Table 4
Detailed review of tests that include traumatic brain injury in the test purpose, applications, or standardization sample. If not stated, test/subtest did
not meet criteria. See Appendix B for definitions for evaluation categories. DNT = Did not test. Criteria for Inter-Rater Reliability = Strict r > .90,
Kappa > .80; Relaxed r > .80, Kappa > .70. Criteria for Internal Consistency (split half) = Strict Cronbach Alpha > .90, Relaxed > .80. Criteria for
Test-Retest reliability = Strict r > .90, Relaxed r > .80.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Aphasia Diagnostic
Profiles
(ADP; Helm-
Estabrooks, 1992)
DNT
Suggested
training in teams.
7/9 subtests met
strict criterion, 1/9
subtests met
relaxed criterion.
1/12 subtests met
strict criterion,
3/12 subtests met
relaxed criterion.
Author is an expert in
aphasiology.
Based on the premise that
individuals with acquired
brain damage will have a
wide range of linguistic
problems and a wide range
of severity.
“Includes tasks that have a
range of difficulty to allow
clinician to identify severely
impaired and relatively
spared language skills and
those in between.”
Designed to classify/profile
aphasia type based on
preserved language skills.
Includes measures of non-
verbal characteristics of
aphasia, e.g. behavioral
profile and alternate
communication profile.
19 subjects with aphasia
improved scores on 9
subtests at re-testing at
11 days after initial test,
as expected during
recovery from stroke.
Intercorrelations among
scores on subtests and
coefficients are mostly in
the moderate to
moderately high range
(.40-.80).
Four-factor solution
generated on 217
subjects; 54% of variance
accounted for by first
factor, described as
general language
expression factor.
DNT
American Speech
Language Hearing
Association
Functional
Assessment of
Communication Skills
for Adults
(ASHA FACS;
Frattali, Thompson,
Holland, Wohl, &
Ferketic, 1995)
Completed by 2
examiners for
same subject
within 48 hours
after a minimum
of 3 subject-
examiner
contacts, for 35
subjects. Overall
scores r=.95;
Social and Basic
Needs r=.92,
Reading/writing/
#’s and Planning
r= .89. For
Qualitative
Dimensions,
overall r=.88,
Adequacy .82,
Comm. Sharing
.84,
Appropriateness
Majority of items
(all but 3)
covered full 7-pt
range of scale –
including in mild
group (supports
sensitivity).
Inter-item
correlations
consistently high;
also between
items and total
score except for
4 items; mean
correlation =.82.
Analyzed item-
total correlations
for each
diagnostic group
and all but 4
(aphasia) and 3
(TBI) items had
Calculated intra-
rater reliability (i.e.
test-retest) on 38
subjects within 2-5
days: all r’s >.94.
For Qualitative
Dimensions
overall r=.99. All 4
dimensions >.94.
Effects of race, education
and sex considered,
though generalizability
limited by “insufficient or
unbalanced samples”.
Education and sex were
significant factors for
Reading/writing, Daily
Planning, Overall
independence score, and
for Appropriateness,
Promptness and Adequacy
in qualitative domains.
Sign differences by sex for
qualitative adequacy.
Consumer survey used to
rate each item in re: its
“importance to functional
communication” – 69% or
more of 179 respondents
rated each item as
“somewhat important” or
Factor analysis: all 44
items loaded on one
factor that accounted for
59% of the total variance;
4 factor solution
accounted for “more than”
73% of total variance:
factors appeared to be 4
assessment domains –
Social communication,
Communication of basic
needs, Reading/writing
and # concepts, Daily
planning.
Overall communication
independence scores
correlated .93 with overall
qualitative scores.
Compared to:
1) FIM scales for
comprehension,
expression,
problem solving,
memory, social
interaction: a) for
aphasia - all
significant but
modest; range
=.53-.87; b) for
cognitive-
communication
disorders – all
significant;
range=.61-.87.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Dimensions,
overall r=.88,
Adequacy .82,
Comm. Sharing
.84,
Appropriateness
.77, Promptness
.72.
total correlations
for each
diagnostic group
and all but 4
(aphasia) and 3
(TBI) items had
moderate-high
correlations.
Items with low
correlations kept
anyway because
they were
considered
pragmatically
important and
important to
communication
with family and
friends.
Communication
independence
mean scores for
domains
correlated with
overall score .89-
.99. Qualitative
domain scores
correlated with
overall score .91-
.96
rate each item in re: its
“importance to functional
communication” – 69% or
more of 179 respondents
rated each item as
“somewhat important” or
“very important”. 31
clinicians rated usability
and most had positive
feedback (detailed results
provided).
7 peer reviewers reviewed
the test and made largely
positive comments and
suggestions for
improvement (details
provided). Published
version has some changes
based on this – 3 deletions
and one wording change in
instructions.
correlated .93 with overall
qualitative scores.
Significant difference
between cognitive-
communication and
aphasia groups for Social
communication, and
Reading/writing and #
concepts.
Significant difference
between fluent and
nonfluent aphasia for all 4
communication
independence domains.
Qualitative domains also
significantly better both in
cognitive-communication
disorders than aphasia
and in less severely
impaired subjects, except
Appropriateness
(explained more by
severity)
2) WAB subtest
and AQ scores for
subjects with
aphasia: all
significant; range
=.38-.81.
3) Scales of
Cognitive Ability
for Traumatic
Brain Injury for
subjects with
cognitive-
communication
disorders: all
significant;
range=.53-.79
4) Correlated
domain and
overall
independence
scores with Rating
of Communication
Abilities from
families and
clinicians.
Correlations
ranged from .43-
.64. Families did
not report higher
measures than
clinicians (range of
correlations = .49-
.73).
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Behavior Rating
Inventory of Executive
Function
(BRIEF; Gioia,
Isquith, Guy, &
Kenworthy, 1996)
Parent vs.
teacher ratings
had low
correlation (.30-
.50), consistent
with the literature
and probably
related to
differences in
context in which
ratings were
made.
All index scores
and 19/32
subscale scores
for Parent and
Teacher version
for Normative
and combined
clinical samples
(i.e., 8 subscales
per group) met
strict criterion.
Remaining 13/32
subscales met
relaxed criterion.
Re-tested over 2-6
week period.
Teacher form:
Normative sample
met strict criterion
for index scores,
relaxed criterion
for all subscales.
Clinical sample
DNT.
Parent form: met
relaxed criterion
for all index scores
and 4/8 subscale
scores for both
Normative and
Clinical samples.
Developed based on
review of literature and
identification of domains of
executive function.
Domains then reviewed by
colleagues in
neuropsychology and were
the basis for the BRIEF
normative edition. Revised
based on factor analysis
from first stage of
standardization. Designed
to be easy to administer
and score and yield
clinically useful information.
Each scale “built from
items whose content
reflects the intended
domain.” (p. 35) Items
collected from high-
frequency issues raised in
clinical interviews rather
than generated by authors.
Inter-rater agreement on
item clusters determined by
calculating responses of
neuropsychologists with
expertise in the area.
Tryouts conducted.
Inconsistency and
negativity scales added to
address response bias. No
significant effect of ethnic
group membership. Low
SES scores associated
with lower BRIEF scores.
Other tests reviewed to
ensure completeness of
scales and divergence
from existing measures.
Age groupings determined
by examination of data
(although developmental
trends not evident on
visual inspection of those
data).
Factor analysis on
normative sample,
combined clinical samples
showed 2 factor solution;
inclusion of related
measures (i.e. Child
Behavior
Checklist/Teacher Rating
Form, Behavior
Assessment System for
Children, ADHD Rating
Scale-IV) for normative
and clinical samples
showed discriminant
validity (i.e. added new
factors where constructs
were hypothesized to
differ, loaded on same
factors where they were
hypothesized to be
similar).
Studied on children with
Attention Deficit
Hyperactivity Disorder
(ADHD), brain injury,
Tourette’s Disorder (with
and without ADHD),
Reading Disorder, low
birth weight, high
functioning autism,
pervasive developmental
disorders, mental
retardation, and children
with documented frontal
vs. extrafrontal lesions.
Overall, discriminated
between controls and
other groups, except for
children with
Phenylketenuria and
Mental retardation –
groups differed on only 2
and 1 score respectively.
No other executive
function tests of
this type existed,
so examined
patterns of
relationships by
correlating
individual
subscales with
related measures
e.g., ADHD Rating
Scale-IV, Child
Behavior
Checklist,
Teacher’s Report
Form, Behavior
Assessment
System for
Children, Connors’
Ratings Scale.
Correlations were
high where
constructs were
hypothesized to
converge (e.g.,
Inhibit cluster vs.
Hyperactivity-
Impulsivity on the
ADHD-IV scale)
and low where
constructs were
hypothesized to
differ (e.g.,
between Inhibit
and Somatic
Complaints on
Child Behavior
Checklist).
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
between controls and
other groups, except for
children with
Phenylketenuria and
Mental retardation –
groups differed on only 2
and 1 score respectively.
Brief Test of Head
Injury
(BTHI; Helm-
Estabrooks & Hotz,
1991)
DNT
Severity score
and total score
met strict
criterion; 5/7
subtests met
relaxed criterion.
19/53 patients with
TBI retested within
57 days of injury
(mean time
between testing =
19.1 days): met
relaxed criterion
for BTHI total raw
score.
18 patients with
TBI in
standardization
sample were
retested within 86
days after injury
(mean time
between testing =
16.6 days). Did not
meet criterion:
r=.46 for BTHI
total raw score
(may be consistent
with early stage of
recovery).
Constructed by experts
based on observed clinical
need for quick probes of a
variety of skills in a wide
range of patients.
Targeted areas were
chosen based on literature
on sequelae from TBI.
Designed to be used with
patients who are evolving
quickly and have limited
tolerance for test
administration.
Raw score accurately
classified 100% of 41
uninjured subjects and
85% of 52 patients with
TBI when testing within 30
days of injury with a cutoff
score of 43. Within 7
days of testing, test
accurately classified
100% of uninjured
subjects and 64% of
subjects with TBI, using a
cutoff score of 38.
No significant correlation
with age and time post-
onset.
Total raw scores
differed
significantly
between patients
classified via
Glasgow Coma
Scale (GCS)
scores as severe
vs. mild, and
moderate vs. mild,
but not moderate
vs. severe.
Mean scores
differed across
levels of Rancho
Los Amigos Levels
of Cognitive
Functioning (LCF)
except IV and V in
193 subjects in
standardization
sample.
Significant
correlation
between total
score and GCS at
admission, GCS at
test administration,
LCF at test
administration.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Communication
Activities of Daily
Living, Second
Edition
(CADL-2; Holland,
Frattali, & Fromm,
1999)
Stanine scores
calculated by 2
ProEd Inc.
research staff for
30 scored
protocols, met
strict criterion
(though they
didn’t have two
people actually
score one
examinee).
Met strict
criterion.
Also calculated
SEM=sqr(1-
alpha) for 68%
and 95%
confidence
intervals.
Met relaxed
criterion for adults
with chronic
aphasia at 2-32
week test-retest
interval.
Test designed to answer
specific questions
regarding functional
communication (e.g. What
do normal readers read?
What are some elementary
adaptive strategies for
clarifying
miscommunications?).
10 certified speech
language pathologists
answered questionnaires
about scope and
appropriateness of items,
categories, stimulus
pictures.
Item discrimination (i.e.
extent to which each item
correlates with total score)
analyzed using point-
biserial correlation
technique and 49/50 items
met .2 criterion for
significant coefficient.
Significant differences
between people with
aphasia and those
without.
Item discriminant validity
high (see point bi-serial
correlations).
Moderate negative
relationship
between Global
Communication
Disorder ratings
(1=mild,
2=moderate,
3=severe) and test
scores of 127
participants in
standardization
sample (r=-.50)
Comprehensive
Assessment of
Spoken Language
(CASL; Carrow-
Woolfolk, 1999)
DNT
Internal
consistency for
15 subtests and 6
index scores for
the 12 age
groups range =
.64-.94; 37/252
(i.e., across all
ages, subtests,
and index scores)
met strict
criterion; 57/252
met relaxed
criterion; core
composite met
strict criterion
across all ages;
all index scores
met strict criterion
except 3
receptive indexes
(age 7, 8, 9) that
met relaxed
criterion.
Test administered
twice to 148
randomly selected
examinees in 3
age groups: 5.0-
6.11 (n=41), 8.0-
10.11 (n=38), and
14.0-16.11 (n=69);
interval between
tests 7-109 days,
median=6 weeks.
All Core
Composites met
strict criterion; 4/5
Index Scores at
age 8-10 and 3/3
at age 14-16 met
strict criterion, 1/5
at age 8-10 met
relaxed criterion.
Test purports to measure
oral language; subtest
content and structure is
consistent with test model.
Developmental
progression of scores: a
table of means and
standard deviations of test
raw scores for the
standardization sample by
age “reveals steady
increases in mean raw
scores throughout the age
range, with greater
increases in the early
years and more gradual
changes in later years”.
One discrepancy noted in
table: Nonliteral language
subtest score for 19.0-
21.11 was 4.3 (typo in
manual?).
Intercorrelations among
subtests: coefficients
range from .30 to .79.
Factor structure of
Category and Processing
Indexes Subtests
developed and organized
on the basis of the
Integrative Language
Theory.
Test of Auditory
Comprehension of
Language -
Revised (ages 5.0-
5.11) and CASL
correlations
ranged from .43
on syntax
construction to .76
on pragmatic
judgment.
Oral and Written
Language Scales
(ages 7.10-10.11)
vs. CASL
correlations
ranged from .34
on nonliteral
subtest to .89 on
the syntactic and
expressive
indices; Peabody
Picture Vocabulary
Test (Third
Edition) and
Expressive One
Word Vocabulary
Test (ages 7.1-
10.11) vs. CASL
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
met relaxed
criterion.
Category and Processing
Indexes Subtests
developed and organized
on the basis of the
Integrative Language
Theory.
Factor analysis results
consistent with design –
e.g., 1 factor model for 3-6
and no index scores for
this age, receptive vs.
expressive factors for 7-
21.
The following groups
scored significantly lower
than age-matched
controls: 50 children
previously diagnosed with
language delay, 42
children with language
impairment, 44 children
and adolescents with
Mental Retardation, 30
12-18 year olds with a
learning disability in
reading, and a group of
children with a variety of
learning disabilities
Scores of 50 8-10 year old
children with speech
impairments were not
significantly lower than the
control group.
Picture Vocabulary
Test (Third
Edition) and
Expressive One
Word Vocabulary
Test (ages 7.1-
10.11) vs. CASL
correlations
ranged from .45
on syntax
construction to .85
on
lexical/semantic
index.
Kaufman Brief
Intelligence Test
(ages 14.0-17.9)
vs. CASL
correlations
ranged from .40
on pragmatic
judgment to .81 on
lexical/semantic
index.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Clinical Evaluation of
Language
Fundamentals – Third
Edition
(CELF-3; Semel,
Wiig, & Secord, 1995)
Compared 2
randomly-
selected raters
on 590
Formulated
Sentences and
599 Word
Associations
record forms at
ages 6, 11 and
16 years.
Formulated
sentences: met
strict criterion for
age 6, relaxed
criterion for age
11, r=.70 for age
16.
Word
Associations:
met strict
criterion for all
three ages.
Receptive
language: met
relaxed criterion
for 4/12 age
groups, strict
criterion for 8/12
age groups.
Expressive
language: met
relaxed criterion
for 3/12 age
groups, strict
criterion for 9/12
age groups.
Total language:
met strict criterion
for all 12 age
groups.
Individual
subtests variable:
range from .54-
.91 across
subtests and age
groups.
152 examinees
ages 7, 10, and 13
years re-tested
after 1-4 weeks.
Receptive
language: met
relaxed criterion
for 2/3 age groups;
r=.77 for age 13.
Expressive
language: met
strict criterion for
2/3 age groups,
r=.77 for age 13.
Total language:
met strict criterion
for 2/3 age groups,
relaxed criterion
for 1/3 age groups.
Individual subtests
variable: range
from .45-.94
across subtests
and age groups.
Choice of subtests is
based on a model of
language (form and
content), and expressive
vs. receptive language.
Includes subtests that
“focus on memory of
spoken language,”
although, purpose is not to
test memory.
In third edition, “gender
and ethnic biases have
been eliminated.”
Each subtest based on
relevant literature.
Factor structure supports
single factor measured by
test (i.e., receptive and
expressive language are
not differentiated in factor
structure).
Compared children with
vs. without diagnosed
language disorders,
matched for age, gender,
demographic
characteristics (n=272).
Test correctly classified
71.3% of students as LD
vs. non-LD.
Correlations
between CELF-3
composite
standard scores
and Wechsler
Intelligence Scales
for Children
Verbal,
Performance and
Full-Scale IQ
ranged from .56-
.75.
CELF-3 classified
students with TBI
who were
receiving speech
and language
services as LD but
not students with
communication
disorders who
were not receiving
services; subtest
score variability
not significantly
greater in students
with TBI (Turkstra,
1999).
Cognitive Linguistic
Quick Test
(CLQT; Helm-
Estabrooks, 2001)
Clock Drawing: 2
raters, 170 tests,
met relaxed
criterion.
Generative
Naming: 2
raters, 81 tests,
met strict
criterion.
DNT
Normal sample
(n=46) re-tested
80-140 days from
first test:
Across 10 tasks,
test-retest
correlations
ranged from .03-
.81 with 1/10 tasks
meeting relaxed
criterion. Across 5
cognitive domains
correlations
ranged from .61-
.90: 1/5
(Language) met
relaxed criterion,
1/5 (Executive
Functions) met
strict criterion.
Average absolute
score differences
were less than 1
point for 8/10
tasks – i.e.,
Based on a framework that
cognitive functions are
needed in daily life; 5
functions identified and
tasks designed to test
them.
Items based on a review of
the literature on cognitive
function and other tests.
Author’s expert opinion
determined percent each
task contributed to each
cognitive domain; author
reviewed standardization
sample to confirm this.
Spanish version developed
concurrently and included
only items that would be
familiar to both English and
Spanish speakers.
Confirmatory factor
analysis completed and
goodness-of-fit statistics
>.97 across items from 5
cognitive domains.
Clock drawing correlated
.74 with total composite.
Age and severity cutoffs
determined by author
based on examination of
normal subjects and
subjects with neurological
impairments.
Significant difference
between clinical and
normal subjects matched
by age group, education,
sex.
Author reviewed
patient charts to
confirm diagnosis.
Data presented by
etiology but
diagnostic utility
not directly tested.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Functions) met
strict criterion.
Average absolute
score differences
were less than 1
point for 8/10
tasks – i.e.,
statistics limited in
normal subjects
because of ceiling
effects and
relatively small
number of points
available for some
questions.
Absolute score
differences on
cognitive domains
ranged from 1.22-
16.57.
Did not test in
clinical samples.
familiar to both English and
Spanish speakers.
Stories designed to use
everyday vocabulary and
minimize cultural or
socioeconomic bias, and
reviewed by Spanish
speakers with modifications
based on pilot study
feedback.
Children’s Orientation
and Amnesia Test
(COAT; Ewing-
Cobbs, Levin,
Fletcher, Miner, &
Eisenberg, 1990)
During the sub-
acute stage of
recovery in 11
children with
TBI, compared
scores obtained
by 2 examiners:
met strict
criterion.
DNT
DNT
Developed by experts in
pediatric TBI.
Based on an adult test
(Galveston Orientation and
Amnesia Test) that has
established reliability and
validity.
No effect of sex on scores.
Test based on current
concepts of anterograde
and retrograde amnesia.
Scores increase with age
from 3 to 15 years.
Glasgow Coma
Scale (GCS)
scores significantly
higher in subjects
with post-traumatic
amnesia (PTA)
duration < 7 days
(Group 1, n=21)
than in subjects
with PTA duration
8-14 days (Group
2, n=7) or PTA
duration > 14 days
(Group 3, n=9).
Fewer focal
lesions in Group 1
(52%) than in
Group 2 (57%)
and Group 3
(66%).
PTA as
determined by the
COAT correlated -
.61 with admission
GCS score and
.48 with duration
of unconscious-
ness.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
GCS score and
.48 with duration
of unconscious-
ness.
PTA as
determined by the
COAT correlated
significantly with
verbal consistent
long-term retrieval
(CLTR) at
baseline.
Discourse
Comprehension Test
(DCT; Brookshire &
Nicholas, 1993)
DNT
Performance of
40 control
subjects was
“virtually
identical” across
2 sets of 5
stories. No
statistics
calculated.
Met relaxed
criterion for
aphasic subjects
and strict criterion
for subjects with
right hemisphere
damage.
Design based on test
authors’ belief that a test of
discourse comprehension
must assess
comprehension and
retention of both main
ideas and details, as well
as comprehension of both
directly stated and implied
info.
10 judges (graduate
students and speech-
language pathologists)
verified that questions
actually tested stated and
implied main ideas and
details.
Passage Dependency
Index (PDI) calculated for
each story based on the
performance of 10 adults
without brain damage. 9/10
PDI scores were at or
above the level “generally
considered acceptable”
(.40).
Analysis of word
frequency and word types
showed that DCT content
words are more like adult-
to-adult conversations
than newspapers.
Pearson product-
moment
correlation
coefficients
calculated for DCT
total scores vs.
auditory
comprehension
subtests of the
Boston Diagnostic
Examination of
Aphasia (r=.76),
the 4 subtest
version of the
Porch Inventory of
Communicative
Abilities (r=.64),
and the Sentence
Completion Test
(r=.53) for 20
aphasic adults.
Authors cite
Katsuki-
Nakamura,
Brookshire, &
Nicolas (1988) as
evidence that DCT
predicts
comprehension in
daily life: Purpose
was to compare
paragraph
comprehension for
stimuli read as
monologues vs.
dialogues. Study
stimuli were DCT
precursors.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
comprehension for
stimuli read as
monologues vs.
dialogues. Study
stimuli were DCT
precursors.
For aphasic
subjects, no
significant
correlation of total
errors between
stories read as
monologues and
stories read as
dialogues.
No significant
correlation of main
ideas recalled for
monologues vs.
dialogues,
significant
correlation of
details recalled in
monologues vs.
dialogues; both for
aphasic and
control subjects
combined.
Galveston Orientation
and Amnesia Test
(GOAT; Levin,
O'Donnell, &
Grossman, 1979)
Ratings by 2
examiners who
alternated
interviewing
patients
demonstrated a
“significant level”
of inter-observer
agreement (no
statistics
reported).
DNT
DNT
Developed by experts
based on current concepts
of memory and stages of
recovery after injury.
DNT
In 32 patients
“association was
highly significant
to duration of post-
traumatic
amnesia, as
measured by the
GOAT, and acute
neurologic
impairment,
measured by eye,
motor, and verbal
components of the
Glasgow Coma
Scale. ”
Serial GOAT
scores strongly
correlated with the
severity of acute
cerebral
disturbance after
TBI.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
severity of acute
cerebral
disturbance after
TBI.
LaTrobe
Communication
Questionnaire
(LCQ; Douglas,
O'Flaherty, & Snow,
2000)
DNT
Item analysis
results met
relaxed criterion.
Tested 24 primary
subjects and 24
close others (self-
selected
conversation
partners) over an
8-week period: Did
not meet criteria.
For close others,
authors cited
inconsistency of
subject’s choice of
close other as
reason for low
reliability.
Developed for use with
individuals with brain
damage and close others.
In primary subjects (i.e. vs.
close others), age and
education did not account
for significant percent of
variance in self-perceived
communication ability.
Females reported
significantly fewer
difficulties than males with
2 items: keeping track of
main details of
conversation, changing
speech style according to
situation.
No effect of age in adults
tested.
Completed using self-
report data as sample size
for close others was
insufficient.
Factor analysis based on
147 responses yielded 6
factors accounting for
49% of variance:
conversational tone,
conversational
effectiveness,
conversational flow,
conversational
engagement,
conversational (partner)
sensitivity, and
conversational
attention/focus.
Age and social
desirability poorly
correlated with
LCQ scores.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Mount Wilga High
Level Language Test
(Christie, Clark, &
Mortensen, 1986)
DNT
DNT
DNT
Developed by individuals
working with TBI clients
with mild language
problems.
Designed for assessment
of linguistic skills without
ceiling effects of standard
aphasia tests, and with
consideration of
communication efficacy
and appropriateness (i.e.
impairments that are
common after TBI).
Revisions based on a
normative study (Beverley
& Scott, 1986), but current
version not normed;
authors stated need for
collection of normative
data.
DNT
DNT
Repeatable Battery
for the Assessment of
Neuropsychological
Status
(RBANS; Randolph,
2001)
3 scorers on 20
random Figure
Copy/Recall
subtests from
standardization
sample. Met
relaxed criterion.
Booklet A vs. B:
repeated
measures
ANOVA of Index
scores showed
no significant
difference for any
age except
Language Raw
Score – Semantic
Fluency
increased 4
points on second
test and scores
for Form B
adjusted to
compensate.
Split half: Total
scale scores met
strict criterion for
all 6 age groups;
3/5 index scores
met relaxed
criterion for all
age groups; 2/5
index scores met
relaxed criterion
for 4/6 age
groups.
Total score met
relaxed criterion.
Index scores did
not meet criteria.
Designed to meet 3 test
goals: 1) alternative forms
developed for repeated
testing, 2) overall length
permits use as screening
and portable test, 3) range
of item difficulty so it can
be used as a core test for
dementia.
Tests cognitive areas
known to be affected in
dementia.
Existing tests are
excessively difficult for
older subjects – designed
to meet needs of that
population and be sensitive
enough to detect mild
impairment.
Tasks used based on a
review of the literature on
dementia.
Standardized on younger
subjects because it
became apparent that it
could also be used to
screen cognitive function in
younger patients e.g. acute
care, tracking recovery
after injury,
neuropsychological
screening for non-
neuropsychologists.
Calculated statistical
probability of differences
between index scores at
p>.15 and p>.05, and
base rates at which
differences occurred.
Intercorrelations among
subtests range from .28-
.44, and among index
scores range from .20-.40
– i.e., measuring different
constructs.
Preliminary comparison to
data from Wechsler Adult
Intelligence Scales
(WAIS) Revised,
Wechsler Memory Scales
Revised, Benton
Judgment of Line
Orientation, Complex
Figure Test, Wide Range
Achievement Test Third
Edition, Boston Naming
Test, Controlled Oral
Word Association – but
sample sizes small and
patient groups had mixed
etiologies: largest group
was for WAIS-Full Scale
IQ vs. RBANS and
correlation between these
was .78.
DNT
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
relaxed criterion
for 4/6 age
groups.
screen cognitive function in
younger patients e.g. acute
care, tracking recovery
after injury,
neuropsychological
screening for non-
neuropsychologists.
patient groups had mixed
etiologies: largest group
was for WAIS-Full Scale
IQ vs. RBANS and
correlation between these
was .78.
Rivermead
Behavioural Memory
Test (RBMT; Wilson,
Baddeley, Cockburn,
& Hiorns, 1985)
2 raters scoring
40 subjects met
strict criterion.
DNT
Tested by
comparing Version
A vs. B, C, D.
For Screening
scores:
A vs. B and A vs.
C met strict
criterion.
For Standardized
profile scores:
all met relaxed
criterion.
Test developed by experts
and based on the idea that
patients with brain have
substantially more memory
problems than controls.
Designed to represent
everyday memory
demands.
Overall level of
performance between
patients and controls
revealed that patients
have substantially lower
scores than the control
subjects.
Performance on RBMT
compared to performance
on standardized memory
tests: Warrington’s (1984)
Recognition Memory Test
for words and faces, Digit
Span, Spatial Span, and
the Paired-Associate
Learning subtest taken
from Randt, Brown and
Osborne (1980). Results
showed that the RBMT
had moderate but
significant correlations
with all of the tests and
can be regarded as
measuring overall
memory performance
when assessed in terms
of existing tests. Designed
to reflect everyday
memory, and performance
on the RBMT correlated
with observation made by
therapists on memory
lapses of 80 patients for a
mean of 35 hours of
observation per patient.
Correlation was -0.71 for
the screening score and -
0.75 for the standardized
profile score. Results
DNT
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
lapses of 80 patients for a
mean of 35 hours of
observation per patient.
Correlation was -0.71 for
the screening score and -
0.75 for the standardized
profile score. Results
suggest that the RBMT
offers a good measure of
everyday memory
performance
Ross Information
Processing Ability,
Second Edition
(RIPA-2; Ross-Swain,
1996)
3 clinicians
independently
scored a
videotaped test
administration.
Inter-scorer
agreement met
strict criterion.
Analyzed using
126 protocols.
4/10 subtests met
strict criterion,
5/10 met relaxed
criterion.
DNT
Test construction based on
theoretical framework of
cognitive-linguistic function.
Rationale for test format
and item/subtest selection
based on the literature and
expert opinion.
Panel of three
professionals assessed
face validity.
Item analysis “satisfied
requirements.”
Inter-subtest correlations
relatively high (r’s = .56-
.89).
Factor analysis reveals
that RIPA-2 is a one-
factor test, with subtest
factor loadings of .56-.89.
Does not support
reporting of individual
subtest scores.
Test stated purpose is to
“quantify and qualify
cognitive-linguistic
deficits” and permit
clinician to analyze “task-
by-task performance.” Not
supported by the test’s
single factor structure.
RIPA-2 subtests
compared to
selected subtests
of the Woodcock
Johnson Tests of
Cognitive Ability.
Coefficients range
from .02 – .78.
Scales of Cognitive
Ability for Traumatic
Brain Injury (SCATBI;
Adamovich &
Henderson, 1992)
DNT
Based on 164
tests.
Raw scores on
the five
subscales met
strict criterion.
33 subjects with
TBI retested within
8 months.
Statistically
significant
increase in mean
scores on all five
subscales. 1/5
subtests met strict
criterion, 4/5 did
not meet relaxed
criterion.
Not based on any single
model of cognitive function.
Instead, constructed to
have hierarchy of cognitive
processes that typically are
impaired after TBI. Based
on TBI literature and child
cognitive development
research on progression of
cognitive task difficulty.
Use of principles from child
development based on
premise that cognitive
development and cognitive
recovery progress in similar
hierarchy over time.
Significant but low
correlation between 5/5
subscale scores and age
in TBI subjects; 1/5
subtest scores in non-
injured subjects.
Factor analysis yielded 4-
factor solution that
supports subtest
structure.
Discriminant analysis on
164 TBI subjects vs. 78
demographically matched
non-injured subjects,
using subscale raw
scores: 2 subscales
accounted for 50% of
variance in group
membership; remaining
3/5 add nothing;
classification rate = 95.7
Correlations
between raw
scores on five
subscales and
Rancho Levels =
.50-.60, all
statistically
significant.
Significant
increase in 4/5
subscale scores
between Rancho
Levels of
Cognitive
Functioning (LCF)
V and VI, and VI
and VII. No
significant
differences
between LCF IV
vs. V, and VII vs.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
recovery progress in similar
hierarchy over time.
Task difficulty increases in
multiple domains – e.g.
stimulus complexity, time
requirement.
Target areas (e.g.
attention, memory) are
those that must improve
through treatment, per the
authors. Items based on
authors’ detailed study of
changes on 39
standardized and
nonstandardized
measures, in 14 patients
with TBI over six months of
treatment vs. 9 uninjured
controls.
Scales organized around
common cognitive and
perceptual functions with
similar items.
Included both auditory and
visual stimuli, with
“reasonable balance where
possible”.
using subscale raw
scores: 2 subscales
accounted for 50% of
variance in group
membership; remaining
3/5 add nothing;
classification rate = 95.7
accuracy for non-injured,
79.2 for TBI subjects
Functioning (LCF)
V and VI, and VI
and VII. No
significant
differences
between LCF IV
vs. V, and VII vs.
VIII.
Low but significant
correlation
between 4/5
subscale scores
and days post-
injury.
The Speed and
Capacity of Language
Processing Test
(SCOLP; Baddeley,
Emslie, & Nimmo-
Smith, 1992)
DNT
Internal
consistency not
measured.
In 25 normal
adults, correlation
between two
halves of the
Speed of
Comprehension
test met relaxed
criterion at test
and at re-test.
25 subjects re-
administered
Speed of
Comprehension
an average of 19.4
days on Version A
vs. Version B.
Met strict criterion.
50 subjects
completed Spot-
the-Word Version
A and Version B.
Correlation did not
meet relaxed
criterion.
224 normal adults
in standardization
sample:
correlation
between Speed of
Comprehension
Version A and
Version B met
relaxed criterion;
between Spot-the-
Word Version A
Based on the premise that
brain damage slows the
rate of information
processing.
Developed by experts in
cognitive psychology and
rehabilitation.
Based on an measure
originally designed to test
retrieval from semantic
memory.
Vocabulary is subtracted
from speeded processing,
to “differentiate between a
subject who has always
been slow and a subject
whose performance has
been impaired as a result
of brain damage or some
other stressor.”
In 76 normal subjects,
significant correlation
between Speed of
Comprehension Test and
other measures of
speeded verbal fluency:
category generation r=.52,
color naming r=.56; letter-
matching tasks r=.40
(name match), r=.34
(physical match);
vocabulary r=.51.
In 50 normal subjects,
Spot-the-Word Test
correlated with Mill Hill
Vocabulary test .60 for
Version A, .71 for Version
B.
In 224 adults in
standardization sample,
correlation between
Speed of Comprehension
and vocabulary (.60) and
grammatical reasoning
(.60).
In 76 normal
subjects,
correlation
between Speed of
Comprehension
and Raven’s
Standard
Progressive
Matrices r=20; age
r=.03.
Higher correlation
with Vocabulary
than measure of
general
intelligence, in
normal elderly
subjects (n not
stated).
Interpreted as
evidence that test
is not measuring
general
intelligence.
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
between Speed of
Comprehension
Version A and
Version B met
relaxed criterion;
between Spot-the-
Word Version A
and Version B met
relaxed criterion.
other stressor.”
Lexical decision (Spot-the-
Word) task designed to be
taken without the need to
read aloud (e.g.
pronounce) words, given in
a group, adapted to any
language, minimize
embarrassment (via silent
reading).
Significant effects of age,
social class, and sex on
Speed, and of age and
social class on Spot.
standardization sample,
correlation between
Speed of Comprehension
and vocabulary (.60) and
grammatical reasoning
(.60).
In 224 normal adults,
correlation between
Speed and Spot = .57 –
i.e., measuring a different
but related construct.
evidence that test
is not measuring
general
intelligence.
The Awareness of
Social Inference Test
(TASIT; McDonald,
Flanagan, Kinch, &
Rollins, 2002)
Subtest abbreviations
used:
ER = Emotion
recognition
SI-M = Social
Inference –Minimal
SI-E = Social
Inference – Enriched
DNT
Correlations not
calculated. ER:
Form A and Form
B have no
significant
difference in
normal subjects;
scores for
individual
emotions differed
between tests.
SI-M and SI-E:
Form A and Form
B have no
significant
difference in
normal subjects;
scores for
individual
inference items
differed between
tests.
Practice effects
studied by re-
administration in a
single setting to
subgroups of 127
normal subjects
from development
sample, and 6
months later to 6
adults from
community.
Results analyzed
with sign test;
correlations not
calculated.
Immediate re-test:
all parts except
Form B Emotions
improved, range of
point increase = 1-
6 (max possible
score = 28 for
emotions, 20 for
social inference
both subtests
combined); no
significant change
6 months later.
Test developed based on
literature on sequelae of
TBI and existing measures
of social cognition and
emotion.
Items developed with
feedback from three
consecutive cohorts of
normal viewers, with
revisions based on each.
7 viewers with TBI gave
feedback re: clarity of
instructions.
Significant correlation
between education (-.41)
and age (.53) on ER Form
B, no significant correlation
on Form A or with
intelligence and scores on
either Form A or B.
No significant correlation of
age or education on SI
subtests, but significant
correlation with intelligence
and scores on both SI-M
Form A (.37) and B (.49)
and SI-E Form A (.36) and
Form B (.38).
12 adults with severe TBI
M=4 weeks post
emergence from post-
traumatic amnesia, all
with normal performance
on language, facial
recognition, immediate
memory and information
processing speed;
compared to 12 age- and
education-matched
controls: significant
differences in total scores
for Emotion and SI-E, not
SI-M.
DNT
Test of Everyday
Attention for Children
(TEA-Ch; Manly,
Robertson, Anderson,
& Nimmo-Smith,
1999)
DNT
DNT
Correlation
coefficients
reported for 9
subtests based on
55 children. Test-
retest interval not
stated.
Based on: 1) published
research in adults
suggesting that attention is
a multi-process construct
(selective, sustained, and
spatial); 2) analysis of the
Test of Everyday Attention
Adult version suggesting at
least three different
attentional factors
(selective, sustained,
Factor analysis performed
with three factor model
determined a prior
(selective, sustained,
switching attention).
Comparative fit index for
this model = .97, Normed
Fit index = .91 and Non-
Normed Fit Index = .96
(all indicate good fit).
Regression coefficients
24 boys with DSM
IV-based
diagnosis of
attention deficit
hyperactivity
disorder had
significantly lower
scores than
vocabulary- and
age-matched
controls from
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
& Nimmo-Smith,
1999)
55 children. Test-
retest interval not
stated.
Correlation
coefficients met
relaxed criterion
for Sky Search
time-per-target,
Sky Search DT
decrement, Same
World time,
Opposite World
Time.
Percent
agreement
reported for 3
subtests with
ceiling effects.
Range = 71.0-
76.2%.
Reliable change
scores calculated
for practice effects
from Version A to
Version B on
children retested
between 5-20
days (n not stated)
and presented in
Table 13.
(selective, sustained, and
spatial); 2) analysis of the
Test of Everyday Attention
Adult version suggesting at
least three different
attentional factors
(selective, sustained,
switching), 3) research on
development of attention in
children; and 4) well-
validated measures of
attention in adults. Bias
addressed indirectly in
factor analysis (i.e. factor
analysis should be affected
by confounds)
switching attention).
Comparative fit index for
this model = .97, Normed
Fit index = .91 and Non-
Normed Fit Index = .96
(all indicate good fit).
Regression coefficients
for individual subtests on
each factor ranged from
.44-.79.
Creature Counting, Sky
Search time-per-target
and Same World time not
included in the factor
analysis (no reason
given).
96 children also received
Stroop Test, Trails,
Matching Familiar Figures
Test. Correlations are
modest and in directions
predicted by test model.
As predicted, no
significant correlation with
Wechsler Intelligence
Scales for Children Third
Edition (WISC-III).
Small but significant
correlations between
some subtests and
aspects of achievement.
hyperactivity
disorder had
significantly lower
scores than
vocabulary- and
age-matched
controls from
standardization
sample on all
subtests except
Sky Search. Same
with
standardization
sample children
matched on Block
Design Scores.
Significant
differences
between 18
children with TBI
and 18 age and
sex-matched
controls on all
subtests but Score
and Creature
Counting time,
with no significant
difference in
scores on WISC-III
Block Design or
Object Assembly
(groups differed
significantly on
Vocabulary and
Similarities
scores).
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Test of Language
competence
(TLC; Wiig & Secord,
1989)
Based on 16
raters for Level 1
and 8 raters for
Level 2, rating 2
protocols each.
Subtests 3 and 4
met strict
criterion for
Levels 1 and 2.
Level 1
composite score
met strict criterion
for all 8 age
groups except 8-
0 to 8-5 and 9-6
to 9-11, which
met relaxed
criterion. Level 2
composite score
reliabilities met
relaxed criterion
for oldest 3/8 age
groups and did
not meet criterion
for 5/8 age
groups.
For Level 1, based
on 88 students
ages 6 and 8
years with there
was 4-8 week test-
retest interval.
1/7 subtests and
composite score
met strict criterion;
5/7 subtests met
relaxed criterion.
For Level 2, based
on 92 students
ages 9-18 years
with 3 week test-
retest interval.
None met criteria.
Designed to measure a
single construct termed
“language competence”.
Subtests viewed as a
series of tasks designed to
discriminate between
competent and non-
competent language use.
Subtests may be
categorized according to a
model of language
competence. Model
focuses on the levels of
semantics, semantic-
syntactic interfaces, and
pragmatics. Within these
levels, the model identifies
the content of each of the
subtests as propositions in
narrow contexts or
propositions in
communication-like
contexts.
Most subtest
intercorrelations moderate
(.17-.57).
Factor analysis performed
on subtest correlation
matrix for each age
interval.
For Level 2, more than
86% of common variance
explained by the first
unrotated factor. For Level
1, all common variance
explained by the first
factor at each age. Should
be viewed as a one-factor
test. With factor analysis
of item intercorrelation
matrix, Level 1 range from
.36-.55 and Level 2 range
.27-.49.
Classification of Language
Learning Disability (LLD)
compared between TLC-E
vs. Test of Adolescent
Language (TOAL): LLD =
96% for TLC-E vs. 89%
for TOAL; non-LLD = 89%
for TLC-E vs. 86% for
TOAL. Discriminated
children with vs. without
TBI in subsequent study
(Towne & Entwisle, 1993).
Compared to a
priori classification
of students by
school districts as
LLD or normally
achieving.
Accuracy of TLC =
96% LLD, 93%
non-LLD.
Correlations
between the TLC
subtests and
Wechsler
Intelligence Scales
for Children
Revised (WISC-
R), Test of
Language
Development
Second Edition,
Peabody Picture
Vocabulary Test
Revised (PPVT-R)
as predicted by
content – i.e.
relatively high with
verbal measures
(e.g. .39-.55 with
PPVT) and low
with nonverbal
measures (e.g.,
.07-.12 with WISC-
R Performance
IQ).
Test
Inter-rater
Internal
consistency
Test-retest
reliability
Content/Face validity
Construct validity
Criterion-related
validity:
Concurrent
Western Aphasia
Battery
(WAB; Kertesz,
1982)
Based on 3
raters rating 10
videotaped
patients at two
time points 6
months apart.
9/9 subtests met
strict criterion.
Met strict
criterion.
7/9 subtests and
Aphasia Quotient
met strict criterion;
1/9 subtest and
Cognitive Quotient
met relaxed
criterion; Praxis
subtest did not
meet criterion.
Used clinical and
neurolinguistic principles,
concepts, and some of the
subtests developed by
Goodglass and Kaplan.
WAB “tests all language
modalities and is
comparable in content with
other batteries assessing
language impairment in
aphasia; praxis and
construction subtests,
while not included in all
standard aphasia batteries,
provide an assessment of
more general higher-
cortical functioning”.
Assesses language
content areas common to
all aphasia batteries.
Items within subtests
similar to those of the
BDAE.
15 subjects administered
WAB and NCCEA within
2-week interval.
Correlations ranged from
.82 to .92.
2 summary comparisons
computed for comparable
WAB and Neurosensory
Center Comprehensive
Examination for Aphasia
subtests, correlation =.97.
Normal control subjects
(n=31) differed
significantly from 70
individuals with non-
dominant-hemisphere
lesions, 17 subjects with
diffuse brain lesions, and
117 subjects with acute
aphasia except
anomic group.
Non-left hemisphere
groups did not differ
significantly among
themselves.
Across aphasia subtypes,
all groups differed
significantly except
Broca’s vs. Wernicke’s
aphasic types.
Scores related to
site of lesion – i.e.
subjects with right
hemisphere
lesions did not test
in aphasic range.
AQ correlation
with Raven’s
Standard
Progressive
Matrices scores =
.55 for 140
patients.
References
Adamovich, B., & Henderson, J. (1992). Scales of Cognitive Ability for Traumatic Brain Injury (Normed ed.): The Riverside Publishing Company.
Baddeley, A., Emslie, & Nimmo-Smith, I. (1992). Speed and Capacity of Language Processing Test (First ed.). Suffolk, England: Thames Valley
Test Company.
Brookshire, R., & Nicholas, L. (1993). Discourse Comprehension Test (First ed.): BRK Publishers.
Carrow-Woolfolk, E. (1999). Comprehensive Assessment of Spoken Language (First ed.). Circle Pines, MN: American Guidance Service, Inc.
Christie, J., Clark, W., & Mortensen, L. (1986). Mount Wilga High Level Language Test (First ed.). Sidney, NSW: Mt. Wilga Rehabilitation Centre:
Unpublished Manuscript.
Douglas, J., O'Flaherty, C., & Snow, P. (2000). Measuring perception of communication ability: The development and evaluation of the La Trobe
communication questionnaire. Aphasiology, 14, 251-268.
Ewing-Cobbs, L., Levin, H., Fletcher, J., Miner, M., & Eisenberg, H. (1990). The Children's Orientation and Amnesia Test: Relationship to severity
of acute head injury and to recovery of memory. Neurosurgery, 27, 683-691.
Frattali, C., Thompson, C., Holland, A., Wohl, C., & Ferketic, M. (1995). American Speech Language Hearing Association Functional Assessment of
Communication Skills for Adults (First ed.). Rockville, MD: American Speech Language Hearing Association.
Gioia, G., Isquith, P., Guy, S., & Kenworthy, L. (1996). Behavior Rating Inventory of Executive Function (First ed.). Odessa, FL: Psychological
Assessment Resources, Inc.
Helm-Estabrooks, N. (1992). Aphasia Diagnostic Profiles (First ed.): Applied Symbolix.
Helm-Estabrooks, N. (2001). Cognitive Linguistic Quick Test (First ed.). San Antonio, TX: Psychological Corporation.
Helm-Estabrooks, N., & Hotz, G. (1991). Brief Test of Head Injury (First ed.): Riverside Publishing Company.
Holland, A., Frattali, C., & Fromm, D. (1999). Communication Activities of Daily Living (Second ed.). Austin, TX: Pro-Ed, Inc.
Kertesz, A. (1982). Western Aphasia Battery (First ed.). San Antonio, TX: The Psychological Corporation.
Levin, H., O'Donnell, V., & Grossman, R. (1979). The Galveston Orientation Amnesia Test: A practical scale to assess cognition after head injury.
Journal of Nervous and Mental Disease, 167, 675-684.
Manly, T., Robertson, I., Anderson, V., & Nimmo-Smith, I. (1999). The Test of Everyday Attention for Children (First ed.). Suffolk, England: Thames
Valley Test Company.
McDonald, S., Flanagan, S., Kinch, J., & Rollins, J. (2002). The Awareness of Social Inference Test (First ed.). Suffolk, England: Thames Valley
Test Company.
Randolph, C. (2001). Repeatable Battery for the Assessment of Neuropsychological Status (First ed.). San Antonio, TX: Psychological Corporation.
Ross-Swain, D. (1996). Ross Information Processing Assessment (Second ed.). Austin, TX: Pro-Ed.
Semel, E., Wiig, E., & Secord, W. (1995). Clinical Evaluation of Language Fundamentals (Three ed.). San Antonio, TX: The Psychological
Corporation.
Towne, R. L., & Entwisle, L. M. (1993). Metaphoric comprehension in adolescents with traumatic brain injury and in adolescents with language
learning disability. Language, Speech and Hearing Services in Schools, 24, 100-107.
Turkstra, L. (1999). Language testing in adolescents with brain injury: A consideration of the CELF-3. Language, Speech, and Hearing Services in
Schools, 30, 132-140.
Wiig, E., & Secord, W. (1989). Test of Language Competence-Expanded Edition (Expanded ed.). San Antonio, TX: Psychological Corporation.
Wilson, B., Baddeley, A., Cockburn, J., & Hiorns, R. (1985). The Rivermead Behavioural Memory Test (Second Supplement ed.): Western
Psychological Services.