Content uploaded by Cathy Lussier
Author content
All content in this area was uploaded by Cathy Lussier on Apr 10, 2021
Content may be subject to copyright.
Reading
disabilities
in
children:
A
selective
meta-analysis
of
the
cognitive
literature
§
Milagros
F.
Kudo *,
Cathy
M.
Lussier,
H.
Lee
Swanson *
University
of
California,
Riverside,
United
States
1.
Introduction
A
popular
assumption
is
that
children
with
reading
disabilities
(RD)
have
specific
localized
low-order
processing
deficits.
A
component
consistently
implicated
in
reading
disabilities
is
phonological
awareness.
Phonological
awareness
is
‘‘the
ability
to
attend
explicitly
to
the
phonological
structure
of
spoken
words’’
(Scarborough,
1998,
p.
95).
Abundant
evidence
shows
that
children
with
RD
have
problems
in
processing
phonological
information
(e.g.,
Gottardo,
Collins,
Baciu,
&
Gebotys,
2008;
Melby-
Lerva
˚g,
Lyster,
&
Hulme,
2012;
Nelson,
Lindstrom,
&
Lindstrom,
2012;
Scarborough,
2009;
Stanovich
&
Siegel,
1994;
Vellutino,
Fletcher,
Snowling,
&
Scanlon,
2004;
Wagner
&
Torgesen,
1987;
Waterman
&
Lewandowski,
1993).
Recently,
some
studies
have
suggested
other
processes
may
be
involved
in
reading
acquisition
that
are
as
important
as
phonological
awareness
(e.g.,
Swanson,
Harris,
&
Graham,
2003;
Swanson
&
Jerman,
2007).
Although
several
studies
show
that
reading
deficiencies
are
related
to
phonological
awareness
(e.g.,
Badian,
2001,
2005;
Bus
&
van
Ijzendoorn,
1999;
Catts,
Gillispie,
Leonard,
Kail,
&
Miller,
2002;
Morris
et
al.,
1998;
Stanovich,
1988),
additional
studies
suggest
other
processes
such
as
those
related
to
rapid
naming
(e.g.,
Bonifacci
&
Snowling,
2008;
Compton,
2003;
Cronin,
2013;
Kirby,
Parrila,
&
Pfeiffer,
2003;
Schatschneider,
Fletcher,
Francis,
Carlson,
&
Foorman,
2004),
orthography
(e.g.,
Cunningham
&
Stanovich,
1990;
McBride-Chang,
Manis,
Seidenberg,
Research
in
Developmental
Disabilities
40
(2015)
51–62
A
R
T
I
C
L
E
I
N
F
O
Article
history:
Received
10
April
2014
Received
in
revised
form
29
December
2014
Accepted
16
January
2015
Available
online
26
February
2015
Keywords:
Reading
disabilities
Cognition
Meta-analysis
A
B
S
T
R
A
C
T
This
article
synthesizes
literature
that
compares
the
academic,
cognitive,
and
behavioral
performance
of
children
with
and
without
reading
disabilities
(RD).
Forty-eight
studies
met
the
criteria
for
the
meta-analysis,
yielding
735
effect
sizes
(ESs)
with
an
overall
weighted
ES
of
0.98.
Small
to
high
ESs
in
favor
of
children
without
RD
emerged
on
measures
of
cognition
(rapid
naming
[ES
=
0.89],
phonological
awareness
[ES
=
1.00],
verbal
working
memory
[ES
=
0.79],
short-term
memory
[ES
=
0.56],
visual–spatial
memory
[ES
=
0.48],
and
executive
processing
[ES
=
0.67]),
academic
achievement
(pseudoword
reading
[ES
=
1.85],
math
[ES
=
1.20],
vocabulary
[ES
=
0.83],
spelling
[ES
=
1.25],
and
writing
[ES
=
1.20]),
and
behavior
skills
(ES
=
0.80).
Hierarchical
linear
modeling
indicated
that
specific
cognitive
process
measures
(verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-
term
memory)
and
intelligence
measures
(general
and
verbal
intelligence)
significantly
moderated
overall
group
effect
size
differences.
Overall,
the
results
supported
the
assumption
that
cognitive
deficits
in
children
with
RD
are
persistent.
ß
2015
Published
by
Elsevier
Ltd.
§
This
research
was
supported
by
an
Institute
of
Education
Science
(IES),
Cognition
and
Student
Learning,
grant
numbers
R324B080002
and
R324A090002
awarded
to
H.
Lee
Swanson.
*
Corresponding
author
at:
University
of
California,
Riverside,
900
University
Avenue,
Riverside,
CA
9252,
United
States.
Tel.:
+19518274734.
E-mail
addresses:
mkudo001@ucr.edu
(M.F.
Kudo),
lee.swanson@ucr.edu
(H.L.
Swanson).
Contents
lists
available
at
ScienceDirect
Research
in
Developmental
Disabilities
http://dx.doi.org/10.1016/j.ridd.2015.01.002
0891-4222/ß
2015
Published
by
Elsevier
Ltd.
Custodio,
&
Doi,
1993;
Shany
&
Share,
2011;
Spencer
&
Hanley,
2004;
Zaretsky,
Kraljevic,
Core,
&
Lencek,
2009),
semantics
(e.g.,
Crisp
&
Lambon
Ralph,
2006;
Nation
&
Snowling,
1998),
and
memory
span
(e.g.,
Das
&
Mishra,
1991;
Nevo
&
Breznitz,
2013;
Parrila,
Kirby,
&
McQuarrie,
2004)
contribute
statistically
significant
amounts
of
variance
to
reading.
The
current
literature
weighs
heavily
on
the
side
of
phonological
deficits
as
the
major
sources
of
reading
difficulties.
Nevertheless,
an
understanding
of
the
interplay
between
multiple
processes
is
necessary
before
one
has
an
adequate
account
on
the
major
information
processing
variables
that
contribute
to
RD.
More
important,
little
is
known
about
potential
moderator
variables
(e.g.,
age
of
the
sample,
severity
of
reading
problems)
that
influence
the
magnitude
of
the
effect
sizes
between
children
with
and
without
RD
on
cognitive
measures.
In
the
present
study,
we
sought
to
investigate
the
evidence
on
cognitive
differences
between
children
with
and
without
RD.
Thus,
our
interest
in
reading
ability
was
narrowly
confined
to
word
reading
and
those
variables
(e.g.,
phonological
awareness,
rapid
naming
speed)
that
have
been
identified
in
the
literature
as
critically
related
to
reading
disabilities
(see
Siegel,
2003,
for
a
comprehensive
review).
We
are
also
interested
in
investigating
potential
cognitive
processes
(e.g.,
spelling,
orthography,
vocabulary,
memory)
that
may
also
play
an
important
role
in
predicting
reading
disabilities.
The
study
used
meta-analytic
procedures
to
aggregate
the
research
literature.
The
three
primary
purposes
of
this
synthesis
were
to
(a)
conduct
a
meta-analysis
of
differences
between
children
with
and
without
RD,
(b)
identify
some
of
the
variables
that
moderate
effect
sizes
between
children
with
and
without
RD
(e.g.,
age
groups,
socioeconomic
status
[SES],
and
types
of
criterion
reading
measures
used
to
classify
skilled
and
readers
at-risk),
and
(c)
to
see
if
interactions
between
variables
moderated
the
effect
sizes
between
the
two
groups
of
children.
Two
main
research
questions
directed
this
synthesis:
1.
Which
performance
domains
(i.e.,
intellectual,
academic,
cognitive)
make
the
largest
contribution
to
the
differences
between
children
with
RD
and
their
average-achieving
counterparts?
In
other
words,
which
array
of
measures
show
the
largest
magnitude
of
effect
sizes
that
explain
the
similarities
and
differences
between
the
two
groups?
2.
What
performance
similarities
or
differences
among
children
with
and
without
RD
are
a
function
of
variations
in
age,
intelligence
quotient
(IQ),
ethnicity,
and
gender?
For
example,
we
determine
if
some
of
the
same
deficits
(as
reflected
in
the
magnitude
of
effect
size)
that
emerge
in
studies
that
include
older
participants
with
RD
in
secondary
school
also
occur
when
the
sample
is
early
elementary
school
age.
To
answer
these
questions,
the
present
synthesis
uses
hierarchical
linear
modeling
(HLM)
procedure
to
identify
key
constructs
(e.g.,
IQ,
reading,
math,
memory,
and
phonological
processing)
that
contribute
unique
(independent)
variance
to
defining
differences
and
similarities
between
children
with
and
without
RD.
2.
Method
2.1.
Identification
of
studies
(literature
search)
2.1.1.
Data
gathering
Several
approaches
were
used
to
locate
the
relevant
studies
published
in
peer-reviewed
journals.
First,
a
computer
search
located
studies
comparing
children
with
reading
disabilities
and
without
reading
disabilities
on
psychological,
occupational,
and
vocational
variables
using
the
Web
of
Knowledge,
PsycINFO,
and
ERIC
databases.
The
computer
search
used
the
following
terms:
‘‘cognitive
processes,
cognition,
memory,
speed,
phonological’’
coupled
with
‘‘learning
disabilities,
dyslexia,
reading
disorders,
orthographic,
and
specific
reading
disabilities.’’
Entry
of
these
terms
yielded
11,432
references.
Additional
terms
were
entered
into
the
search
such
as
‘‘IQ’’
and
‘‘assessment,’’
but
these
results
were
found
to
produce
results
that
overlapped
with
the
earlier
terms.
A
refinement
of
the
search
focused
only
on
empirical
studies
and
journal
articles
published
in
English.
The
sample
search
obtained
articles
using
the
above
descriptors
that
ranged
in
publication
dates
from
1957
(the
earliest
year
of
the
earliest
article
found
using
the
descriptors)
and
March
2013.
Second,
published
articles
by
primary
researchers
(i.e.,
Badian,
Berninger,
Bowers,
Bull,
Chiappe,
Das,
De
Jong,
Fletcher,
Johnson,
Naglieri,
Pennington,
Siegel,
Stanovich,
Swanson,
Vellutino,
Willcutt,
and
Wolf)
were
also
analyzed
for
possible
inclusion.
Finally,
a
manual
search
of
journals
where
the
majority
of
reading
disability
articles
is
published
was
conducted
(e.g.,
Journal
of
Educational
Psychology,
Journal
of
Learning
Disabilities,
and
Learning
Disability
Quarterly).
From
this
pool
of
literature,
articles
were
eliminated
that
focused
on
children
with
below
average
intelligence
(mild
mental
retardation
range,
<85)
and/or
were
not
comparative
or
data-based
studies.
Focusing
on
comparative
studies
(children
with
RD
vs.
children
without
RD)
that
were
published
in
English
journals
narrowed
the
search
down
to
588
studies.
The
588
potential
studies
were
further
evaluated
to
determine
their
relevance
to
the
current
review.
To
be
included
in
the
meta-analysis,
each
study
had
to
satisfy
the
following
criteria:
1.
Children
with
RD
(a
combination
of
various
labels
were
used
in
articles,
e.g.,
reading
disabilities,
dyslexia,
students
with
learning
disabilities
in
reading,
Individualized
Educational
Plan’s
[IEP’s)
which
indicated
a
focus
on
reading
difficulties,
etc.)
were
compared
to
chronologically
age
matched
children
without
RD
group
(i.e.,
no
indication
of
a
learning
or
behavior
deficit)
on
at
least
one
cognitive
measure
(e.g.,
phonological
awareness,
naming
speed,
memory,
executive
processing,
etc.).
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
52
2.
The
study
should
include
school-aged
children;
that
is,
children
that
are
6
years,
0
months
to
15
years,
11
months
old
(or
72–191
months).
3.
The
RD
subgroup
had
no
reported
co-morbidity
(e.g.,
math
disabilities,
attention-deficit/hyperactivity
disorder,
traumatic
brain
injury,
etc.).
4.
Each
study
reported
a
mean
score
and
standard
deviation
on
a
standardized,
norm-referenced
measure
of
intelligence
for
each
comparison
group
(e.g.,
Wechsler
Intelligence
tests
or
selected
subtests).
5.
Each
study
reported
a
mean
score
and
standard
deviation
from
a
standardized,
norm-referenced
measure
of
reading
for
each
comparison
group.
6.
Each
study
was
required
to
include
participant
samples
with
IQ
scores
above
80
in
the
analysis
(and
no
indication
that
the
majority
of
participants
fell
below
80).
Studies
were
excluded
if:
(a)
they
were
not
published
in
refereed
journals,
(b)
they
failed
to
provide
enough
quantitative
data
to
calculate
the
ESs,
(c)
they
failed
to
include
a
chronologically
age-matched
average
achieving
comparison
group,
(d)
they
failed
to
provide
information
on
each
ability
group
performance
on
a
standardized
(norm
referenced)
reading
and/or
IQ
test,
(d)
participants
were
reading
languages
other
than
English
(controlling
for
different
types
of
language
formats),
(e)
and/
or
the
article
did
not
provide
enough
description
to
assure
of
methodologically
sound
research
practices
being
utilized
(Slavin,
1986).
From
the
588
potential
studies,
only
48
met
the
aforementioned
inclusion
criteria.
The
majority
of
studies
were
excluded
because
they
failed
to
provide
enough
quantitative
data
to
calculate
the
ESs
or
they
did
not
aggregate
their
data
by
RD
and
non-RD
groups.
2.2.
Coding
procedure
Each
study
was
coded
for
the
following
information:
(a)
sample
characteristics,
(b)
classification
measures,
and
(c)
performance
measures.
2.2.1.
Attributes
of
the
study
Each
study
provided:
(a)
the
year
of
the
study,
(b)
the
name
of
the
first
author,
(c)
the
number
of
co-authors,
and
(d)
the
country
where
the
study
was
carried
out.
2.2.2.
Attributes
of
the
participants
According
to
the
inclusion
criteria,
each
study
provided
at
least
one
RD
and
one
non-RD
(NRD)
comparison
group.
Other
attributes
of
the
participants
coded
included:
(a)
the
number
of
participants
in
each
subgroup,
(b)
the
number
of
males
in
each
subgroup,
(c)
the
mean
age
of
the
group
(converted
into
months),
and
(d)
the
participants’
primary
language.
Studies
were
also
coded
for
(e)
SES
and
(f)
ethnicity.
An
estimated
age
was
calculated
for
studies
that
only
listed
the
grade
level
with
no
specific
reported
age
in
months
or
years.
Further,
some
studies
subdivided
the
RD
and
NRD
children
(i.e.,
high
span
and
low
span
memory).
In
this
scenario,
the
sub-groups
were
combined
across
their
means
and
standard
deviations
to
yield
single
RD
and
NRD
groups.
2.2.3.
Comparison
measures
All
classification
measures
were
converted
to
standard
scores.
In
those
cases
in
which
only
a
range
was
reported,
a
middle
value
was
assigned.
Classification
measures
included
measures
of
general
intelligence
(performance
and
nonverbal),
word
recognition,
and
reading
comprehension.
Comparative
measures
(those
not
included
as
part
of
the
classification
criteria),
were
organized
into
several
categories:
rapid
naming,
phonological
awareness,
pseudoword
reading,
math,
vocabulary,
spelling,
writing,
problem
solving
and
reasoning,
memory
and
cognitive
monitoring,
perceptual
motor
skills,
visual
perception
skills,
auditory
perception
skills,
general
information-facts,
and
behavior.
Because
of
the
small
number
of
effect
sizes
(ESs),
some
of
the
above
categories
were
aggregated
into
broader
domains.
For
example,
measures
of
rapid
naming
of
objects,
letters,
and
numbers
were
included
under
the
category
of
naming
speed.
The
visual–spatial
category
included
measures
of
both
visual-motor
and
non-visual-motor
tasks.
Although
comparative
tasks
not
used
in
the
classification
criteria
were
used
in
calculating
effect
sizes,
some
of
the
categories
for
analysis
were
no
doubt
related
(e.g.,
word
attack)
to
the
classification
variable
(e.g.,
word
recognition).
Thus,
children
with
RD
were
compared
to
their
counterparts
on
measures
related
to
the
following
categories:
2.3.
Classification
measures
1.
Reading
comprehension.
This
category
focused
on
measures
of
text
or
passage
comprehension.
The
majority
of
dependent
measures
in
this
domain
included
reading
comprehension
and
general
reading
measures
(e.g.,
the
Reading
Cluster
subtest
in
the
Woodcock
Johnson
Psychoeducational
Battery
[WJPB],
the
Passage
Comprehension
test
in
the
Woodcock
Reading
Mastery
Test-Revised
[WRMT-R],
and
the
Nelson
Denny
Reading
Test).
2.
Word
recognition.
This
category
focused
on
the
sight
recognition
of
real
words.
Sample
tasks
included
measures
of
irregular
and
regular
words,
experimental
words,
and
real
word
identification
(e.g.,
the
Word
Identification
subtest
in
the
Woodcock
Reading
Mastery
Tests
[WRMT-R]
and
the
Reading
subtest
in
the
Wide
Range
Achievement
Test
[WRAT]).
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
53
3.
General
intelligence
(IQ).
This
category
focused
on
standardized
measures
taken
from
tests
of
general
intelligence.
Sample
tasks
included
Raven’s
Progressive
Matrices
test
and
full
scale
IQ
(subtests
of
vocabulary,
and
verbal
IQ
were
assigned
to
the
categories
below).
4.
Verbal
intelligence
(Verbal
IQ).
This
domain
included
general
measures
of
verbal
intelligence
(e.g.,
the
Verbal
IQ
subtest
in
the
Wechsler
Intelligence
Scale
for
Children-III
[WISC-III]).
2.4.
Related
achievement
measures
5.
Math.
This
category
focused
on
measures
related
to
calculation.
Sample
tasks
included
measures
of
arithmetical
calculation
and
word
problem
solving
(e.g.,
WRAT,
Peabody
Individual
Achievement
Test,
and
WJPB-Math
Cluster).
6.
Vocabulary.
This
category
focused
on
measures
related
to
word
meaning.
Sample
tasks
included
measures
of
word
knowledge,
semantic
processing
measures
(e.g.,
Peabody
Picture
Vocabulary
Test,
Wechsler
Intelligence
Scale
for
Children-Vocabulary
subtest),
and
the
Stanford-Binet
Vocabulary
Test.
7.
Spelling.
This
category
focused
on
real-word
spelling
skills.
Sample
tasks
included
spelling
subtests
taken
from
standardized
tests
(e.g.,
WRAT,
Iowa
Test
of
Basic
Skills).
8.
Writing.
This
category
focused
on
written
language
(e.g.,
WJPB
written
language
cluster)
and
included
measures
of
syntax
and
grammar
(e.g.,
Test
of
Written
Language).
2.5.
Comparative
measures
9.
Rapid
naming
(RAN).
This
category
focused
on
measures
of
speed
of
processing
(timed
trials)
related
to
the
overt
verbalizing
of
letters,
sounds,
words,
objects,
or
colors.
Sample
tasks
included
color
naming,
digit
naming,
picture
naming,
number
naming,
letter
naming,
object
naming,
and
naming
of
words.
10.
Phonological
awareness.
This
category
focused
on
oral
tasks
that
required
dividing
spoken
words
into
segments
of
sounds
smaller
than
a
syllable
or
learning
about
individual
phonemes
(Torgesen
&
Mathes,
2000).
Sample
tasks
included
spoonerisms,
blending
sounds,
naming
letter
sounds,
phoneme
deletion,
phoneme
elision,
phoneme
segmentation,
phonemic
blending,
phonological
awareness,
phonological
oddity,
phonological
skills,
rhyme,
rhyme
judgment,
rhyming
letter
naming,
word
analysis,
sound
categorization,
syllable
deletion,
and
phoneme
detection.
11.
Pseudoword
reading
(word
attack).
This
category
focused
on
measures
of
word
attack
skills
and
was
considered
as
a
separate
entity
of
phonological
processing.
These
tasks
required
the
reading
of
printed
nonwords.
This
category
fits
most
closely
with
the
definition
provided
by
Siegel
(1993)
that
states
phonological
processing
is
‘‘the
understanding
of
the
grapheme-phonological
conversion
rules
and
the
exceptions
to
these
rules’’
(p.38).
Sample
tasks
included
the
reading
of
nonwords
(pseudowords)
or
sounding
out
of
nonwords
of
increasing
complexity
from
the
WRMT-R.
12.
Problem
solving/reasoning.
This
category
focused
on
general
problem
solving
on
measures
assumed
to
gauge
fluid
intelligence
(e.g.,
critical
thinking,
block
design,
picture
arrangement,
and
WJPB-Cognitive
Cluster).
13.
Memory
and
cognitive
monitoring.
This
category
focused
on
span
measures
related
to
digits,
words,
sentences,
and
objects.
Some
tasks
were
considered
measures
of
cognitive
monitoring,
such
as
the
Tower
of
Hanoi
and
Trail
Making.
Phonemic
memory
(recall
of
isolated
sounds)
was
coded
as
measures
of
phonological
awareness
and,
therefore,
was
not
included
in
this
category.
This
category
was
subdivided
into
verbal
working
memory,
visual–spatial
working
memory,
executive
processing,
and
verbal
short-term
memory.
14.
Perception
and
motor
skills.
This
category
focused
on
measures
of
tactical
performance
and
balance.
15.
Auditory
processing.
This
category
focused
on
auditory-perceptual
motor
or
listening
tasks.
16.
General
information.
This
domain
included
measures
that
tapped
previous
knowledge
or
memory
for
general
information
(e.g.,
WISC-III-Information
subtest,
WJPB-Knowledge
Cluster,
and
answered
questions
such
as
‘‘What’s
the
capital
of
California?’’).
17.
Behavior.
This
category
related
to
measures
of
attention
and
hyperactivity.
2.6.
Calculation
of
effect
sizes
For
each
measure,
an
effect
size
(ES)
was
computed,
a
Cohen’s
d
(Cohen,
1988),
and
then
weighted
by
the
reciprocal
of
the
sampling
variance
with
Hedge’s
g
(Hedges
&
Olkin,
1985).
The
dependent
measure
for
the
estimate
of
ES
was
defined
as
est
=
d/(1/
v
),
where
d
(Mean
of
RD
Mean
of
comparison
group/average
of
standard
deviation
for
both
groups),
and
v
is
the
inverse
of
the
sampling
variance,
v
=
(N
RD
+
N
nRD
)/(N
RD
N
nRD
)
+
d
2
/[2(N
RD
+
N
nRD
)]
(Hedges
&
Olkin,
1985).
Means
and
standard
deviations
were
used
in
the
computation
of
100%
of
the
ESs.
Thus,
ESs
were
computed
with
each
effect
size
weighted
by
the
reciprocal
of
its
variance,
a
procedure
that
gives
more
weight
to
effect
sizes
that
are
more
reliably
estimated.
The
overall
results
for
children
with
and
without
RD
are
shown
in
Table
1.
As
suggested
by
Hedges
and
Olkin
(1985),
outliers
were
removed
from
the
analysis
of
main
effects.
Outliers
were
defined
as
ESs
lying
beyond
the
first
gap
of
at
least
one
standard
deviation
between
adjacent
ES
values
in
a
positive
direction
(Bollen,
1989).
Cohen’s
criterion
was
used
for
the
interpretation
of
the
magnitude
of
the
ESs
(Cohen,
1988).
According
to
Cohen’s
criterion,
an
effect
size
(in
their
absolute
values)
of
0.20
is
small,
0.50–0.80
is
moderate,
and
0.80
or
greater
is
large.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
54
We
also
determined
whether
a
set
of
ds
shared
a
common
ES
by
category
(i.e.,
whether
they
were
consistent
across
studies).
The
analysis
of
each
category
of
measure
reported
separately
is
shown
in
Table
1.
Shown
are
the
mean
weighted
ES,
standard
error
(SE),
and
95%
confidence
interval
at
the
lower
and
upper
bounds.
For
the
category
of
each
dependent
measure,
a
homogeneity
statistic
Q
was
computed
to
determine
whether
separate
ESs
within
each
category
shared
a
common
ES
(Hedges
&
Olkin,
1985).
The
statistic
Q
has
a
distribution
similar
to
the
distribution
of
chi-square
with
k1
degrees
of
freedom,
where
k
is
the
number
of
ESs.
A
significant
chi-square
indicated
that
the
study
features
significantly
moderated
the
magnitude
of
ESs.
If
the
homogeneity
was
not
achieved,
which
is
usually
the
case,
then
the
influence
of
outliers
was
assessed
using
a
95%
confidence
interval.
The
Q
statistic
however,
has
been
criticized
in
the
literature
because
it
is
directly
tied
to
the
number
of
studies,
resulting
in
poor
power
to
detect
differences
when
a
small
number
of
studies
were
analyzed
and
excessive
power
to
detect
negligible
differences
when
a
large
number
of
studies
are
available.
Critics
of
the
Q
statistic
recommend
reporting
the
I
2
statistic,
which
provides
a
percentage
index
of
the
amount
of
variability
and
is
calculated
using
the
following
formula:
I
2
¼Q
ðk
1Þ
Q
I
2
indices
of
25%,
50%,
and
75%
are
classified
as
low,
medium,
and
high
heterogeneity,
respectively
(Higgins
&
Thompson,
2002).
As
shown
in
Table
1,
there
is
an
extremely
high
percentage
of
variability
across
the
majority
of
measures.
Because
we
expected
the
absence
of
homogeneity,
moderating
variables
were
determined.
For
example,
the
subsequent
analyses
determined
how
the
characteristics
of
the
sample
(e.g.,
reading
level,
chronological
age)
and
category
(e.g.,
phonological
awareness,
naming
speed)
of
the
various
studies
contributed
to
the
variability
and
the
heterogeneity
of
ESs.
To
determine
the
relationship
between
moderating
variables
and
the
magnitude
of
ESs,
a
conditional
model
was
analyzed.
Categorical
models,
analogous
to
an
Analysis
of
Variance,
showed
whether
the
heterogeneity
in
effect
sizes
were
isolated
to
a
particular
variable
(e.g.,
severity
of
reading
performance).
The
procedure
for
calculating
categorical
models
provides
a
between
class
effect.
This
procedure
was
considered
helpful
in
determining
if
certain
moderator
variables
made
a
significant
contribution
to
ES.
2.7.
Statistical
analysis
The
data
reflected
ES
nested
within
and
between
studies
(category
of
the
dependent
measure).
Thus,
hierarchical
linear
modeling
(HLM;
Raudenbush
&
Bryk,
2002;
Singer,
2002)
was
used
to
analyze
ES
nested
between
studies.
To
examine
ES,
we
used
a
random
effects
model
(Singer,
2002);
the
unconditional
model
is
expressed
as:
y
i
j
¼
b
o1
þ
U
o1
j
þ
R
i
j
where
y
ij
was
the
dependent
variable
(i.e.,
effect
size),
b
o1
was
the
grand
mean.
U
o1j
was
the
random
intercept
for
study
j
in
the
sample
representing
variation
between
studies,
U
o2j
was
the
second
random
intercept
for
study
x
domain
in
the
sample
representing
variation
between
studies,
and
R
ij
was
the
variation
of
effect
sizes
within
studies.
A
simple
conditional
model
can
be
expressed
as:
y
i
j
¼
b
o
þ
b
o1
ðdomainÞ
þ
U
o1
j
þ
U
o2
j
þ
R
i
j
where
y
ij
was
the
dependent
variable
(e.g.,
effect
size),
b
o
was
the
grand
mean,
and
b
o1
was
the
binary
variable
related
to
domain
comparison
(e.g.,
phonological
awareness
domain
vs.
other
domains).
The
domain
variables
were
entered
as
binary
variables
(e.g.,
phonological
awareness
+
1,
other
domains
0).
The
same
random
effect
and
the
residual,
as
included
in
the
Table
1
Summary
of
demographic
information
on
norm
standardized
tests.
Variable
Reading
disabled
Average
achieving
Effect
size
K
M
SD
M
SD
M
SD
Age
(months)
204
125.32
23.13
122.61
23.50
Gender
ratio
110
0.59
0.15
0.51
0.15
Normed
standard
scores
Reading
comprehension
28
83.54
6.44
107.55
7.03
1.90
0.98
Word
recognition
31
82.52
5.81
105.71
6.66
2.33
0.72
General
intelligence
45
100.93
7.26
107.62
6.08
0.54
0.58
Verbal
intelligence
21
97.31
6.27
108.44
6.98
0.93
0.58
Rapid
naming
10
95.69
11.25
96.67
26.06
1.17
0.74
Phonological
awareness
11
94.02
7.19
104.24
8.18
0.97
0.55
Pseudoword
reading
12
83.05
3.57
105.56
5.81
2.11
0.58
Math
8
95.38
5.90
105.33
4.35
0.85
0.45
Vocabulary
17
95.73
11.93
104.39
12.59
0.68
0.45
Spelling
7
86.40
9.01
110.87
15.83
0.68
0.45
Note:
K
=
number
of
effect
sizes;
M
=
mean;
SD
=
standard
deviation.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
55
unconditional
model,
were
also
included
in
the
conditional
model.
The
fixed
and
random
effect
parameter
estimates
were
obtained
using
PROC
MIXED
in
SAS
9.3
(SAS
Institute,
Inc.,
2010).
We
tested
three
conditional
models:
(1)
a
model
that
included
covariates,
age,
and
reading
level
of
children;
(2)
a
cognitive
model
that
included
memory
variables
(verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory);
and
(3)
a
full
model
that
included
cognitive
and
other
domains
that
may
have
contributed
significantly
to
the
magnitude
of
ESs.
Specifically,
we
tested
whether
adding
one
or
more
predictors
to
the
model
reduced
the
magnitude
of
the
various
random
components
related
to
study
effects.
The
random
effects
of
the
unconditional
model
represented
the
proportion
of
variance
in
the
effects
that
were
parameter-specific
rather
than
error
variance.
To
evaluate
the
compatibility
of
the
data
with
our
conditional
model,
we
tested
the
significance
of
the
model
change.
This
was
done
by
using
a
chi-square
difference
test
from
the
unconditional
and
conditional
models
by
subtracting
their
deviance
values.
The
degrees
of
freedom
for
the
Dx
2
equals
the
number
of
independent
constraints
imposed.
A
significant
chi-square
change
would
indicate
that
the
conditional
model
showed
a
better
fit
to
the
data
than
the
unconditional
model.
In
general,
models
with
lower
deviance
fit
better
than
ones
with
higher
deviance.
Snijders
and
Bosker
(2003)
argued
that
the
power
to
detect
significant
parameters
in
multilevel
research
is
frequently
low
because
of
reductions
in
parameter
reliability.
For
this
reason,
we
maintained
all
multiple
comparisons
at
p
<
0.05.
We
tested
the
models
using
both
restricted
maximum-likelihood
(REML)
and
maximum-likelihood
(ML)
estimations
to
compute
the
parameters
in
the
various
models.
However,
because
we
compared
variations
in
both
the
fixed
effects
and
random
effects,
the
results
of
the
ML
estimation
were
shown
in
Table
2.
Prior
to
the
analysis,
we
computed
the
intraclass
correlation
for
ESs.
The
intraclass
correlation
exceeded
0.10
indicating
that
ESs
within
studies
were
more
likely
to
have
outcomes
similar
to
ESs
within
the
various
categories
than
to
ESs
in
other
studies
(i.e.,
ESs
were
not
independent
of
one
another).
Thus,
it
was
necessary
to
portion
the
total
outcome
variance
into
between-study
variance
and
within-study
variance
(i.e.,
residual
error).
2.8.
Interrater
agreement
Two
post-doctoral
fellows
and
a
doctoral
student
coded
all
the
studies.
The
overall
structure
of
the
coding
system
yielded
a
reliable
percentage
of
interrater
agreement
across
all
codes
(>88%
agreement
initially
and,
eventually,
100%
after
clarification).
3.
Results
3.1.
Study
characteristics
The
meta-analysis
results
generated
48
studies
with
a
total
of
735
ESs
across
categories
between
children
with
RD
and
average
readers,
yielding
a
mean
ES
of
0.98.
These
studies
were
most
frequently
published
in
Journal
of
Educational
Table
2
Weighted
effect
sizes,
standard
errors,
confidence
intervals,
and
homogeneity
of
Q.
Comparisons
between
reading
disabled
and
average
readers
K
ES
SE
Lower
Upper
Homogeneity
Q
I
2
Total
across
categories
735
0.98
0.05
0.88
1.09
205.31
***
Reading
comprehension
52
1.63
0.04
1.56
1.71
553.83
***
0.91
Reading
recognition
52
2.20
0.04
2.12
2.28
251.07
***
0.80
General
IQ
72
0.54
0.02
0.49
0.59
399.60
***
0.82
Verbal
IQ
33
0.87
0.04
0.80
0.94
120.88
***
0.74
Rapid
naming
55
0.89
0.03
0.84
0.95
181.27
***
0.70
Phonological
awareness
66
1.00
0.04
0.93
1.07
296.29
***
0.78
Pseudoword
reading
19
1.85
0.07
1.71
1.98
83.67
*
0.78
Math
28
1.20
0.05
1.10
1.29
157.17
***
0.83
vocabulary
39
0.83
0.04
0.76
0.91
229.17
***
0.83
Spelling
17
1.25
0.06
1.13
1.37
328.00
***
0.95
writing
6
1.20
0.11
0.99
1.41
22.30
**
0.78
Problem
solving/reasoning
7
0.43
0.1
0.24
0.62
56.16
**
0.89
Verbal
working
memory
68
0.79
0.03
0.72
0.85
230.67
***
0.71
Visual–spatial
memory
35
0.48
0.04
0.40
0.57
111.54
***
0.70
Executive
processing
46
0.67
0.04
0.60
0.75
146.78
***
0.69
Short-term
memory
98
0.56
0.03
0.51
0.61
533.02
***
0.82
Perceptual
motor
skills
5
0.25
0.09
0.08
0.42
28.53
***
0.86
Auditory
processing
8
0.63
0.07
0.50
0.76
27.69
**
0.75
General
information
9
0.69
0.08
0.53
0.86
184.99
***
0.96
Behavior
19
0.80
0.05
0.70
0.90
115.91
***
0.84
Note:
K
=
number
of
effect
sizes.
*
p
<
0.05.
**
p
<
0.01.
***
p
<
0.001.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
56
Psychology,
Journal
of
Experimental
Child
Psychology,
Reading
and
Writing,
Journal
of
Learning
Disabilities,
and
Learning
Disability
Quarterly.
Publication
dates
for
the
included
studies
ranged
from
1988
to
2012,
with
the
average
year
of
publication
in
2003.
The
number
of
authors
ranged
from
1
to
9,
with
30
of
the
included
studies
conducted
in
the
United
States,
9
in
Canada,
6
in
England,
2
in
Australia,
and
1
in
both
the
US
and
Canada.
The
average
age
of
students
across
studies
was
10
years
old
(M
=
130.03
months),
and
the
majority
of
students
were
male.
There
were
425
effect
sizes
that
were
associated
with
a
male
to
female
gender
ratio
of
participants,
and
13
studies
reported
ethnicity
within
each
sample
of
RD.
Initial
testing
of
the
gender
and
ethnicity
ratio
resulted
in
non-significant
differences
in
ESs,
so
they
were
removed
from
subsequent
analyses.
In
addition,
all
studies
failed
to
report
performance
outcomes
as
a
function
of
gender
or
ethnicity.
Further,
the
majority
of
studies
either
did
not
report
SES
or
did
not
separate
it
between
the
children
with
and
without
RD,
so
these
variables
were
not
considered
further.
Table
1
shows
the
overall
performance
of
children
with
and
without
RD
on
norm-referenced
psychometric
information
(e.g.,
IQ
and
reading).
Results
showed
that
on
standardized
classification
measures
of
reading
comprehension
and
word
recognition,
children
with
RD
were
approximately
1
standard
deviation
below
the
mean
(where
M
=
100
and
SD
=
15).
Some
of
the
comparative
measures,
such
as
phonological
awareness,
word
attack
(pseudoword
reading),
and
spelling
skills
were
particular
points
of
weaknesses
for
children
with
RD
when
compared
to
children
without
RD,
many
scoring
at
or
below
the
25th
percentile
on
standard
scores.
3.2.
Domain
categories
The
weighted
means
and
standard
errors
for
the
ESs
within
each
category
are
reported
in
Table
2,
along
with
the
upper
and
lower
bounds
for
the
95%
confidence
interval
and
the
homogeneity
Q.
Using
criteria
provided
by
Cohen
(1988),
results
indicated
large
effect
sizes
(>0.80)
for
the
following
categories:
reading
comprehension,
word
recognition,
verbal
IQ,
rapid
naming,
phonological
awareness,
pseudoword
reading,
math,
vocabulary,
spelling,
writing,
and
behavior.
There
were
also
moderate
ESs
(0.50–0.80)
for
the
categories
of
general
IQ,
verbal
working
memory,
executive
processing/inhibition,
short-
term
memory,
auditory
processing
(auditory
perception
skills),
and
general
information.
The
remaining
categories,
problem
solving/reasoning,
visual–spatial
memory,
and
perceptual
motor
skills,
yielded
relatively
low
ESs
(<0.50).
3.3.
Mixed-level
modeling
Multi-level
modeling
was
used
to
identify
those
domains
that
significantly
moderated
overall
ESs
between
children
with
and
without
RD.
In
the
models
shown
in
Table
3,
level
2
represented
between-study
intercept
and
the
between
study
x
word
identification
variance.
For
the
analysis
of
random
effects,
word
recognition
values
were
converted
to
moderate
(>16th
percentile)
and
severe
(<16th
percentile)
values.
The
dependent
variable
of
interest
was
the
overall
weighted
ES
differences
between
children
with
and
without
RD.
Model
testing
examined
the
contributions
of
various
cognitive
or
academic
domains
(verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory),
in
moderating
overall
ES
differences
between
children
with
and
without
RD.
Because
the
reading
levels
and
the
age
of
children
varied
across
studies,
they
were
included
as
covariates
in
the
modeling.
As
shown
in
Table
3,
an
unconditional
model
(without
moderating
or
explanatory
variables)
was
computed
as
a
baseline
for
comparison
to
the
conditional
models.
Two
conditional
models
were
tested.
The
first
conditional
model
included
all
moderating
variables.
The
second
conditional
model
was
a
reduced
or
parsimonious
model
that
included
only
significant
explanatory
variables
identified
in
the
first
conditional
model.
The
conditional
models
were
compared
to
each
other,
as
well
as
to
an
unconditional
model.
This
comparison
was
done
by
determining
the
differences
between
the
deviance
values
(i.e.,
the
likelihood
value
for
the
correspondence
between
model
and
data)
from
the
unconditional
and
conditional
growth
model.
These
differences
are
chi-square
values,
and
the
number
of
parameters
added
for
the
conditional
model
served
as
degrees
of
freedom.
A
significantly
lower
deviance
score
for
the
conditional
model
indicated
that
the
conditional
model
showed
a
better
fit
to
the
data
than
the
unconditional
model.
Two
additional
indices
of
model
fit
were
included
in
the
analyses.
The
Akaike’s
Information
Criterion
(AIC)
allowed
for
a
comparison
of
models
that
were
not
nested,
and
the
Bayesian
information
criterion
(BIC)
allowed
for
a
comparison
of
nested
models
(Hox,
2010,
pp.
47–50).
In
general,
models
with
lower
deviance,
AIC,
and
BIC
values
fit
better
than
models
with
higher
values.
When
one
or
more
predictors
(explanatory
variables)
are
introduced
into
the
conditional
model,
the
reductions
in
the
magnitude
of
the
various
components
when
compared
to
the
unconditional
means
model
are
analogous
to
effect
sizes
(Snijders
&
Bosker,
1999).
For
the
unconditional
model,
Table
3
shows
a
fixed
effect
estimate
of
1.23
with
a
standard
error
(SE)
of
0.06.
The
variance
for
random
effect
was
significant
for
the
intercept
variance
between
studies
(0.07),
intercept
between
studies
and
word
recognition
(0.55),
and
the
estimated
variance
for
within
study
effects
(0.22).
The
full
model
included
all
possible
cognitive
moderators,
as
long
as
the
number
of
studies
that
included
the
cognitive
measures
exceeded
15
(K
>
15).
The
full
model
accounted
for
53%
of
the
explainable
variance
when
compared
to
the
unconditional
(0.55
+
0.02)
(0.05
+
0.24)/(0.55
+
0.02)]
and
provided
a
significantly
better
fit
than
the
unconditional
model
(
Dx
2
(15)
=
192.90
(1558.5–1365.6,
p
<
0.0001).
The
intercept
for
the
full
model
was
1.21
(SE
=
0.05).
In
addition,
several
cognitive
measures
(i.e.,
verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory)
were
significant
moderators
of
the
intercept.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
57
To
provide
a
parsimonious
account
of
the
data,
however,
a
reduced
model
was
tested
that
removed
all
the
nonsignificant
parameters
from
the
nested
model.
The
reduced
model
was
not
significantly
different
from
the
full
(nested)
model
(
Dx
2
(6)
=
13.70
(BIC:
1379.3–1365.6),
p
>
0.05.
Thus,
the
reduced
model
provided
a
more
parsimonious
fit
to
the
data.
When
compared
to
the
unconditional
model,
the
reduced
model
accounted
for
48%
of
the
explainable
variance
between
studies
(0.62
0.32/0.62).
The
reduced
model
revealed
a
significant
relationship
between
the
overall
ES
between
children
with
and
without
RD
and
various
moderators
related
to
cognition
(verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory)
and
academic
domains
(pseudoword
reading,
vocabulary,
and
spelling).
3.4.
Summary
In
summary,
this
synthesis
showed
that
large
effect
sizes
emerged
when
comparing
children
with
and
without
RD
for
reading
comprehension,
word
recognition,
verbal
IQ,
rapid
naming,
phonological
awareness,
pseudoword
reading,
math,
vocabulary,
spelling,
writing,
and
behavior.
Moderate
effect
sizes
were
also
found
for
general
IQ,
verbal
working
memory,
executive
processing,
short-term
memory,
auditory
processing
(auditory
perception
skills),
and
general
information.
Hierarchical
linear
modeling,
that
examined
the
various
cognitive
and
academic
domains,
found
that
several
cognitive
(verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory)
and
some
academic
(pseudoword
reading,
vocabulary,
and
spelling)
variables
significantly
moderated
the
overall
ESs,
when
accounting
for
the
initial
reading
levels
and
age
of
students
reported
in
the
studies.
The
reduced
model
accounted
for
approximately
48%
of
the
explainable
variance
across
the
studies.
4.
Discussion
This
study
synthesized
the
cognitive
process
differences
between
children
with
and
without
RD.
The
study
focused
on
two
issues.
The
first
focused
on
identifying
those
domains
that
contributed
to
the
largest
to
differences
between
children
with
and
without
RD.
As
shown
in
Table
2,
moderate
to
high
ESs
in
favor
of
children
without
RD
emerged
on
measures
of
cognition
(rapid
naming
speed
[ES
=
0.89],
phonological
awareness
[ES
=
1.00],
verbal
working
memory
[ES
=
0.68],
short-
term
memory
[ES
=
0.64],
and
perceptual
motor
skills
[ES
=
0.79]),
academic
achievement
(pseudoword
reading
[ES
=
1.84],
math
[ES
=
1.20],
vocabulary
[ES
=
0.83],
spelling
[ES
=
1.25],
and
writing
[ES
=
1.20]),
and
behavior
skills
(ES
=
0.80).
Hierarchical
linear
modeling
indicated
that
both
cognitive
(verbal
working
memory,
visual–spatial
memory,
executive
Table
3
Hierarchical
linear
modeling
predicting
effect
sizes
between
reading
disabled
and
average
readers.
Unconditional
model
Full
model
Reduced
model
Estimate
SE
Estimate
SE
Estimate
SE
Random
effects
Study
0.07
**
0.03
0.05
*
0.03
0.06
***
0.03
Study
*
word
ID
0.55
***
0.06
0.24
**
0.03
0.26
***
0.04
Residual
s
2
0.22
**
0.02
0.22
***
0.02
0.22
**
0.02
Fit
statistics
2
log
likelihood
1558.5
1365.6
1379.3
AIC
1566.5
1405.6
1405.3
BIC
1558.5
1365.6
1379.3
Fixed
effects
Intercept
1.23
***
0.06
1.21
***
0.05
1.22
***
0.05
Word
Id
1.12
***
0.13
1.25
***
0.13
Intelligence
0.87
***
0.13
0.78
***
0.13
Chronological
age
0.003
0.002
Rapid
naming
0.29
0.15
Phonological
awareness
0.008
0.14
Pseudoword
reading
0.76
***
0.18
0.84
***
0.18
Mathematics
0.26
0.18
Vocabulary
0.50
***
0.14
0.40
***
0.14
Spelling
0.68
***
0.24
0.80
**
0.24
Problem
solving
0.54
0.44
Verbal
WM
0.52
**
0.14
0.40
**
0.14
Visual–spatial
memory
0.70
**
0.17
0.59
***
0.18
Executive
processing
0.60
**
0.16
0.47
***
0.17
Short-term
memory
0.68
*
0.30
0.42
***
0.16
Behavior
0.41
0.28
*
p
<
0.05.
**
p
<
0.01.
***
p
<
0.001.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
58
processing,
short-term
memory,
and
perceptual
motor
skills)
and
academic
(pseudoword
reading,
vocabulary,
spelling)
measures
contributed
unique
and
significant
variance
to
the
overall
group
effect
size
differences.
The
second
issue
focused
on
whether
performance
differences
between
children
with
and
without
RD
was
moderated
differences
in
IQ
levels,
age,
ethnicity,
or
gender.
During
our
investigation
we
found
that
few
eligible
studies
reported
on
the
specific
effects
of
ethnicity
and
gender
and
therefore,
the
influence
of
these
variables
on
performance
outcomes
could
not
be
pursued
further.
As
shown
in
Table
3,
chronological
age
was
not
a
significant
moderator
of
the
overall
ES
between
children
with
and
without
RD.
In
contrast,
IQ
moderated
the
overall
outcomes,
even
when
competing
measures
(e.g.,
the
cognitive
and
academic
measures
that
had
been
found
to
be
significant)
were
entered
into
the
HLM
analysis.
The
present
synthesis
extends
the
literature
on
the
assessment
of
children
with
RD
in
three
major
ways.
First,
this
meta-
analysis
showed
that
RD
students
have
numerous
cognitive
lags
in
several
areas
compared
to
average
reading
students,
and
specifically
identifies
the
areas
in
need
of
intervention.
Those
areas
found
specifically
that
contributed
to
understanding
the
differences
between
RD
children
and
average
readers
were
in
phonological
processing
skills,
including
word
attack,
as
well
as
cognitive
and
academic
areas
including
spelling
and
vocabulary,
verbal
working
memory,
visual–spatial
memory,
executive
processing,
and
short-term
memory.
Furthermore,
while
this
and
previous
studies
(e.g.,
Johnson,
Humphrey,
Mellard,
Woods,
&
Swanson,
2010)
found
large
ESs
for
processing
speed
(i.e.,
rapid
naming)
and
math
ability,
our
HLM
analysis
did
not
find
processing
speed
or
math
ability
to
be
predictors
of
the
ES
between
children
with
and
without
RD.
It
should
be
noted,
however,
that
processing
speed
and
math
ability
are
predictors
of
effect
size
between
adults
without
and
without
RD
(Swanson,
2012;
Swanson
&
Hsieh,
2009).
These
results
suggest
that
understanding
the
differences
between
children
with
and
without
RD
among
children
should
not
include
processing
speed
or
math
ability,
but
that
these
areas
should
be
included
when
studying
adult
populations.
Future
research
is
needed
to
examine
the
differential
effects
of
processing
speed
and
math
ability
on
reading
ability
among
children
and
adult
populations.
Second,
direct
comparisons
were
made
across
studies
in
terms
of
variations
in
IQ
and
reading
level.
We
found
support
for
the
notion
that
IQ
was
a
valid
component
in
the
assessment
of
RD.
Several
researchers
have
suggested
eliminating
IQ
from
the
classification
of
RD.
Our
results
found
that
general
IQ
significantly
moderated
ES
differences
across
a
broad
array
of
measures.
That
is,
the
HLM
analysis
showed
that
variations
in
reading
did
not
partial
out
the
influence
of
general
IQ
in
predicting
ES
differences
between
children
with
and
without
RD.
Further,
we
found
variables
in
the
cognitive
and
academic
domains
that
there
were
significant
moderators
and
independently
predicted
ESs
between
RD
and
NRD
children
after
the
influence
of
all
other
variables
had
been
entered
into
the
analysis.
These
results
are
consistent
with
assessment
models
emphasizing
both
IQ
and
various
cognitive
measures
(i.e.,
verbal
working
memory,
visual–spatial
memory,
executive
processing,
short-term
memory,
and
perceptual
motor
skills)
in
the
assessment
of
RD.
Finally,
we
found
support
for
the
notion
that
problems
in
RD
extend
beyond
a
phonological
core
deficit.
Although
initially
the
analyses
found
clear
indications
of
weaknesses
in
comparative
processing
between
RD
and
skilled
readers
on
measures
of
phonological
awareness
(as
well
as
pseudoword
reading
and
spelling),
our
results
did
not
maintain
this
significance
upon
closer
examination.
Instead
the
outcomes
were
also
more
in
line
with
those
studies
indicating
that
other
cognitive
processes,
independent
of
phonological
awareness,
are
related
to
differences
between
RD
and
NRD
children,
such
as
those
on
memory
span
(Swanson
&
Jerman,
2007)
that
indicate
cognitive
processes
contribute
significant
amounts
of
variance
toward
reading
capability.
No
doubt,
the
above
finding
creates
a
conceptual
problem
when
one
attempts
to
link
RD
in
children
to
a
specific
or
core
phonological
processing
deficit.
Perhaps
one
obvious
means
of
reconciling
this
conceptual
problem
is
to
suggest
that
relationships
among
cognitive
processes
reflect
‘‘bootstrapping
effects’’
(see
Stanovich,
1986,
p.
364,
for
an
earlier
discussion
of
this
concept).
As
stated
by
Stanovich
(1986),
‘‘Many
things
that
facilitate
further
growth
in
reading.
.
.general
knowledge,
vocabulary.
.
.are
developed
by
reading
itself’’
(p.
364).
Thus,
due
to
the
mutual
facilitation
between
reading
and
cognitive
processing,
such
interrelationships
would
be
expected
to
increase
with
skill
improvement.
The
implicit
assumption
is
that
the
deficits
in
word
recognition
skills
(e.g.,
phonological
skills)
underlie
such
bootstrapping
effects.
Another
means
of
reconciling
the
phonological
core
issue
is
to
suggest
that
high-order
cognitive
processing
problems
can
exist
in
children
with
RD,
independent
of
their
specific
problems
in
low-order
processes,
such
as
phonological
processing.
Children
with
RD
may
be
viewed
as
having
difficulty
accessing
high-level
information
(as
reflected
in
their
reading
comprehension
and
vocabulary
scores)
and/or
lower-order
skills
(phonological
codes),
or
switching
between
the
two
levels
of
processing.
Thus,
one
may
speculate
that
the
processing
problems
in
children
with
RD
reflect
a
system
that
fails
to
compensate
for
(or
effectively
coordinate)
deficiencies
in
lower-order
specialized
processes.
This
lack
of
compensatory
processing
may
be
characterized
by
a
processing
system
either
not
contributing
enough
information
to
a
specialized
system
or
failing
to
provide
an
adequate
capacity
of
processing
resources
(i.e.,
because
of
verbal
memory
deficiencies),
given
that
there
are
problems
in
a
specialized
system.
Future
research
will
have
to
focus
on
the
interaction
between
higher
and
lower
order
processing
during
the
act
of
reading
to
disentangle
these
issues.
4.1.
Limitations
There
are
several
caveats
in
our
synthesis
that
limit
our
generalizations.
Three
are
considered.
First,
it
is
important
to
note
we
selected
studies
that
classified
children
with
RD
performing
at
various
levels
on
either
a
word
recognition
and/or
reading
comprehension
continuum.
Each
of
these
measures
draws
upon
different
processes
and
therefore,
may
have
obscured
the
results.
It
is
also
important
to
note
in
our
studies
that
one
of
the
most
frequent
identifiers
of
children
with
RD
provided
by
the
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
59
primary
authors
was
an
existing
discrepancy
between
the
targeted
sample’s
IQ
and
his/her
current
reading
achievement.
Thus,
our
use
of
aggregated
scores
related
to
IQ
and
reading
may
not
reflect
variables
that
underlie
how
subjects
were
selected
in
the
first
place.
The
second
was
that
although
we
selected
studies
that
included
only
samples
with
at
least
average
IQ
score
above
80
in
the
analysis
(and
no
indication
that
the
majority
were
below
80),
no
cap
was
placed
on
the
upper
limits
of
IQ
scores.
Some
studies
had
mean
IQ
scores
that
fell
within
a
level
that
might
be
referred
to
as
‘‘high,’’
and
therefore,
these
children
may
have
experienced
more
specific
deficits
in
cognitive
and
language
areas
than
children
closer
to
an
IQ
of
80.
It
is
possible
that
if
the
parameters
for
defining
the
non-discrepancy
groups
focused
on
higher
and
lower
IQ
groups,
as
well
as
different
types
of
IQ
measures
(e.g.,
nonverbal,
performance),
different
outcomes
may
have
emerged.
The
final
issue
was
that
our
selection
of
studies
was
biased
toward
those
that
included
samples
with
designated
labels
of
RD
(or
a
related
term)
as
well
as
samples
that
had
reported
both
intelligence
and
reading
scores.
Given
these
restrictions,
however,
this
meta-analysis
supports
previous
syntheses
showing
that
IQ
is
important
in
predicting
effect
size
differences
across
language,
behavioral
and
cognitive
variables
(Fuchs,
Fuchs,
Mathes,
&
Lipsey,
2000;
Hoskyn
&
Swanson,
2000).
4.2.
Summary
In
general,
several
cognitive
measures
yielded
a
unique
contribution
to
the
overall
ES
differences
between
groups.
For
example,
verbal
working
memory,
visual–spatial
memory,
executive
processing,
short-term
memory,
and
perceptual
motor
skills)
as
well
as
academic
(pseudoword
reading,
vocabulary,
and
spelling)
domains
were
found
to
be
significant
contributors
to
group
differences
in
ESs.
Because
chronological
age
was
not
found
to
be
a
significant
predictor
of
overall
differences,
it
appears
that
the
processing
difficulties
of
children
with
RD
remain
fairly
stable
across
a
broad
age
range.
References
1
Badian,
N.
(2001).
Phonological
and
orthographic
processing:
Their
roles
in
reading
prediction.
Annals
of
Dyslexia,
51(1),
177–202.
http://dx.doi.org/10.1007/
s11881-001-0010-5
Badian,
N.
(2005).
Does
a
visual-orthographic
deficit
contribute
to
reading
disability?
Annals
of
Dyslexia,
55(1),
28–52.
http://dx.doi.org/10.1007/s11881-005-
0003-x
Bollen,
K.
A.
(1989).
Structural
equations
with
latent
variables.
New
York:
John
Wiley
&
Sons
Inc.
*Bonifacci,
P.,
&
Snowling,
M.
J.
(2008).
Speed
of
processing
and
reading
disability:
A
cross-linguistic
investigation
of
dyslexia
and
borderline
intellectual
functioning.
Cognition,
107(3),
999–1017.
http://dx.doi.org/10.1016/j.cognition.2007.12.006
Bus,
A.
G.,
&
van
Ijzendoorn,
M.
H.
(1999).
Phonological
awareness
and
early
reading:
A
meta-analysis
of
experimental
training
studies.
Journal
of
Educational
Psychology,
91(3),
403–414.
http://dx.doi.org/10.1037/0022-0663.98.4.714
Catts,
H.
W.,
Gillispie,
M.,
Leonard,
L.
B.,
Kail,
R.
V.,
&
Miller,
C.
A.
(2002).
The
role
of
speed
of
processing,
rapid
naming,
and
phonological
awareness
in
reading
achievement.
Journal
of
Learning
Disabilities,
35(6),
510–525.
http://dx.doi.org/10.1177/00222194020350060301
Cohen,
J.
(1988).
Statistical
power
analysis
for
the
behavioral
sciences
(2nd
ed.).
Hillsdale,
NJ:
Erlbaum.
Compton,
D.
L.
(2003).
Modeling
the
relationship
between
growth
in
rapid
naming
speed
and
growth
in
decoding
skill
in
first-grade
children.
Journal
of
Educational
Psychology,
95(2),
225–239.
http://dx.doi.org/10.1037/0022-0663.95.2.225
Crisp,
J.,
&
Lambon
Ralph,
M.
A.
(2006).
Unlocking
the
nature
of
the
phonological–deep
dyslexia
continuum:
The
keys
to
reading
aloud
are
in
phonology
and
semantics.
Journal
of
Cognitive
Neuroscience,
18(3),
348–362.
http://dx.doi.org/10.1162/jocn.2006.18.3.348
Cronin,
V.
S.
(2013).
RAN
and
double-deficit
theory.
Journal
of
Learning
Disabilities,
46(2),
182–190.
http://dx.doi.org/10.1177/0022219411413544
Cunningham,
A.
E.,
&
Stanovich,
K.
E.
(1990).
Assessing
print
exposure
and
orthographic
processing
skill
in
children:
A
quick
measure
of
reading
experience.
Journal
of
Educational
Psychology,
82(4),
733–740.
http://dx.doi.org/10.1037/002-0663.82.4.733
Das,
J.
P.,
&
Mishra,
R.
K.
(1991).
Relation
between
memory
span,
naming
time,
speech
rate,
and
reading
competence.
Journal
of
Experimental
Education,
59(2),
129–
139.
http://dx.doi.org/10.1080/00220973.1991.10806556
Fuchs,
D.,
Fuchs,
L.,
Mathes,
P.
G.,
&
Lipsey,
M.
(2000).
Reading
differences
between
low-achieving
students
with
and
without
learning
disabilities:
A
meta-
analysis.
In
R.
Gersten,
E.
P.
Schiller,
&
S.
Vaughn
(Eds.),
Contemporary
special
education
research:
Synthesis
of
knowledge
base
of
critical
issues
(pp.
72–94).
Mahwah,
NJ:
Erlbaum.
Gottardo,
A.,
Collins,
P.,
Baciu,
I.,
&
Gebotys,
R.
(2008).
Predictors
of
grade
2
word
reading
and
vocabulary
from
grade
1
variables
in
Spanish-speaking
children:
Similarities
and
differences.
Learning
Disabilities
Research
&
Practice,
23(1),
11–24.
http://dx.doi.org/10.1111/j.1540-5826.2007.00259.x
Hedges,
L.
V.,
&
Olkin,
I.
(1985).
Statistical
methods
for
meta-analysis.
San
Diego,
CA:
Academic
Press.
Higgins,
J.,
&
Thompson,
S.
(2002).
Quantifying
heterogeneity
in
a
meta-analysis.
Statistics
in
Medicine,
21(11),
1539–1558.
http://dx.doi.org/10.1002/sim.1186
Hoskyn,
M.,
&
Swanson,
H.
L.
(2000).
Cognitive
processing
of
low
achievers
and
children
with
reading
disabilities:
A
selective
meta-analytic
review
of
the
published
literature.
School
Psychology
Review,
29(1),
102–119.
Hox,
J.
(2010).
Multilevel
analysis:
Techniques
and
applications.
Mahwah,
NJ:
Erlbaum.
Johnson,
E.
S.,
Humphrey,
M.,
Mellard,
D.
F.,
Woods,
K.,
&
Swanson,
H.
L.
(2010).
Cognitive
processing
deficits
and
students
with
specific
learning
disabilities:
A
selective
meta-analysis
of
the
literature.
Learning
Disability
Quarterly,
33(1),
3–18.
doi:10:622165831.
Kirby,
J.
R.,
Parrila,
R.
K.,
&
Pfeiffer,
S.
L.
(2003).
Naming
speed
and
phonological
awareness
as
predictors
of
reading
development.
Journal
of
Educational
Psychology,
95(3),
453–464.
http://dx.doi.org/10.1037/0022-0663.95.3.453
McBride-Chang,
C.,
Manis,
F.
R.,
Seidenberg,
M.
S.,
Custodio,
R.,
&
Doi,
L.
M.
(1993).
Print
exposure
as
a
predictor
of
word
reading
and
reading
comprehension
in
disabled
and
nondisabled
readers.
Journal
of
Educational
Psychology,
85(2),
230–238.
http://dx.doi.org/10.1037/0022-0663.85.2.230
Melby-Lerva
˚g,
M.,
Lyster,
S.
H.,
&
Hulme,
C.
(2012).
Phonological
skills
and
their
role
in
learning
to
read:
A
meta-analytic
review.
Psychological
Bulletin,
138(2),
322–
352.
http://dx.doi.org/10.1037/a0026744
Morris,
R.
D.,
Stuebing,
K.
K.,
Fletcher,
J.
M.,
Shaywitz,
S.
E.,
Lyon,
G.
R.,
Shankweiler,
D.
P.,
et
al.
(1998).
Subtypes
of
reading
disability:
Variability
around
a
phonological
core.
Journal
of
Educational
Psychology,
90(3),
347–373.
http://dx.doi.org/10.1037/0022-0663.90.3.347
Nation,
K.,
&
Snowling,
M.
J.
(1998).
Semantic
processing
and
the
development
of
word-recognition
skills:
Evidence
from
children
with
reading
comprehension
difficulties.
Journal
of
Memory
and
Language,
39(1),
85–101.
http://dx.doi.org/10.1006/jmla.1998.2564
Nelson,
J.
M.,
Lindstrom,
J.
H.,
&
Lindstrom,
W.
(2012).
The
structure
of
phonological
processing
and
its
relationship
to
basic
reading.
Exceptionality,
20(3),
179–196.
http://dx.doi.org/10.1080/09362835.2012.694612
1
Asterisk
denoting
that
articles
were
part
of
the
meta
analysis.
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
60
Nevo,
E.,
&
Breznitz,
Z.
(2013).
The
development
of
working
memory
from
kindergarten
to
first
grade
in
children
with
different
decoding
skills.
Journal
of
Experimental
Child
Psychology,
114(2),
217–228.
http://dx.doi.org/10.1016/j.jecp.2012.09.004
Parrila,
R.,
Kirby,
J.
R.,
&
McQuarrie,
L.
(2004).
Articulation
rate,
naming
speed,
verbal
short-term
memory,
and
phonological
awareness:
Longitudinal
predictors
of
early
reading
development?
Scientific
Studies
of
Reading,
8(1),
3–26.
http://dx.doi.org/10.1207/s1532799xssr
0801_2
Raudenbush,
S.
W.,
&
Bryk,
A.
S.
(2002).
Hierarchical
linear
models:
Applications
and
data
analysis
methods
(2nd
ed.).
Thousand
Oaks,
CA:
Sage
Publications
Inc.
SAS
Institute
(2010).
SAS/STAT
software:
Changes
and
enhancements
through
release
9.3.
Cary,
NC:
SAS
Institute
Inc.
Scarborough,
H.
S.
(1998).
Early
identification
of
children
at
risk
for
reading
disabilities:
Phonological
awareness
and
some
other
promising
predictors.
In
B.
Shapiro,
P.
Accardo,
&
A.
Capute
(Eds.),
Specific
reading
disability:
A
view
of
the
spectrum
(pp.
75–119).
Timonium,
MD:
York
Press.
Scarborough,
H.
S.
(2009).
Connecting
early
language
and
literacy
to
later
reading
(dis)abilities:
Evidence,
theory,
and
practice.
In
F.
Fletcher-Campbell,
J.
M.
Soler,
&
G.
Reid
(Eds.),
Approaching
difficulties
in
literacy
development:
Assessment,
pedagogy
and
programmes
(pp.
23–38).
Thousand
Oaks,
CA:
Sage
Publications,
Inc.
Schatschneider,
C.,
Fletcher,
J.
M.,
Francis,
D.
J.,
Carlson,
C.
D.,
&
Foorman,
B.
R.
(2004).
Kindergarten
predication
of
reading
skills:
A
longitudinal
comparative
analysis.
Journal
of
Educational
Psychology,
96(2),
265–282.
http://dx.doi.org/10.1037/0022-0663.96.2.265
Siegel,
L.
S.
(1993).
Phonological
processing
deficits
as
the
basis
of
a
reading
disability.
Developmental
Review,
13(3),
246–257.
http://dx.doi.org/10.1006/
drev.1993.1011
Siegel,
L.
S.
(2003).
IQ-discrepancy
definitions
and
the
diagnosis
of
LD:
Introduction
to
the
special
issue.
Journal
of
Learning
Disabilities,
36(1),
2–3.
http://dx.doi.org/
10.1177/00222194030360010101
Singer,
J.
D.
(2002).
Fitting
individual
growth
models
using
SAS
PROC
MIXED.
In
D.
S.
Moskowitz
&
S.
Hersberger
(Eds.),
Modeling
intraindividual
variability
with
repeated
measures
data:
Methods
and
applications
(pp.
135–170).
Mahwah,
NJ:
Erlbaum.
Shany,
M.,
&
Share,
D.
L.
(2011).
Subtypes
of
reading
disability
in
a
shallow
orthography:
A
double
dissociation
between
accuracy-disabled
and
rate-disabled
readers
of
Hebrew.
Annals
of
Dyslexia,
61(1),
64–84.
http://dx.doi.org/10.1007/s11881-010-0047-4
Snijders,
T.
A.,
&
Bosker,
R.
J.
(2003).
Multilevel
analysis.
Thousand
Oaks,
CA:
Sage.
Slavin,
R.
E.
(1986).
Best-evidence
synthesis:
An
alternative
to
meta-analytic
and
traditional
reviews.
Educational
Researcher,
15(9),
5–11.
http://dx.doi.org/
10.3102/0013189X015009005
Spencer,
L.
H.,
&
Hanley,
J.
R.
(2004).
Learning
a
transparent
orthography
at
five
years
old:
Reading
development
of
children
during
their
first
year
of
formal
reading
instruction
in
Wales.
Journal
of
Research
in
Reading,
27(1),
1–14.
http://dx.doi.org/10.1111/j.1467-9817.2004.00210.x
Stanovich,
K.
E.
(1986).
Matthew
effects
in
reading:
Some
consequences
of
individual
differences
in
the
acquisition
of
literacy.
Reading
Research
Quarterly,
21(4),
360–407.
http://dx.doi.org/10.1598/RRQ.21.4.1
Stanovich,
K.
E.
(1988).
Explaining
the
differences
between
the
dyslexic
and
the
garden
variety
poor
reader:
The
phonological-core
variable
difference
model.
Journal
of
Learning
Disabilities,
21(10),
590–604.
http://dx.doi.org/10.1177/002221948802101003
Stanovich,
K.
E.,
&
Siegel,
L.
S.
(1994).
Phenotypic
performance
profile
of
children
with
reading
disabilities:
A
regression
based
test
of
the
phonological–core
variable
difference
model.
Journal
of
Educational
Psychology,
86(1),
24–53.
http://dx.doi.org/10.1037/0022-0663.86.1.24
Swanson,
H.
L.
(2012).
Adults
with
reading
disabilities:
Converting
a
meta-analysis
to
practice.
Journal
of
Learning
Disabilities,
45(1),
17–30.
http://dx.doi.org/
10.1177/0022219411426856
Swanson,
H.
L.,
Harris,
R.
K.,
&
Graham,
S.
G.
(2003).
Handbook
of
learning
disabilities.
New
York,
NY:
Guilford
Press.
Swanson,
H.
L.,
&
Hsieh,
C.
(2009).
Reading
disabilities
in
adults:
A
selective
meta-analysis
of
the
literature.
Review
of
Educational
Research,
79(4),
1362–1390.
http://dx.doi.org/10.3102/0034654309350931
*Swanson,
H.
L.,
&
Jerman,
O.
(2007).
The
influence
of
working
memory
on
reading
growth
in
subgroups
of
children
with
reading
disabilities.
Journal
of
Experimental
Child
Psychology,
96(4),
249–283.
http://dx.doi.org/10.1016/j.jecp.2006.12.004
Torgesen,
J.
K.,
&
Mathes,
P.
G.
(2000).
A
basic
guide
to
understanding,
assessing
and
teaching
phonological
awareness.
Austin,
TX:
Pro-Ed.
Vellutino,
F.
R.,
Fletcher,
J.
M.,
Snowling,
M.
J.,
&
Scanlon,
D.
M.
(2004).
Specific
reading
disability
(dyslexia):
What
have
we
learned
in
the
past
four
decades?
Journal
of
Child
Psychology
and
Psychiatry,
45(1),
2–40.
http://dx.doi.org/10.1046/j.0021-9630.2003.00305.x
Wagner,
R.
K.,
&
Torgesen,
J.
K.
(1987).
The
nature
of
phonological
processing
and
its
causal
role
in
the
acquisition
of
reading
skills.
Psychological
Bulletin,
101(2),
192–212.
http://dx.doi.org/10.1037/0033-2909.101.2.192
Waterman,
B.,
&
Lewandowski,
L.
(1993).
Phonologic
and
semantic
processing
in
reading-disabled
and
nondisabled
males
at
two
age
levels.
Journal
of
Experimental
Child
Psychology,
55(1),
87–103.
http://dx.doi.org/10.1006/jecp.1993.1004
Zaretsky,
E.,
Kraljevic,
J.
K.,
Core,
C.,
&
Lencek,
M.
(2009).
Literacy
predictors
and
early
reading
and
spelling
skills
as
a
factor
of
orthography:
Cross-linguistic
evidence.
Written
Language
&
Literacy,
12(1),
52–81.
http://dx.doi.org/10.1075/wll.12.1.03zar
Further
reading
*Bishop,
D.
V.
M.,
McDonald,
D.,
Bird,
S.,
&
Hayiou-Thomas,
M.
E.
(2009).
Children
who
read
words
accurately
despite
language
impairment:
Who
are
they
and
how
do
they
do
it?
Child
Development,
80(2),
593–605.
http://dx.doi.org/10.1111/j.1467-8624.2009.01281.x
*Bowers,
P.
G.
(1995).
Tracing
symbol
naming
speeds
unique
contributions
to
reading
disabilities
over
time.
Reading
and
Writing,
7(2),
189–216.
http://dx.doi.org/
10.1007/BF01027185
*Bowers,
P.
G.,
Steffy,
R.,
&
Tate,
E.
(1988).
Comparisons
of
the
effects
of
IQ
control
methods
on
memory
and
naming
speed
predictors
of
reading-disability.
Reading
Research
Quarterly,
23(3),
304–319.
http://dx.doi.org/10.2307/748044
*Bruno,
J.
L.,
Manis,
F.
R.,
Keating,
P.,
Sperling,
A.
J.,
Nakamoto,
J.,
&
Seidenberg,
M.
S.
(2007).
Auditory
word
identification
in
dyslexic
and
normally
achieving
readers.
Journal
of
Experimental
Child
Psychology,
97(3),
183–204.
http://dx.doi.org/10.1016/j.jecp.2007.01.005
*Compton,
D.
L.,
Fuchs,
L.
S.,
Fuchs,
D.,
Lambert,
W.,
&
Hamlett,
C.
(2012).
The
cognitive
and
academic
profiles
of
reading
and
mathematics
learning
disabilities.
Journal
of
Learning
Disabilities,
45(1),
79–95.
http://dx.doi.org/10.1177/0022219410393012
*D’Anguilli,
A.,
&
Siegel,
L.
S.
(2003).
Cognitive
functioning
as
measured
by
the
WISC-R:
Do
children
with
learning
disabilities
have
distinctive
patterns
of
performance?
Journal
of
Learning
Disabilities,
36(1),
48–58.
http://dx.doi.org/10.1177/00222194030360010601
*Desroches,
A.
S.,
Joanisse,
M.
F.,
&
Robertson,
E.
K.
(2006).
Specific
phonological
impairments
in
dyslexia
revealed
by
eye
tracking.
Cognition,
100(3),
B32–B42.
http://dx.doi.org/10.1016/j.cognition.2005.09.001
*Fletcher,
J.
M.,
Shaywitz,
S.
E.,
Shankweiler,
D.
P.,
Katz,
L.,
Liberman,
I.
Y.,
Stuebing,
K.
K.,
et
al.
(1994).
Cognitive
profiles
of
reading
disability:
Comparisons
of
discrepancy
and
low
achievement
definitions.
Journal
of
Educational
Psychology,
86(1),
6–23.
http://dx.doi.org/10.1037/0022-0663.86.1.6
*Foorman,
B.
R.,
Francis,
D.
J.,
Fletcher,
J.
M.,
&
Lynn,
A.
(1996).
Relationship
between
phonological
and
orthographic
processing
in
early
reading:
Comparing
two
approaches
to
the
regression
based
reading
level
match
designs.
Journal
of
Educational
Psychology,
88(4),
639–652.
http://dx.doi.org/10.1037/0022-
0663.88.4.639
*Fuchs,
D.,
Compton,
D.
L.,
Fuchs,
L.
S.,
Bryant,
V.
J.,
Hamlett,
C.
L.,
&
Lambert,
W.
(2012).
First-grade
cognitive
abilities
as
long-term
predictors
of
reading
comprehension
and
disabilty
status.
Journal
of
Learning
Disabilities,
45(3),
217–231.
http://dx.doi.org/10.1177/0022219412442154
*Gibson,
L.
Y.,
Hogben,
J.
H.,
&
Fletcher,
J.
(2006).
Visual
and
auditory
processing
and
component
reading
skills
in
developmental
dyslexia.
Cognitive
Neuropsychology,
23(4),
621–642.
http://dx.doi.org/10.1080/02643290500412545
*Goswami,
U.,
Gerson,
D.,
&
Astruc,
L.
(2010).
Amplitude
envelope
perception,
phonology
and
prosodic
sensitivity
in
children
with
develomental
dyslexia.
Reading
and
Writing,
23(8),
995–1019.
http://dx.doi.org/10.1007/s11145-009-9186-6
*Irannejad,
S.,
&
Savage,
R.
(2012).
Is
a
cerebellar
deficit
the
underlying
cause
of
reading
disabilities?
Annals
of
Dyslexia,
62(1),
22–52.
http://dx.doi.org/10.1007/
s11881-011-0060-2
*Jerman,
O.,
Reynolds,
C.,
&
Swanson,
H.
L.
(2012).
Does
growth
in
working
memory
span
or
executive
processes
predict
growth
in
reading
and
math
in
children
with
reading
disabilities?
Learning
Disability
Quarterly,
35(3),
144–157.
http://dx.doi.org/10.1177/0731948712444276
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
61
*Katzir,
T.,
Kim,
Y.,
Wolf,
M.,
Kennedy,
B.,
Lovett,
M.,
&
Morris,
R.
(2006).
The
relationship
of
spelling
recognition,
RAN,
and
phonological
awareness
to
reading
skills
in
older
poor
readers
and
younger
reading-matched
controls.
Reading
and
Writing,
19(8),
845–872.
http://dx.doi.org/10.1007/s11145-006-0913-2
*Kibby,
M.
Y.
(2009).
There
are
multiple
contributors
to
the
verbal
short-term
memory
deficit
in
children
with
developmental
reading
disabilty.
Child
Neuropsychology,
15(5),
485–506.
http://dx.doi.org/10.1080/09297040902748218
*Locascio,
G.,
Mahone,
E.
M.,
Eason,
S.
H.,
&
Cutting,
L.
E.
(2010).
Executive
dysfunction
among
children
with
reading
comprehension
deficits.
Journal
of
Learning
Disabilities,
43(5),
441–454.
http://dx.doi.org/10.1177/0022219409355476
*Marshall,
C.
R.,
Harcourt-Brown,
S.,
Ramus,
F.,
&
Van
Der
Lely,
H.
K.
J.
(2009).
The
link
between
prosody
and
language
skills
in
children
ith
speific
language
impairment.
International
Journal
of
Languge
&
Communication
Disorders,
44(4),
466–488.
http://dx.doi.org/10.1080/13682820802591643
*McArthur,
G.
M.,
Ellis,
D.,
Atkinson,
C.
M.,
&
Coltheart,
M.
(2008).
Auditory
processing
deficits
in
children
with
reading
and
language
impairments:
Can
they
(and
should
they)
be
treated?
Cognition,
107(3),
946–977.
http://dx.doi.org/10.1016/j.cognition.2007.12.005
*Mellanby,
J.,
Anderson,
R.,
Campbell,
B.,
&
Westwood,
E.
(1996).
Cognitive
determinants
of
verbal
underachievement
at
secondary
school
level.
British
Journal
of
Educational
Psychology,
66(4),
483–500.
http://dx.doi.org/10.1111/j.2044-8279.1996.tb01214.x
*Robertson,
E.
K.,
&
Joanisse,
M.
F.
(2010).
Spoken
sentence
comprehension
in
children
wih
dyslexia
and
language
impairment:
The
roles
of
syntax
and
working
memory.
Applied
Psycholinguistics,
31(1),
141–165.
http://dx.doi.org/10.1017/S0142716409990208
*Robertson,
E.
K.,
Joanisse,
M.
F.,
Desroches,
A.
S.,
&
Ng,
S.
(2009).
Categorical
speech
perception
deficits
distinguish
language
and
reading
impairments
in
children.
Developmental
Science,
12(5),
753–767.
http://dx.doi.org/10.1111/j.1467-7687.2009.00806x
*Shanahan,
M.
A.,
Pennington,
B.
F.,
Yerys,
B.
E.,
Scott,
A.,
Boada,
R.,
Willcutt,
E.
G.,
et
al.
(2006).
Processing
speed
deficits
in
attention
deficit/hyperactivity
disorder
and
reading
disability.
Journal
of
Abnormal
Child
Psychology,
34(5),
585–602.
http://dx.doi.org/10.1007/s10802-006-0937-8
*Shaywitz,
S.
E.,
Holahan,
J.
M.,
Freudenheim,
D.
A.,
Fletcher,
J.
M.,
Makuch,
R.
W.,
&
Shaywitz,
B.
A.
(2001).
Heterogeneity
within
the
gifted:
Higher
IQ
boys
exhibit
behaviors
resembling
boys
with
learning
disabilities.
Gifted
Child
Quarterly,
45(1),
16–23.
http://dx.doi.org/10.1177/001698620104500103
*Swanson,
H.
L.
(1988).
Memory
subtypes
in
learning
disabled
readers.
Learning
Disability
Quarterly,
11(4),
342–357.
http://dx.doi.org/10.2307/1510781
*Swanson,
H.
L.
(1993a).
An
information
processing
analysis
of
learning
disabled
children’s
problem
solving.
American
Educational
Research
Journal,
30(4),
861–
893.
http://dx.doi.org/10.2307/1163207
*Swanson,
H.
L.
(1993b).
Executive
processing
in
learning-disabled
readers.
Intelligence,
17(2),
117–149.
http://dx.doi.org/10.1016/0160-2896(93)90024-Y
*Swanson,
H.
L.
(1993c).
Individual
differences
in
working
memory:
A
model
testing
and
subgroup
analysis
of
learning-disabled
and
skilled
readers.
Intelligence,
17(3),
285–332.
http://dx.doi.org/10.1016/0160-2896(93)90003-N
*Swanson,
H.
L.
(1999).
Reading
comprehension
and
working
memory
in
learning-disabled
readers:
Is
the
phonological
loop
more
important
than
the
executive
system?
Journal
of
Experimental
Child
Psychology,
72(1),
1–31.
http://dx.doi.org/10.1006/jecp.1998.2477
*Swanson,
H.
L.,
&
Alexander,
J.
E.
(1997).
Cognitive
processes
as
predictors
of
word
recognition
and
reading
comprehension
in
learning-disabled
and
skilled
readers:
Revisiting
the
specificity
hypothesis.
Journal
of
Educational
Psychology,
89(1),
128–158.
http://dx.doi.org/10.1037/0022-0663.89.1.128
*Swanson,
H.
L.,
&
Ashbaker,
M.
H.
(2000).
Working
memory,
short-term
memory,
speech
rate,
word
recognition
and
reading
comprehension
in
learning
disabled
readers:
Does
the
executive
system
have
a
role?
Intelligence,
28(1),
1–30.
http://dx.doi.org/10.1016/S0160-2896(99)00025-2
*Swanson,
H.
L.,
&
Howard,
C.
B.
(2005).
Children
with
reading
disabilities:
Does
dynamic
assessment
help
in
the
classification?
Learning
Disability
Quarterly,
28(1),
17–34.
http://dx.doi.org/10.2307/4126971
*Swanson,
H.
L.,
Howard,
C.
B.,
&
Saez,
L.
(2006).
Do
different
components
of
working
memory
underlie
different
subgroups
of
reading
disabilities?
Journal
of
Learning
Disabilities,
39(3),
252–269.
http://dx.doi.org/10.1177/00222194060390030501
*Swanson,
H.
L.,
Kehler,
P.,
&
Jerman,
O.
(2010).
Working
memory,
strategy
knowledge,
and
strategy
instruction
in
children
with
reading
disabilities.
Journal
of
Learning
Disabilities,
43(1),
24–47.
http://dx.doi.org/10.1177/0022219409338743
*Swanson,
H.
L.,
Orosco,
M.
J.,
&
Lussier,
C.
M.
(2011).
Cognition
and
literacy
in
English
language
learners
at
risk
for
reading
disabilties.
Journal
of
Educational
Psychology,
103(4),
1–19.
http://dx.doi.org/10.1037/a0026225
*Swanson,
H.
L.,
&
Ramalgia,
J.
M.
(1992).
The
relationship
between
phonological
codes
on
memory
and
spelling
tasks
for
students
with
and
without
learning
disabilities.
Journal
of
Learning
Disabilities,
25(6),
396–407.
http://dx.doi.org/10.1177/002221949202500608
*Swanson,
H.
L.,
Reffel,
J.,
&
Trahan,
M.
(1991).
Naturalistic
memory
in
learning-disabled
and
skilled
readers.
Journal
of
Abnormal
Child
Psychology,
19(2),
117–147.
http://dx.doi.org/10.1007/BF00909975
*Swanson,
H.
L.,
&
Sachse-Lee,
C.
(2001a).
A
subgroup
analysis
of
working
memory
in
children
with
reading
disabilities:
Domain-general
or
domain-specific
deficiency?
Journal
of
Learning
Disabilities,
34(3),
249–263.
http://dx.doi.org/10.1177/002221940103400305
*Swanson,
H.
L.,
&
Sachse-Lee,
C.
(2001b).
Mathematical
problem
solving
and
working
memory
in
children
with
learning
disabilities:
Both
executive
and
phonological
processes
are
important.
Journal
of
Experimental
Child
Psychology,
79(3),
294–321.
http://dx.doi.org/10.1006/jecp.2000.2587
*Swanson,
H.
L.,
Sa
´ez,
L.,
Gerber,
M.,
&
Leafstedt,
J.
(2004).
Literacy
and
cognitive
functioning
in
bilingual
and
nonbilingual
children
at
or
not
at
risk
for
reading
disabilities.
Journal
of
Educational
Psychology,
96(1),
3–18.
http://dx.doi.org/10.1037/0022-0663.96.1.3
*Swanson,
H.
L.,
&
Trahan,
M.
(1990).
Naturalistic
memory
in
learning-disabled
children.
Learning
Disability
Quarterly,
13(2),
82–95.
http://dx.doi.org/10.2307/
1510653
*Swanson,
H.
L.,
&
Trahan,
M.
(1996).
Learning
disabled
and
average
readers’
working
memory
and
comprehension:
Does
metacognition
play
a
role?
British
Journal
of
Educational
Psychology,
66(3),
333–355.
http://dx.doi.org/10.1111/j.2044-8279.1996.tb01201.x
*Tal,
N.
F.,
&
Siegel,
L.
S.
(1996).
Pseudoword
reading
errors
of
poor,
dyslexic,
and
normally
achieving
readers
on
multisyllable
pseudowords.
Applied
Psycholinguistics,
17(2),
215–232.
http://dx.doi.org/10.1017/S0142716400007645
*Tanaka,
H.,
Black,
J.
M.,
Hulme,
C.,
Stanley,
L.
M.,
Kesler,
S.
R.,
Whitfield-Gabrieli,
S.,
et
al.
(2011).
The
brain
basis
of
the
phonologial
deficit
in
dyslexia
is
independent
of
IQ.
Psychologial
Science,
22(11),
1442–1451.
http://dx.doi.org/10.1177/0956797611419521
*Thomson,
J.
M.,
Richardson,
U.,
&
Goswami,
U.
(2005).
Phonological
similarity
neighborhoods
and
children’s
short-term
memory:
Typical
development
and
dyslexia.
Memory
&
Cognition,
33(7),
1210–1219.
http://dx.doi.org/10.3758/BF03193223
*White,
S.,
Milne,
E.,
Rosen,
S.,
Hansen,
P.,
Swettenham,
J.,
Frith,
U.,
et
al.
(2006).
The
role
of
sensorimotor
impairments
in
dyslexia:
A
multiple
case
study
of
dyslexic
children.
Developmental
Science,
9(3),
237–255.
http://dx.doi.org/10.1111/j.1467-7687.2006.00483.x
M.F.
Kudo
et
al.
/
Research
in
Developmental
Disabilities
40
(2015)
51–62
62
