hyperactivity disorder
function al MRI. America n Journal of Psychiatry,
order motor control: A study with
156(6), 891-896.

419
References

Rubin, H. H., Goldman, J. J., & Rosenfeld, J. G. (1985). A comparison of WISC-R and
WAIS-R IQs in a mentally retarded residential population. Psychology in the
Schools, 22, 392-397.
Rudel, R. G., & Denckla, M. B. (1974). Relation of forward and backward digit repeti-
tion to neurological impairment in children with learning disabilities.
Neuropsychologia, 12, 109-118.
Rugel, R. P. (1974). WISC subtest scores of disabled readers: A review with respect to
Bannatyne’s recategorization. Journal of Learning Disabilities, 7(1), 57-64.
Russell, E. W. (1975). A multiple scoring method for the assessment of complex
memory functions. Journal of Consulting and Clinical Psychology, 43(6), 800-809.
Ryan, C., & Butters, N. (1980). Learning and memory impairments in young and old
alcoholics: Evidence for premature-aging hypothesis. Alcoholism: Clinical and
Experimental Research, 4(3), 288-293.
Ryan, C., & Butters, N. (1983). Cognitive deficits in alcoholics. In B. Kissen & H.
Begleiter, (Eds.). The pathogenesis of alcoholism: Biological factors (pp. 485-538).
New York: Plenum Press.
Ryan, J. J. (1999). Two types of tables for use with the seven-subtest short forms of
the WAIS-III. Journal of Psychoeducational Assessment, 17, 145-151.
Ryan, J. J., Ament, P. A., & Arb, J. D. (2000). Supplementary WMS-III tables for
determining primary subtest strengths and weaknesses. Psychological Assessment,
12(2), 193-196.
Ryan, J. J., & Lewis, C. V. (1988). Comparison of normal controls and recently
detoxified alcoholics on the Wechsler Memory Scale—Revised. The Clinical
Neuropsychologist, 2(2), 173-180.
Ryan, J. J., Lopez, S. J., & Werth, T. R. (1998). Administration time estimates for
WAIS-III subtests, scales, and short forms in a clinical sample. Journal of
Psychoeducational Assessment, 16, 315-323.
Ryan, J. J., Paolo, A. M., & Brungardt, T. M. (1990). Standardization of the Wechsler
Adult Intelligence Scale—Revised for persons 75 years and older. Psychological
Assessment, 2(4), 404-411.
Ryan, J. J., Paolo, A. M., Oehlert, M. E., & Coker, M. C. (1991). Relationship of sex, race,
age, education, and level of intelligence to the frequency of occurrence of a
WAIS-R marker for dementia of the Alzheimer's type. Developmental
Neuropsychology, 7(4), 451-458.
Ryan, J. J., Sattler, J. M., & Lopez, S. J. (2000). Age effects on Wechsler Adult
Intelligence Scale-III subtests. Archives of Clinical Neuropsychology, 15(4),
311-317.
Ryan, J. J., & Ward, L. C. (1999). Validity, reliability, and standard errors of measure-
ment for two seven-subtest short forms of the Wechsler Adult Intelligence
Scale-III. Psychological Assessment, 11(2), 207-211.
Ryan, L., Clark, C. M., Klonoff, H., Li, D., & Paty, D. (1996). Patterns of cognitive
impairment in relapsing-remitting multiple sclerosis and their relationship to
neuropathology on magnetic resonance images. Neuropsychology, 10(2), 176-193.
Saklofske, D. H., Hildebrand, D. K., & Gorsuch, R. L. (2000). Replication of the factor
structure of the Wechsler Adult Intelligence Scale-Third Edition with a Canadian
sample. Psychological Assessment, 12(4), 436-439.

420
 References

Salmon, D. P, Butters, N., & Heindel, W. C. (1993). Alcoholic dementia and related
disorders. In R. W. Parks, R. E Zec, & R. S. Wilson (Eds.), Neuropsychology of
Alzheimer’s disease and other dementias (pp. 186-209). New York: Oxford
University Press.
Sandoval, J., Sassenrath, J., & Penaloza, M. (1988). Similarity of WISC-R and WAIS-R
scores at age 16. Psychology in the Schools, 25(4), 373-379.
Sattler, J. M. (1992). Assessment of children: WISC-III and WPPSI-R supplement. San
Diego, CA: Author
Sattler, J. M., & Ryan, J. J. (1999). Assessment of children: WAIS-III supplement. La
Mesa, CA: Jerome M. Sattler Publishing.
Saykin, A. J., Gur, R. C., Gur, R. E., Mozley, D., Mozley, L. H., Resnick, S. M., et al.
(1991). Neuropsychological function in schizophrenia: Selective impairment in
memory and learning. Archives of General Psychiatry, 48, 618-624.
Saykin, A. J., Shtasel, D. L., Gur, R. E., Kester, D. B., Mozley, L. H., Stafiniak, P, et al.
(1994). Neuropsychological deficits in neuroleptic naive patients with first-
episode schizophrenia. Archives of General Psychiatry, 51, 124-131.
Schoenberg, B. S. (1987). Epidemiology of movement disorders. In C. D. Marsden &
S. Fahn (Eds.), Movement disorders (pp. 17-32). Boston: Butterworth.
Schuckit, M. A., Smith, T. L., Anthenelli, R., & Irwin, M. (1993). Clinical course of alco-
holism in 636 male inpatients. American Journal of Psychiatry, 150(5), 786-792.
Schultz, R. T., Cho, N. K., Staib, L. H., Kier, L. E., Fletcher, J. M., Shaywitz, S. E., et al.
(1994). Brain morphology in normal and dyslexic children: The influence of sex
and age. Annals of Neurology, 35(6), 732-742.
Schwean, V. L., Saklofske, D. H., Yackulic, R. A., & Quinn, D. (1992). WISC-III perfor-
mance of ADHD children. Journal of Psychoeducational Assessment {WISC-III
Monograph], 56-70.
Schweinberger, S. R., Buse, C., Freeman, R. B., Jr., Schonle, P. W., & Sommer, W. (1992).
Memory search for faces and digits in patients with unilateral brain lesions.
Journal of Clinical and Experimental Neuropsychology, 14(5), 839-856.
Seidman, L. J., Yurgelun—Todd, D., Kremen, W. S., Woods, B. T., Goldstein, J. M.,
Faraone, S. V,, et al. (1994). Relationship of prefrontal and temporal lobe MRI mea-
sures to neuropsychological performance in chronic schizophrenia. Biological
Psychiatry, 35, 235-246.
Semrud-Clikeman, M., Steingard, R. J., Filipek, P, Biederman, J., Bekken, K., &
Renshaw, P. E (2000). Using MRI to examine brain-behavior relationships in males
with attention deficit disorder with hyperactivity. Journal of the American Academy
of Child and Adolescent Psychiatry, 39(4), 477-484.
Shakow, D. (1963). Psychological deficit in schizophrenia. Behavioral Science, 8(4),
275-305.
Shapiro, R. M. (1993). Regional neuropathology in schizophrenia: Where are we?
Where are we going? Schizophrenia Research, 10(3), 187-239.
Sheikh, J. I., & Yesavage, J. A. (1986). Geriatric Depression Scale (GDS): Recent evi-
dence and development of a shorter version. Clinical Gerontology, 5, 165-173.
Shepard, L. A. (1980). An evaluation of the regression discrepancy method for identi-
fying children with learning disabilities. Journal of Special Education, 14(1), 79-91.

421
References

Shimamura, A. P, & Squire, L. R. (1984). Paired-associate learning and priming effects

in amnesia: A neuropsychological study. Journal of Experimental Psychology:
General, 113(4), 556-570.
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater relia-
bility. Psychological Bulletin, 86(2), 420-428.
Shum, D. H. K., McFarland, K., & Bain, J. D. (1994). Effects of closed-head injury on
attentional processes: Generality of Sternberg’s additive factor method. Journal of
Clinical and Experimental Neuropsychology, 16(4), 547-555.
Silverstein, A. B. (1968). Validity of a new approach to the design of WAIS, WISC, and
WPPSI short forms. Journal of Consulting and Clinical Psychology, 32(4), 478-479.
Silverstein, A. B. (1981). Reliability and abnormality of test score differences. Journal
of Clinical Psychology, 37(2), 392-394.
Silverstein, A. B. (1982). Two- and four-subtest short forms of the Wechsler Adult
Intelligence Scale—Revised. Journal of Consulting and Clinical Psychology, 50(3),
415-418.
Simon, C. L., & Clopton, J. R. (1984). Comparison of WAIS and WAIS-R scores of
mildly and moderately mentally retarded adults. American Journal of Mental
Deficiency, 89(3), 301-303.
Smith, G. E., Ivnik, R. J., Malec, J. E, Kokmen, E., Tangalos, E. G., & Kurland, L. T.
(1992). Mayo’s older Americans normative studies (MOANS): Factor structure of a
core battery. Psychological Assessment, 4(3), 382-390.
Smith, G. E., Ivnik, R. J., Malec, J. E, Petersen, R. C., Kokmen, E., & Tangalos, E. G.
(1994). Mayo cognitive factors scales: Derivation of a short battery and norms for
factor scores. Neuropsychology, 8(2), 194-202.
Smith, M. L., & Milner, B. (1989). Right hippocampal impairment in the recall of spa-
tial location: Encoding deficit or rapid forgetting* Neuropsychologia, 27(1), 71-81.
Spalletta, G., Pasini, A., Pau, E, Guido, G., Menghini, L., & Caltagirone, C. (2001).
Prefrontal blood flow dysregulation in drug naive ADHD children without struc-
tural abnormalities. Journal of Neural Transmission, 108(10), 1203-1216.
Sparrow, S. S., Balla, D. A., & Cicchetti, D. V. (1984). Vineland Adaptive Behavior
Scales. Circle Pines, MN: American Guidance Service.
Spearman, C. (1904). “General intelligence,” objectively determined and measured.
American Journal of Psychology, 15, 201-293.
Spearman, C. (1970). The abilities of man: Their nature and measurement. New York:
AMS Press. (Original work published 1932)
Spencer, T., Biederman, J., Wilens, T., & Faraone, S. V. (1994). Is attention-deficit
hyperactivity disorder in adults a valid disorder? Harvard Review of Psychiatry, 1,
326-335.
Spitz, H. H. (1972). Note on immediate memory for digits: Invariance over the years.
Psychological Bulletin, 78(3), 183-185.
Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests:
Administration, norms, and commentary (2nd ed.). New York: Oxford University
Press.
Spruill, J. (1991). A comparison of the Wechsler Adult Intelligence Scale—Revised
with the Stanford-Binet Intelligence Scale (4th edition) for mentally retarded
adults. Psychological Assessment: A Journal of Consulting and Clinical Psychology,
3(1), 1-3.

422
 References

Squire, L. R. (1987). Memory and brain. New York: Oxford Universi

ty Press.
Squire, L. R. (1992). Memory and the hippocampus: A synthesis from
findings with
rats, monkeys and humans. Psychological Review, 99, 195-231.
Squire, L., & Butters, N. (Eds.). (1992). Neuropsychology of memory (2nd
ed.). New
York: Guilford Press.
Stanley, J. C. (1971). Reliability. In R. L. Thorndike (Ed.), Educational measurement
(2nd ed., pp. 356-442). Washington, DC: American Council on Education.
Sternberg, R. J. (1980). Factor theories ofintelligence are all right almost. Educational
Researcher, 9, 6-13, 18.
Sternberg, R. J. (1995). In search of the human mind. Fort Worth, TX: Harcourt Brace
College.
Storandt, M. (1994). General principles of assessment of older adults. In M. Storandt
& G. R. VandenBos (Eds.), Neuropsychological assessment of dementia and depres-
sion in older adults: A clinician's guide (pp. 7-32). Washington, DC: American
Psychological Association.
Storandt, M., & Hill, R. D. (1989). Very mild senile dementia of the Alzheimer type:
Il. Psychometric test performance. Archives of Neurology, 46, 383-386.
Stuss, D. T., Alexander, M. P.,, Palumbo, C. L., Buckle, L., Sayer, L., & Pogue, J. (1994).
Organizational strategies of patients with unilateral or bilateral frontal lobe injury
in word list learning tasks. Neuropsychology, 8(3), 355-373.
Stuss, D. T., & Benson, D. E (1984). Neuropsychological studies of the frontal lobes.
Psychological Bulletin, 95(1), 3-28.
Stuss, D. T., & Benson, D. E (1986). The frontal lobes. New York: Raven Press.
Stuss, D. T., Eskes, G. A., & Foster, J. K. (1994). Experimental neuropsychological stud-
ies of the frontal lobe functions. In E Boller & J. Grafman (Eds.), Handbook of neu-
ropsychology (Vol. 9, pp. 149-185). Amsterdam: Elsevier.
Sullivan, P. M., & Schulte, L. E. (1992). Factor analysis of WISC-R with deaf and hard-
of-hearing children. Psychological Assessment, 4(4), 537-540.
Swanson, H. L. (1993). Working memory in learning disability subgroups. Journal of
Experimental Child Psychology, 56(1), 87-114.
Swanson, H. L., Cochran, K. E, & Ewers, C. A. (1990). Can learning disabilities be
determined from working memory performance? Journal of Learning Disabilities,
23(1), 59-67.
Swirsky-Sacchetti, T., Mitchell, D. R., Seward, J., Gonzales, C., Lublin, E, Knobler, R.,
et al. (1992). Neuropsychological and structural brain lesions in multiple sclerosis:
A regional analysis. Neurology, 42, 1291-1295.
Tamkin, A. S., & Dolenz, J. J. (1990). Cognitive impairment in alcoholics. Perceptual
and Motor Skills, 70, 816-818.
Tanaka, J. S. (1993). Multifaceted conceptions of fit in structural equation models. In
K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 10-39).
Newbury Park, CA: Sage.
Tarter, R. E. (1973). An analysis of cognitive deficits in chronic alcoholics. Journal of
Nervous and Mental Disease, 157(2), 138-147.
Tarter, R. E., Hegedus, A. M., Goldstein, G., Shelly, C., & Alterman, A. (1984). Aa
Adolescent sons of alcoholics: Neuropsychological and personality characteristics.
Alcoholism: Clinical and Experimental Research, 8(2), 216-222.

423
References

Taylor, A. E., Saint-Cyr, J. A., & Lang, A. E. (1990). Memory and learning in early
Parkinson's disease: Evidence for a “frontal-lobe syndrome.” Brain and Cognition,
13, 211-232.
Taylor, M. J., & Heaton, R. K. (2001). Sensitivity and specificity of the WAIS—III/
WMS-III demographically corrected factor scores in neuropsychological assess-
ment. Journal of the International Neuropsychological Society, 7(7), 867-874.
Teasdale, G., & Jennett, B. (1974). Assessment of coma and impaired consciousness.
Lancet, 2, 81-83.
Thorndike, E. L., Lay, W., & Dean, P. R. (1909). The relation of accuracy in sensory dis-
crimination to general intelligence. American Journal of Psychology, 20, 364-369.
Thorndike, R. L., Hagen, E. P, & Sattler, J. M. (1986). Stanford-Binet Intelligence Scale:
Fourth Edition: Technical manual. Chicago: Riverside.
Trahan, D. E., Larrabee, G. J., Quintana, J. W., Goethe, K. E., & Willingham, A. C.
(1989). Development and clinical validation of an expanded paired associate test
with delayed recall. The Clinical Neuropsychologist, 3(2), 169-183.
Trenerry, M. R., Jack, C. R., Jr., Cascino, G. D., Sharbrough, EF W., & Ivnik, R. J. (1996).
Sex differences in the relationship between visual memory and MRI hippocampal
volumes. Neuropsychology, 10(3), 343-351.
Tréster, A. I., Butters, N., Salmon, D. P, Cullum, C. M., Jacobs, D., Brandt, J., et al.
(1993). The diagnostic utility of savings scores: Differentiating Alzheimer’s and
Huntington's diseases with logical memory and visual reproduction tests. Journal
of Clinical and Experimental Neuropsychology, 15(5), 773-788.
Trott, L. A. (1984). Providing school psychological services to hearing-impaired stu-
dents in New Jersey. American Annals of the Deaf, 129, 319-323.
Tucker, L. R., & Lewis, C. (1973). A reliability coefficient for maximum likelihood
factor analysis. Psychometrika, 38(1), 1-10.
Tulsky, D. S., Ivnik, R. J., Price, L. R., & Wilkins, C. (in press). Assessment of cognitive
functioning with WAIS-III and WMS-III: Development of a 6-factor model. In D. S.
Tulsky, et al. (Eds.), Clinical Interpretation of the WAIS-III and WMS-III. San Diego:
Academic Press.
Tulsky, D., & Price, L. R. (in press). The joint WAIS—III and WMS-III factor structure:
Part I. Development of a 6-factor model of cognitive functioning. Journal of
Psychological Assessment.
Tulsky, D. S., Rolfhus, E. L., & Zhu, J. (2000). Two-tailed versus one-tailed base rates of
discrepancy scores in the WAIS-III. Clinical Neuropsychologist, 14(4), 451-460.
Tulsky, D. S., Saklofske, D. H., Wilkins, C., & Weiss, L. (2001). Development of a
general ability index for the Wechsler Adult Intelligence Scale-Third Edition.
Psychological Assessment, 13(4), 566-571.
Tulsky, D. S., & Zhu, J. (2000). Could test length or order affect scores on
Letter-Number Sequencing of the WAIS-III and WMS-III? Ruling out effects of
fatigue. Clinical Neuropsychologist, 14(4), 474-478.
Tulsky, D., Zhu, J., & Vasquez, C. (1998). The clinical utility of the WAIS-III IQ and
index scores in patients with neuropsychological disorders. Annual Convention of
the International Neuropsychological Society, Honolulu. Journal of the
International Neuropsychological Society, 3
Tuokko, H. A., Frerichs, R. J., & Kristjansson, B. (2001). Cognitive impairment, no
dementia: Concepts and issues. International Psychogeriatrics, 13(1), 183-202.

424
 References

Turner, M. L., & Engle, R. W. (1989). Is working memor

y capacity task dependent?
Journal of Memory and Language, 28, 127-154,
U.S. Bureau of the Census. (1995). Current population survey,
March 1995
[Machine-readable data file]. Washington, DC: U.S. Bureau
of the Census
(Producer/ Distributor).
U.S. Bureau of the Census. (1997). P-251095. Washington, DC:
Author,
Ward, L. C. (1990). Prediction of verbal, performance, and full
scale IQs from seven
subtests of the WAIS-R. Journal of Clinical Psychology, 46(4), 436-440.
Ward, L. C., Ryan, J. J., & Axelrod, B. N. (2000). Confirmatory factor
analyses of the
WAIS-III standardization data. Psychological Assessment, 12(3), 341-345.
Watson, B. U., Sullivan, P. M., Moeller, M. P, & Jensen, J. K. (1982). Nonverba
l intelli-
gence and English language ability in deaf children. Journal of Speech and
Hearing
Disorders, 47, 199-204.
Watson, C. G., & Klett, W. G. (1974). Are nonverbal IQ tests adequate substitute
s for
the WAIS? Journal of Clinical Psychology, 30(1), 55-57.
Wechsler, D. (1939). Wechsler—Bellevue Intelligence Scale. New York: The Psychologi
cal
Corporation.
Wechsler, D. (1944). The measurement of adult intelligence (3rd ed.). Baltimore:
Williams & Wilkins.
Wechsler, D. (1945). A standardized memory scale for clinical use. The Journal of
Psychology, 19, 87-95.
Wechsler, D. (1950). Cognitive, conative, and non-intellective intelligence. American
Psychologist, 5, 78-83.
Wechsler, D. (1955). Wechsler Adult Intelligence Scale. New York: The Psychological
Corporation.
Wechsler, D. (1975). Intelligence defined and undefined: A relativistic appraisal.
American Psychologist, 30, 135-139.
Wechsler, D. (1981). Wechsler Adult Intelligence Scale—Revised. San Antonio, TX: The
Psychological Corporation.
Wechsler, D. (1987). Wechsler Memory Scale—Revised. San Antonio, TX: The
Psychological Corporation.
Wechsler, D. (1991). Wechsler Intelligence Scale for Children—Third Edition. San
Antonio, TX: The Psychological Corporation.
Wechsler, D. (1997a). Wechsler Adult Intelligence Scale—Third Edition. San Antonio,
TX: The Psychological Corporation.
Wechsler, D. (1997b). Wechsler Memory Scale—Third Edition. San Antonio, TX: The
Psychological Corporation.
Weinberger, D. R., Berman, K. FE, & Zec, R. FE (1986). Physiologic dysfunction of dorso-
lateral prefrontal cortex in schizophrenia: I. Regional cerebral blood flow evidence.
Archives of General Psychiatry, 43, 114-126.
Weintraub, S., & Mesulam, M.-M. (1985). Mental state assessment of young and
elderly adults in behavioral neurology. In M.-M. Mesulam (Ed.), Principles of
behavioral neurology (pp. 71-123). Philadelphia: FE A. Davis.
Welsh, K., Butters, N., Hughes, J., Mohs, R., & Heyman, A. (1991). Detection of abnor-
mal memory decline in mild cases of Alzheimer’s disease using CERAD neuropsy-
chological measures. Archives of Neurology, 48, 278-281.

425
References

Wilde, N., Strauss, E., Chelune, G. J., Loring, D. W., Martin, R. C., Hermann, B. A,
et al. (2001). WMS-III performance in patients with temporal lobe epilepsy:
Group differences and individual classification. Journal of the International
Neuropsychological Society, 7, 881-891.
Woltz, D. J. (1988). An investigation of the role of working memory in procedural skill
acquisition. Journal of Experimental Psychology: General, 117(3), 319-331.
Woodard, J. L. (1993). Confirmatory factor analysis of the Wechsler Memory Scale—
Revised in a mixed clinical population. Journal of Clinical and Experimental
Neuropsychology, 15(6), 968-973.
Woodcock, R. W. (1990). Theoretical foundations of the WJ—-R measures of cognitive
ability. Journal of Psychoeducational Assessment, 8, 231-258.
Woodcock, R. W. (1997). The Woodcock-Johnson Tests of Cognitive Ability—Revised.
In D. P. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contemporary intellectual
assessment: Theories, tests, and issues (pp. 230-246). New York: Guilford Press.
Worden, PE. (1986). Prose comprehension and recall in disabled learners. In S. J.
Ceci (Ed.), Handbook of Cognitive, Social and Neuropsychological Aspects of
Learning Disabilities (Vol. 1, pp. 241-261). New York: Erlbaum.
Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M., et al., (1983).
Development and validation of a geriatric depression screening scale: A prelimi-
nary report. Journal of Psychiatric Research, 17(1), 37-49.
Zec, R. E (1993). Neuropsychological functioning in Alzheimer’s disease. In R. W.
Parks, R. E Zec, & R. S. Wilson (Eds.), Neuropsychology of Alzheimer’s disease and
other dementias (pp. 3-80). New York: Oxford University Press.
Zhu, J., & Tulsky, D. S. (2000). Co-norming of the WAIS-III and WMS-IIE: Is there a
test-order effect on IQ and memory scores? Clinical Neuropsychologist, 14(1), 1.
Zhu, J., Tulsky, D. S., Price, L., & Chen, H. Y. (2001). WAIS-III reliability data for clinical
groups. Journal of the International Neuropsychological Society, 7, 862-866.

426
«
»*
 Ml
ISBN 015898144-8

9"780158"981444