ArticlePDF Available

Diagnosis and Treatment of Reading Disabilities Based on the Component Model of Reading: An Alternative to the Discrepancy Model of LD


Abstract and Figures

Currently, learning disabilities (LD) are diagnosed on the basis of the discrepancy between students' IQ and reading achievement scores. Students diagnosed with LD often receive remedial instruction in resource rooms. The available evidence suggests that the educational policy based on this discrepancy model has not yielded satisfactory results. This has led researchers to try other paradigms, such as the component model and response to intervention, for dealing with children with reading disabilities. The component model of reading (CMR) described in the present article identifies the reading component that is the source of reading difficulty and targets instruction at that component. Study 1 describes the CMR and reports on its validity. Study 2 describes the successful outcome of a 7-year CMR-based reading instruction program. Compared to the discrepancy model, the CMR has demonstrated several advantages.
Content may be subject to copyright.
Journal of Learning Disabilities
The online version of this article can be found at:
DOI: 10.1177/0022219407310838
2008 41: 67J Learn Disabil
P.G. Aaron, R. Malatesha Joshi, Regina Gooden and Kwesi E. Bentum
Alternative to the Discrepancy Model of LD
Diagnosis and Treatment of Reading Disabilities Based on the Component Model of Reading : An
Published by:
Hammill Institute on Disabilities
can be found at:Journal of Learning DisabilitiesAdditional services and information for Alerts:
What is This?
- Jan 17, 2008Version of Record >>
by guest on June 2, 2013ldx.sagepub.comDownloaded from
Diagnosis and Treatment of Reading
Disabilities Based on the Component
Model of Reading
An Alternative to the Discrepancy Model of LD
P. G. Aaron
Indiana State University
R. Malatesha Joshi
Texas A&M University
Regina Gooden
St. Thomas University
Kwesi E. Bentum
Bi-County Special Education Cooperative, Sterling, Illinois
Currently, learning disabilities (LD) are diagnosed on the basis of the discrepancy between students’ IQ and reading
achievement scores. Students diagnosed with LD often receive remedial instruction in resource rooms. The available evi-
dence suggests that the educational policy based on this discrepancy model has not yielded satisfactory results. This has led
researchers to try other paradigms, such as the component model and response to intervention, for dealing with children
with reading disabilities. The component model of reading (CMR) described in the present article identifies the reading
component that is the source of reading difficulty and targets instruction at that component. Study 1 describes the CMR and
reports on its validity. Study 2 describes the successful outcome of a 7-year CMR-based reading instruction program.
Compared to the discrepancy model, the CMR has demonstrated several advantages.
Keywords: component model of reading; discrepancy model of learning disability; cognitive domain; psychological domain;
ecological domain; differential diagnosis; word-recognition training; comprehension strategy training
early 38% of the children in U.S. fourth-grade class-
rooms have been identified as reading below the basic
reading level (National Center for Education Statistics,
1998). A substantial number of these children have been
identified as having learning disabilities (LD) and receive
special instruction in resource rooms. There is no uniform
educational policy for teaching the remaining poor read-
ers, who are said to fall through the cracks. Therefore, the
LD construct, when applied in its present form, leads to
many poor readers being “left behind.
In spite of the fact that schools have implemented
the LD-based policy for nearly 40 years, many studies
have shown that the existing diagnostic procedure is
unreliable and that the instructional methods are ineffec-
tive. Because of this disappointing outcome, educators,
researchers, and advocacy groups are trying to find better
methods for the identification and treatment of reading
problems. The response to intervention model is one
such approach (Bradley, Danielson, & Doolittle, 2005;
D. Fuchs, Mock, Morgan, & Young, 2003), and the com-
ponent model of reading presented in this article is yet
another approach. Whereas simpler but effective ways of
identification and remediation of reading difficulties
may be possible than the ones that are used in today’s
schools, the present LD policy implementation appears
to be unable to rid itself of its tradition due to the fact that
it is ensnared in its past history.
The existence of the condition known today as LD was
recognized almost 100 years ago when it was noticed that
some children who apparently were intelligent experi-
enced a great deal of difficulty in learning to read. During
the early period of its history, this condition drew the
attention of many investigators, many of whom were
physicians, who described it by labels such as word blind-
ness (Hinshelwood, 1895; Morgan, 1896), strephosymbo-
lia (Orton, 1937), dyslexia, attributed to Dr. Rudolph
Berlin, and finally, learning disability (Kirk, 1963). In
general, physicians tended to describe reading difficulty in
Journal of Learning Disabilities
Volume 41 Number 1
January/February 2008 67-84
© 2008 Hammill Institute on
hosted at
by guest on June 2, 2013ldx.sagepub.comDownloaded from
neurological terms, but educators viewed it as an educa-
tional problem. In spite of these early differences in orien-
tation, reading difficulty came to be recognized as a
serious pedagogical problem that affected many children.
This realization is evident from the wide acceptance of the
term learning disabilities, which was introduced into the
educational scene in 1963 by Samuel Kirk. The concept of
LD gained official status in 1975 with the passing of the
Education for All Handicapped Children Act. The official
recognition that LD are a form of disability made the priv-
ileges and accommodations to which individuals with dis-
abilities are entitled accessible to individuals with reading
problems. It became necessary, therefore, to develop an
objective means of identifying and diagnosing LD, partic-
ularly in children in the school systems. Because LD are
defined in terms of average or above-average intelligence
but below-average reading performance, it appeared that a
logical means of diagnosing LD would be to assess the IQ
of children who are suspected of having LD and compare
their reading achievement scores with their IQ scores. If
an individual’s IQ was in the average range but the read-
ing achievement score was noticeably lower, that individ-
ual was diagnosed as having LD. This way of identifying
LD has, therefore, come to be referred to as the discrep-
ancy model–based procedure. In many schools, children
identified as having LD on the basis of this discrepancy
are placed in resource rooms and receive special instruc-
tion, although recently there has been a move to teach
children with LD in the general education classroom.
As the years passed, it became possible to take a closer
look at the validity and utility of the discrepancy model
and, as a result, it has come to be realized that the model,
as it is used for diagnosing and treating reading problems,
has failed to deliver the expected academic benefits
(Aaron, 1997b). The disappointing outcome of the discrep-
ancy model–based educational policy naturally impelled
researchers to examine the potential reasons for its failure
(e.g., Lyon et al., 2001). This scrutiny of the LD program
identified many reasons for the disappointing outcome, but
the most formidable problem faced by the discrepancy
model is that children who are identified as having LD and
provided with instruction in resource rooms have failed to
show improvement in their reading skills, as documented
by several researchers (Bentum & Aaron, 2003; Carlson,
1997; D. Fuchs & Fuchs, 1995; Haynes & Jenkins, 1986;
Moody, Vaughn, Hughes, & Fischer, 1998; Wleklenski,
1993). In fact, the studies by Wleklenski (1993) and
Bentum and Aaron (2003) found a significant decline in
the verbal IQ and spelling scores of children taught in
resource rooms—an instance of the “Matthew effect.
There could be many reasons for the lack of success
of resource room LD instruction. Among these are the
high pupil–teacher ratio and the placement of children
with behavioral and emotional problems as cohorts of
children with LD in the resource room. The research reports
mentioned earlier, however, have indicated that the pri-
mary reason for the poor outcome of LD instruction is
the unsystematic way in which children are taught in many
of the resource rooms. More specifically, there is a lack
of uniformity in the instructional methods for teaching
children with LD, as the discrepancy model does not pro-
vide the LD teachers with directions for instruction. In a
review study, Vaughn, Levy, Coleman, and Bos (2002)
synthesized studies conducted on students with LD and
reported that the quality of reading instruction was poor,
with excessive time allocated to seatwork and worksheets
but limited time given to reading itself. After observing
what went on in resource rooms, Haynes and Jenkins
(1986) and Moody et al. (1998) noted that the quality of
reading instruction provided was not based on a skills
approach, but was driven by the whole-language philos-
ophy and relied mainly on group work, which disregarded
individual needs. These observations lead to the ques-
tion, “What are the needs of poor readers, and which
skills should be addressed in resource rooms?”
Several studies have shown that not all poor readers are
alike and that reading difficulties are varied in origin
(Aaron, Joshi, & Williams, 1999; Catts, Hogan, & Fey,
2003; Swanson, 1999; Swanson, Howard, & Saez, 2006).
To be more specific, from a cognitive perspective, some
children may have difficulty at the word recognition level,
others at the comprehension level, and still others may be
poor readers because of limited vocabulary. It is reasonable
to expect instruction in word recognition to improve the
performance of the first type of readers, comprehension
instruction to help the second type of readers, and vocabu-
lary instruction to help the third type of readers. Effective
instruction, therefore, requires knowledge about what skills
make up the reading process and how to identify the weak
component that leads to reading difficulty. The component
model of reading provides a tentative answer to these
requirements (Aaron, 1997a; Aaron & Kotva, 1999; Joshi
& Aaron, 2000). More specifically, the component model
of reading identifies the weak component that underlies
reading difficulties and focuses remedial efforts at this
weak component. A component is defined as an elementary
information processing system that operates on internal
representations of objects and symbols. To be considered a
component, the process should be demonstrably independent
of other cognitive processes (Sternberg, 1985).
The Component Model of Reading
The literacy performance of children in the classroom
is affected not only by cognitive factors, but also by
68 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
environmental and psychological factors (e.g., Berninger,
Dunn, Lin, & Shimada, 2004; Dudley-Marling, 2004). The
component model of reading is broadly conceptualized and
takes this fact into account. Components that have an
influence on the acquisition of literacy skills are organized
into three domains and constitute the component model
of reading (CMR). The three domains of the CMR are
(a) the cognitive domain, (b) the psychological domain,
and (c) the ecological domain.
The cognitive domain of the CMR has two components:
word recognition and comprehension. The psychological
domain includes components such as motivation and inter-
est, locus of control, learned helplessness, learning styles,
teacher expectation, and gender differences. The ecologi-
cal domain includes the components of home environment
and culture, parental involvement, classroom environment,
dialect, and speaking English as a second language. It has
to be added that the components of the cognitive domain
can satisfy the condition of independence fairly well,
whereas the components of the psychological and ecolog-
ical domains do not satisfy this requirement nearly as well.
Nevertheless, the CMR provides a framework for teachers
and psychologists for navigating their course through the
various assessment formats and determining remedial
strategies for use in the classroom. The importance of these
psychological and environmental factors has been recog-
nized by educators for a long time and has been also empir-
ically documented (Berninger et al., 2004; Dudley-Marling,
2004). The three domains of the CMR and their constituent
components are shown in Figure 1.
When applied to literacy acquisition, the CMR envis-
ages that a child can fail to acquire satisfactory levels of
literacy skills because of deficits in any component in any
one of these three domains. The focus of the present article
is the constructional and instructional validity of the
components of the cognitive domain of the CMR; it does
not address the psychological and environmental domains.
In the first part of the article, the nature of the compo-
nents of the cognitive domain of the CMR is described,
along with supporting evidence that authenticates the
validity of the model. In the second part of the article, the
outcome of the implementation of instruction based on
the CMR during the course of a 7-year period is com-
pared with the outcome of LD resource room–based
reading instruction.
The inspiration for the CMR comes from a report by
Gough and Tunmer (1986) who presented a simple view
of reading by noting that the two most important con-
stituents of reading are the ability to decode words and the
ability to comprehend text. Gough and Tunmer expressed
this proposition in the form of a formula: R = D × L,
wherein R is reading comprehension, D is decoding, and
L is linguistic comprehension, as assessed by a test of
listening comprehension. They set the value of each vari-
able to range from 0 to 1. It follows, then, if D is 0, then
R is 0; if L is 0, then R is also 0. The validity of the for-
mula was tested by Hoover and Gough (1990) by track-
ing and assessing 254 English–Spanish bilingual children
from Grades 1 through 4. The investigators found that a
substantial proportion of the variance in reading compre-
hension was accounted for by the product of decoding
and listening comprehension (Grade 1, r = .71; Grade 2,
r = .72; Grade 3, r = .83; Grade 4, r = .82).
The linguistic comprehension component, as used in
the Gough-Tunmer formula, represents listening com-
prehension and is assessed by using a test of listening
comprehension. However, it should be noted that the cor-
relation between reading comprehension and listening
comprehension is high, usually in the vicinity of .80, par-
ticularly for children in upper elementary grades and
adolescents, so that listening comprehension can be used
as a predictor of reading comprehension (Joshi, Williams,
& Wood, 1998; Kintsch & Kozminsky, 1977; Palmer,
McCleod, Hunt, & Davidson, 1985). The other compo-
nent, word recognition, includes two processes: (a) the
ability to decode written words, and (b) the ability to
decode words instantly and automatically. Developmental
studies of reading show that many children are slow in
decoding written words until they reach the third grade
or so, by which time most children have learned to decode
written words instantly and automatically, a process tra-
ditionally referred to as sight word reading.
The simple view of reading, as is true of most psy-
chological theories, has not gone unchallenged. Duke
et al. (2006), for example, have faulted the simple view of
reading by noting that it has left out many variables, includ-
ing vocabulary knowledge, motivation, and the cultural
background of the reader. Nevertheless, in defense of the
Aaron et al. / Component Model of Reading 69
Figure 1
Overview of the Component Model of Reading
by guest on June 2, 2013ldx.sagepub.comDownloaded from
simple view of reading, it has to be noted that Gough and
Tunmer (1986) did not imply that reading is a simple
process but that the information-processing aspect of
reading can be explained simply by the product of D and L.
Of course, it is common knowledge that factors such as
vocabulary knowledge, motivation, and the amount of read-
ing that takes place at home are all factors that contribute
to reading achievement.
Duke et al. (2006) also stressed that speed of processing
is another important element left out of the simple view of
reading. After the publication of the report of the National
Reading Panel (2000), this aspect of reading, which is
included in reading fluency, has received a considerable
amount of research attention. There is unanimous agree-
ment among educators and researchers that fluency is a
hallmark of good readers. What is not agreed upon is
whether speed of processing is a component that is inde-
pendent of decoding skill. Studies by Adlof, Catts, Hogan,
and Little (2005), Cho and McBride-Chang (2005), and
Vukovic and Siegel (2006) have shown that speed of pro-
cessing adds little variance to reading performance that is
not explained by word recognition and comprehension
skills. This conclusion can be backed up by the observation
that all poor decoders are also slow readers and that slow
readers, in general, are also poor decoders, thereby making
room for accommodating fluency under the word recogni-
tion component. Therefore, speed of processing is included
in the present study as a further test of its contribution to
reading comprehension.
The CMR, as described in the present article, is an
elaboration of the simple view of reading created by
adding psychological and environmental factors to the
simple view of reading. The present investigation, how-
ever, is limited only to the cognitive domain of reading
and does not include the psychological and environmen-
tal factors. The motivation is to compare the effective-
ness of the diagnostic and instructional practices based
on the discrepancy model and the CMR.
Evidence for concluding that reading is made up of two
cognitive components, word identification and compre-
hension, comes from three sources: experimental, devel-
opmental, and neuropsychological studies.
Experimental Studies of Reading Components
In a study of undergraduate students, Jackson and
McClelland (1979) found that reaction time and compre-
hension in a letter-matching task accounted for nearly all of
the variance seen in reading ability. Similar results were
reported by Palmer et al. (1985). More recently, Catts et al.
(2003), in a longitudinal study of children in early elemen-
tary grades, reported a low correlation of .16 between word
recognition and listening comprehension, indicating their
relative independence. The independent nature of word
recognition and comprehension was also reported by
Carver (1998), Catts and Kamhi (1999), and de Jong and
van der Leij (2002). Other investigators have described
word recognition skills and comprehension skills as lower
level and higher level processing skills, respectively
(Hannon & Daneman, 2001; Pressley, 2000). The verbal
efficiency theory (Perfetti, 1988) captures the essence of
the bicomponential nature of reading by stressing the inde-
pendent role of verbal efficiency and text comprehension.
Neuropsychological Studies of Reading
Neuropsychological studies have indicated that some
patients could comprehend individually presented words
much better than they could pronounce them, whereas
some other patients could decode such words but had
diminished ability to comprehend them. These instances of
reading failures, which have been labeled deep dyslexia
and surface dyslexia, respectively, indicate that word
recognition and comprehension are independent compo-
nents of reading (Coltheart, Patterson, & Marshall, 1980;
Patterson, Marshall, & Coltheart, 1985). This conclusion is
further supported by recent neuroimaging studies. For
instance, Poldrack and Wagner (2004), using neuroimag-
ing techniques, have shown that different cortical structures
are involved in the retention of phonological information
and semantic information in working memory. Using the
functional magnetic resonance imaging (fMRI) technique,
McDermott, Watson, and Ojemann (2006) reported robust
semantic and phonological processing differences in
underlying frontal and temporal language networks.
Developmental Studies of Reading Components
Frith and Snowling (1983) reported that some children
with autism can read aloud much better than they can
comprehend, whereas children with dyslexia can compre-
hend sentences better than they can decode nonwords.
Studies on children with dyslexia and hyperlexia also
showed that comprehension and decoding skills are dis-
sociable (Aaron, Franz, & Manges, 1990; Healy, 1982).
It has also been reported that a substantial number of poor
readers have deficits in decoding skills but have better
comprehension skills, as determined by their perfor-
mance on tests of listening comprehension (Crain, 1989;
Shankweiler et al., 1995). The existence of children who
can decode written words fairly well but cannot compre-
hend what they have read is less well publicized, even
though educators have recognized the existence of this
70 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
type of poor readers for a long time and have described
them as “word callers.” Research has indicated that about
10% of poor readers fall into this category (Stothard, 1994;
Yuill & Oakhill, 1991). According to Mirak, Scarborough,
and Rescorla (2003), some children who had average
word-level processing skills in earlier grades turned out to
have deficits in reading comprehension when they reached
fourth and fifth grades, probably because the ability to
comprehend what is read emerges later in development.
This study further indicated that the two components of
reading show different developmental trajectories and,
therefore, it is important to diagnose which of the two
components is lagging behind, so that appropriate reme-
dial instruction can be provided.
This brief survey of studies indicates that word recogni-
tion and comprehension are separable component skills of
the cognitive domain of reading. The independent nature
of components, therefore, carries with it the potential for
resulting in poor readers of more than one type. Study 1
was conducted to ascertain the validity of this bicompo-
nential nature of reading in elementary school children and
to estimate the relative contribution of word recognition
and comprehension skills to the reading process. It was
noted earlier that speed of processing has also been sug-
gested by some researchers to be a major contributor to
reading performance. For this reason, the contribution of
speed of processing was also investigated in Study 1.
Study 1: Validation of the CMR
Teachers of children from Grades 2 through 5 from
seven different schools in the southwestern part of the
United States were asked to identify children in their
classrooms who were willing to participate in a testing
project. Subsequently, reading-related tests were adminis-
tered to a total of 204 children. These children came from
families that can be broadly defined as middle class.
About 10% of the children were from minorities, but all
children used English as their language of communication
at home. Demographic details regarding these children are
shown in Table 1.
All the children completed the Reading Comprehension
subtest from the Gates-MacGinitie Test of Reading
(MacGinitie & MacGinitie, 1989). Their Listening
Comprehension and Word Attack skills were assessed by
using subtests from the Woodcock Language Proficiency
Battery (Woodcock, 1991). Moreover, an informal letter
naming task was also administered to these children as a
measure of processing speed. The letter naming task was
constructed by typing 40 lowercase letters of the alphabet,
which were printed in 16-point font size. The children were
asked to name the letters as fast as they could without mak-
ing mistakes. They were asked to ignore any error they
thought they made and to move on to the next letter. Their
responses were taped, and later the time they took was
computed. During data analysis, errors of omission and
commission were disregarded, and only the naming time
was taken into account. Letter naming was used instead of
word naming to minimize the confounding effects of
decoding skills. All the tests were administered individu-
ally by graduate students working toward their master’s or
PhD degree in reading education. The results of these tests
are shown in Table 2.
To assess the relative contribution made by word attack,
listening comprehension, and letter naming speed to read-
ing comprehension, the data were analyzed using a multi-
ple regression procedure. The purpose of using multiple
regression analysis was to learn about the relationship
between the independent predictor variables and the
dependent variable. In this study, only three predictor vari-
ables were used, although many other variables belonging
to the psychological and ecological domains as well as the
instructional method used in the classroom could have
contributed to reading comprehension. The present study,
however, was designed to test the validity of the cognitive
components of the CMR and, therefore, only measures of
decoding, linguistic comprehension, and letter naming
speed were included.
The relative independence of comprehension and
decoding skills is expected to minimize the problem of
multicollinearity. Furthermore, visual inspection of the
Aaron et al. / Component Model of Reading 71
Table 1
Number of Children Participating
in Study 1 by Grade and Gender
Gender 2345
Boys 24 29 30 26
Girls 26 35 25 19
Total 50 54 55 45
by guest on June 2, 2013ldx.sagepub.comDownloaded from
data indicated that there were no outliers, thereby reducing
the risk of biased regression coefficients. The results of
the regression analysis are shown in Tables 3 and 4. Multiple
regression analysis showed that the variance accounted
for by products of listening comprehension and decoding
ranged from 37% to 41% and increased progressively
from Grade 2 through 5. The variance accounted for by
letter naming speed ranged from 11% to 2.5%, decreas-
ing progressively from Grade 2 through 5.
A review of studies that investigated the relationship
between IQ and reading achievement scores shows that
the correlation coefficient between these two variables
ranged from .30 to .54 (Stanovich, Cunningham, &
Feeman, 1984). This means that IQ can predict about
25% of the variability seen in reading achievement. The
present study shows that the two components of the CMR
(viz., listening comprehension and decoding) can predict
from 38% to 41% of the variability seen in reading
Apart from decoding and comprehension, speed of
processing has been claimed to be yet another indepen-
dent variable that affects reading performance (Wolf &
Bowers, 1999). Speed of processing cannot be ignored
because it is an important ingredient of information pro-
cessing. Studies that have examined the role of process-
ing speed with reference to the double-deficit hypothesis
have concluded that speed of processing does contribute
to the variability in reading comprehension, even though
its contribution is relatively small. On the other hand,
Adlof et al. (2005) noted that fluency need not be added
as a separate component to the CMR. This suggestion
72 Journal of Learning Disabilities
Table 2
Means and Standard Deviations of Reading-Related Test Scores by Grade
GMG Reading Comprehension 45.45 19.77 51.00 19.50 47.98 22.0 61.80 20.68
WLPB Listening Comprehension 100.51 11.39 107.61 18.59 110.23 16.38 113.48 16.22
WLPB Word Attack 104.61 17.13 104.91 17.18 109.36 16.53 107.70 18.16
Letter naming speed
33.67 9.35 32.39 8.61 25.70 5.0 23.88 8.4
Note: GMG = Gates-MacGinitie Test of Reading (MacGinitie & MacGinitie, 1989) normal curve equivalent score, M = 50, SD = 10; WLPB =
Woodcock Language Proficiency Battery (Woodcock, 1991) standard score, M = 100, SD = 15.
a. Time in seconds taken to name 40 letters.
Table 3
Multiple Regression Analysis: R
Values and Beta
Weights of Reading Components by Grade
Component 2
0.475 0.462 0.470 0.631
Listening comprehension
β 0.37 0.36 0.50 0.41
β 0.31 0.25 0.27 0.40
p .004 .046 .040 .001
Letter naming speed
β−0.25 0.31 0.28 0.16
p .094 .009 .012 .002
a. n = 50.
b. n = 54.
c. n = 55.
d. n = 45.
Table 4
Multiple Regression Analysis: Contributions of
Listening Comprehension, Decoding, and Letter
Naming Speed to Reading Comprehension by Grade
Listening Comprehension × Letter-Naming
Grade Decoding Speed
37 11
38 8
40 7
41 2.5
Note: All values expressed as percentages.
a. n = 50.
b. n = 54.
c. n = 55.
d. n = 45.
by guest on June 2, 2013ldx.sagepub.comDownloaded from
is based on a study in which a total of 522 children
completed many tests of reading. Statistical analysis of
the results indicated that in Grade 2, word recognition
and fluency were undifferentiated as independent con-
structs. In Grade 4, the analysis showed that word recog-
nition and listening comprehension accounted for 97% of
the variance, whereas fluency did not explain any of the
variance. In Grade 8, word recognition and listening
comprehension accounted for all of the variance. In a
study of German children, Hawelka and Wimmer (2005)
found that rapid naming did not add to the variance
already provided by multiletter word naming (i.e., instant
word reading). Based on their studies of Chinese-, Korean-,
and English-speaking children, Cho and McBride-Chang
(2005) concluded that “speed of processing is a basic
resource important across all tests of cognitive process-
ing. However, it does not appear to be linked specifically
to reading” (p. 13). Vukovic and Siegel (2006), who
reviewed the studies on this topic, concluded that “although
there are readers who meet the classification of the double-
deficit subtype, the existence of a naming speed–only
subtype of dyslexia has not been consistently documented”
(p. 44). This conclusion appears to be convincing, because
poor decoders invariably are also slow processors, indi-
cating that phonological processing skills and speed of
processing are intertwined with each other.
In addition to the uncertainty about the independent
status of reading speed, the present data show that there
is a trend for the effect of speed on reading perfor-
mance to decline as children move up in grade. The pre-
sent study shows that the contribution of speed of
processing to reading varies from 2.5% to 8% depend-
ing on grade, with a tendency to decline in higher
grades. By the time they reach Grade 5, children
become proficient in instantly identifying written
words, which also makes it difficult to isolate the effect
of fluency from that of word recognition. That is, indi-
viduals who have good word recognition skills (i.e.,
sight word readers) tend to be fluent readers and vice
versa. Several studies have shown that the ability to
identify words instantly (sight word reading) is not an
isolated skill, but is built on grapheme–phoneme con-
version skills (Aaron, Joshi, Ayotollah, et al., 1999;
Ehri, 1992). These studies have shown that poor
decoders are also dysfluent readers, and conversely,
slow readers are often deficient in decoding skill.
The results of the validation study, together with the
experimental, developmental, and neuropsychological stud-
ies previously mentioned, lend support to the validity of
the component model of reading by showing that word
recognition and linguistic comprehension are two major
but independent components of cognitive domains of read-
ing. Fluency, on the other hand, makes inconsistent contri-
butions to reading comprehension, accounting for a
negligible 2.5% of the variance at the fifth-grade level.
Study 2: CMR-Based Instruction Versus
Discrepancy-Based Instruction
In this study, the educational outcome of seven 1-year
programs of diagnosis and instruction of reading disabil-
ities based on the CMR is compared with the results of
the traditional LD diagnosis and instruction carried out in
resource rooms.
The simple view of reading continues to be influential
in reading research even though its wholesomeness has
been questioned by some (e.g., Duke et al., 2006). As
noted in the introduction, the theoretical validity of the
simple view of reading has been fairly well established
by experimental, neuropsychological, and developmen-
tal studies. However, its instructional utility has not been
fully explored. If its usefulness in remedial reading
instruction can be demonstrated, the CMR can offer
an alternative to the discrepancy model as well as to
the recently proposed response to intervention (RTI)
approach. Study 2 was designed to compare the effec-
tiveness of CMR-based instruction with the outcome of
the traditional LD program. With this goal in mind,
Study 2 was carried out with the intention of answering
the following four questions:
1. Is the word attack gain score of the children in the
treatment group who received word recognition train-
ing under the CMR significantly greater than the word
attack gain score of the children in the traditional LD
comparison group?
2. Is the word attack gain score of the children in
the treatment group who received comprehension
training under the CMR greater than the word attack
gain score of the children in the LD comparison
3. Is the comprehension gain score of children in the
treatment group who received word recognition train-
ing under the CMR greater than the comprehension
gain score of children in the LD comparison group?
4. Is the comprehension gain score of children in
the treatment group who received comprehension
training under the CMR greater than the comprehen-
sion gain score of the children in the LD comparison
Aaron et al. / Component Model of Reading 73
by guest on June 2, 2013ldx.sagepub.comDownloaded from
The reading achievement scores of a total of 330
children from Grades 2 through 5 were used for comparing
the relative effectiveness of instruction based on the CMR
with that of instruction based on the traditional LD model.
Of these 330 children, 171 received remedial reading
instruction based on the CMR (treatment group), and the
remaining 159 children received instruction in LD resource
rooms in their respective schools (comparison group).
Treatment Group
The 171 children in the treatment group were enrolled
in a remedial program named READ (Reading for
Excellence in Academic Development). The READ pro-
gram is intended for children from Grades 2 through 5 and
is conducted on the premises of the College of Education
and supervised by one of the authors of the present article.
During the month of July, starting from the year 1998, an
announcement was placed in the local newspaper inviting
parents who thought that their children were at risk for
reading problems to enroll their children in a remedial
reading program. Parents brought their children for an
hour of instruction after school hours, 4 days a week, for
one semester. The children were taught in small groups of
four to five by seven or eight graduate students who were
enrolled in the School Psychology program.
Children in the treatment group were from seven dif-
ferent cohorts taught during the course of 7 years. These
children came from a midsize Midwest town and were
mostly middle class. Most (90%) of these children were
European American, 8% were African American, and 2%
were of East European or Arab origin. Details of their dis-
tribution based on gender and grade level are shown in
Table 5. The children in the treatment group were matched
with children in the comparison group for grade level,
gender, and socioeconomic status (SES).
Comparison Group
Children in the comparison group were from three
different geographical regions of the United States:
Oklahoma, Illinois, or the state of Washington. The read-
ing scores of children with LD before and after resource
room instruction were obtained from special education
files kept in their respective districts. These files covered
a period of 6 years, from 1998 through 2004. In the LD
programs, posttests were administered 3 years after the
administration of the pretests. The pretest and posttest
word attack scores were available for 62 children, and
reading comprehension scores for 97 children, making
up a total of 159 LD comparison children. The word attack
and reading comprehension scores of the LD group
children had been obtained by the school psychologists in
their schools by administering the Woodcock-Johnson
Tests of Achievement (Woodcock, 1989; Woodcock,
McGrew, & Mather, 2001). One might expect that the
school records would contain both word attack and com-
prehension scores of all the 159 children in the compari-
son group. This, however, was not the case. The
psychologists who administered the tests in the schools
were looking only for a discrepancy between the reading
comprehension and the IQ scores, and once they saw a
significant discrepancy between IQ and reading compre-
hension, they terminated their assessment procedure and
did not administer any other reading-related tests. The
main objective of the school psychologist was to catego-
rize the at-risk reader as having LD or not having LD, not
to look for the source of the reading difficulty of the child.
The LD programs reported here did not classify children
on the basis of their weakness and did not target instruction
at the weakness. In this respect, the LD programs differed
from the CMR-based instruction. It was difficult to gather
detailed information about the nature and type of remedial
instruction that was provided to children in the comparison
group. The teachers could give only rough descriptions of
their instructional procedures. However, the available infor-
mation shows that all these children received instruction for
at least 1 hour per day during the school days. The method
of instruction varied from school to school. Some teachers
worked on improving word recognition skills by asking the
children to read the assigned material. When the children
failed to recognize a word or could not read the word cor-
rectly, the teacher supplied the word. Other teachers used the
time for helping the children with their classroom writing
assignment. The number of children with reading disabilities
in the resource room varied from three to seven.
74 Journal of Learning Disabilities
Table 5
Distribution of Children in Study 2 by Gender,
Grade, and Type of Instruction Received
Treatment Group
Comparison Group
Grade Boys Girls Boys Girls
2 60166510
3 351035 9
4 25 5 15 5
5 15 5 17 3
a. n = 171; 125 children received word recognition instruction; 46
children received comprehension instruction.
b. n = 159.
by guest on June 2, 2013ldx.sagepub.comDownloaded from
The children in the comparison group were matched,
grade by grade, with those in the treatment group on the
basis of their pretest scores on reading-related tests.
Outliers in both groups were eliminated by visual inspec-
tion. The ethnic composition and SES of this group of
children was comparable to those in the treatment group,
except that nearly 2% of the children in the comparison
group were Native American and Asian children. The
distribution of comparison children according to gender,
grade, and type of reading instruction received is shown
in Table 5.
Design and Procedure
Based on the nature of the reading component they
showed a deficit in (word recognition or comprehension),
125 children in the treatment group received word recogni-
tion training, and 46 children received reading comprehen-
sion strategy training. As for the comparison group, pretest
and posttest word attack scores were available for 62
children and reading comprehension scores for 97 children,
making up a total of 159 LD comparison children.
The results reported for the treatment group come from
remedial instruction based on the CMR carried out over a
period of 7 years, from 1998 through 2005, with seven dif-
ferent groups of children. The CMR-based diagnosis and
instruction were carried out every year by a set of new
graduate students who were in the School Psychology
program of the university. These graduate students had
taken courses in reading disability and received training in
remedial instruction before the beginning of the READ
program. The remedial instruction program started in the
month of August and ended by mid-December. Parents
brought their children to the College of Education build-
ing for 1 hour per day after school hours, 4 days per week.
The total number of children served during any 1 year
ranged from 20 to 28. As the number of graduate students
in the School Psychology program ranged from 7 to 10, it
was possible to carry out reading instruction in small
groups of 3 to 5 children per group.
On the 1st day, when the parents brought their
children, the READ program was explained to them.
They were told that the program would not categorize
their children as poor readers with LD and poor readers
without LD, but that by administering reading-related
tests, the weak area of the child (word recognition, vocab-
ulary, or comprehension) would be identified and reme-
dial instruction would target the weak area. The parents
were also told that at the end of the semester, children
would be given posttests, and information about the
progress (or lack thereof) of the children would be made
available to them.
After the pretests were given, the graduate students
and the supervisor (one of the authors) met and discussed
each case and determined a plan of action. On the basis
of the pretest scores, children were placed in one of two
groups—the word recognition training group or the com-
prehension training group (see the next section for more
details). After this, the supervisor and one or two gradu-
ate students met with the parents or guardian of each
child and explained the pretest profile of their child and
the plan of instruction. During this interview with the
parents, noncognitive reasons for the reading problem,
such as home environment and parental involvement in
the child’s reading, were also explored, and, when nec-
essary, appropriate recommendations were made. All
information, including test scores, was kept confidential.
During the course of the instructional period, the super-
visor moved from group to group and observed how
the instruction was proceeding. Once every 3 weeks, a
staffing meeting of the supervisor and graduate instruc-
tors was held, at which time the progress of each child
and the problems encountered were discussed.
In the past, behavioral problems were minimal because
instruction took place in groups of three, four, or five
children. When behavior problems cropped up, the
instructors, being school psychologists, used behavior
modification techniques to minimize the problems. On
occasions such as Halloween, Thanksgiving, and
Christmas season, pizza parties were given and children
had games to play.
Tests Administered
The following tests were administered as pretests, but
only selected tests from this list were given as posttests:
1. From the Woodcock Language Proficiency Battery
(WLPB; Woodcock, 1991)
Listening Comprehension
Passage (reading) Comprehension
Word Attack
Oral Vocabulary (synonyms and antonyms)
Reading Vocabulary (synonyms and antonyms)
2. From the Stanford Diagnostic Reading Test (SDRT;
Karlsen & Gardner, 1996)
Reading Comprehension
3. Test of Word Reading Efficiency (TOWRE; Torgesen,
Wagner, & Rashotte, 1999)
4. From the Comprehensive Test of Phonological
Proficiency (CTOPP; Wagner, Torgesen, & Rashotte,
Rapid color naming
Rapid letter naming
Aaron et al. / Component Model of Reading 75
by guest on June 2, 2013ldx.sagepub.comDownloaded from
Before standardized tests of word reading and phono-
logical proficiency (TOWRE and CTOPP) became avail-
able, a set of informal tests of letter naming, nonword
reading, word reading, and spelling, which were devel-
oped at the Educational Psychology laboratory, were
administered to the children.
Differential Diagnosis
The logic behind the differential identification of the
weak component and the assignment of the children to
different instructional groups is based on the close rela-
tionship between listening comprehension and reading
comprehension. Reading comprehension and listening
comprehension are highly correlated with each other, the
coefficient sometimes being as high as .78 (Palmer et al.,
1985), which led Palmer et al. (1985) to propose that
reading comprehension can be predicted almost per-
fectly by a listening measure. If a child’s listening com-
prehension is in the average range and higher, but his or
her reading comprehension is lower, then the reading dif-
ficulty is most likely due to a weakness in the word
recognition component. Invariably, the child will have
below-average word attack and spelling scores. Children
with a weak word recognition component receive
instruction in word recognition skills. In contrast, if the
child has below-average scores on tests of both reading
and listening comprehension, but has word attack scores
in the average range, then his or her impediment to read-
ing is the weak comprehension component. Such a child
will receive instruction in comprehension strategies.
Children who have deficits in both word recognition and
comprehension skills are started off with word recogni-
tion skills training. If and when they have attained suffi-
cient word recognition skills, they are moved to the
comprehension group.
Assignment to different instructional groups was based
on the children’s performance on tests of reading com-
prehension, listening comprehension, and word attack.
Furthermore, scores from the TOWRE, CTOPP, and infor-
mal nonword reading, spelling, and letter naming tests
were also used. Information obtained from these tests is
helpful when a child cannot be assigned unambiguously to
either of the instructional groups.
In the READ program, the following criteria are used for
diagnosis of a word recognition deficit: the child’s listening
comprehension score is in the average range (standard score
90–110), but his or her reading comprehension score is two
standard errors below this range (80 and below) on one of
the tests of reading comprehension (WLPB or SDRT). The
word attack scores of these children tend to be very similar
to their reading comprehension scores. That is, a child with
a word attack standard score of 80 is unlikely to have a read-
ing comprehension score much above 80. This diagnosis of
a weak word recognition component should be further con-
firmed by poor spelling and slow word reading, as assessed
by the TOWRE. The child is then assigned to the word
recognition instruction group.
Conversely, if a child’s word recognition score (from the
WLPB Word Attack test) is in the average range (standard
score 90–110), but his or her listening and reading com-
prehension scores are noticeably below the word attack
score (80 or below), then the deficit is in the comprehen-
sion component. A child with such a profile is assigned to
the comprehension instruction group.
There are also children who are weak in both word
recognition and comprehension skills. Theoretically speak-
ing, such children need instruction in both word recog-
nition and comprehension. However, experience with
children who have this profile has led to the conclusion
that word recognition skills should be brought up to the
functional level before the child can be expected to com-
prehend what he or she reads. If words cannot be recog-
nized and understood, the passage in which these words
appear also cannot be understood.
Often, children with below-average comprehension
skills also have limited vocabulary. In the READ pro-
gram, vocabulary training is provided as part of compre-
hension training, because words encountered in context
are more readily learned and retained than words learned
in isolation (Carlisle & Katz, 2005)
Occasionally, some children have reading comprehen-
sion scores that are higher than their listening comprehen-
sion scores. This is a profile that is opposite to the one seen
in children with reading disabilities. Research has shown
that children with this type of profile invariably have dif-
ficulty with sustained attention (Aaron, Joshi, Palmer,
Smith, & Kirby, 2002). This is because listening is more
attention demanding than reading. The READ program
does not have a provision to deal directly with attention-
related difficulties. Inconsistent attention is dealt with by
requiring children to summarize, in one or two sentences,
what they have read as material for strategy instruction.
This strategy seems to minimize the attention problem
of some children and has produced satisfactory results.
In general, inconsistent attention has not been a major
impediment to instruction, probably because of the small
size of the reading groups. Some children in the READ
group have also been taking medication for attention-
deficit/hyperactivity disorder (ADHD).
Children whose reading shows a weak word recogni-
tion component are also invariably slow in reading. These
children are described as lacking in fluency, and some
researchers have advocated that direct instruction is
76 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
necessary for improving the reading fluency of such
children. As discussed earlier, the existence of poor
decoders who are nevertheless fast readers has not been
demonstrated unequivocally (Vukovic & Siegel, 2006).
Furthermore, studies of fluency training have provided
mixed results, indicating that rereading improves the
speed of processing practiced passages but may not
transfer to new passages. The READ program, though
not denying the importance of fluency for reading, con-
siders fluency as a product and not a process and does
not provide special training that is aimed at improving
reading speed per se. Children are simply provided with
more opportunity to read. Furthermore, during word recog-
nition and comprehension strategy training, children scan
the passages and reread words and sentences with which
they have difficulty. It is expected that when word recog-
nition skill improves, a concomitant increase in speed
would be the outcome.
It is not uncommon that some children cannot be
assigned to any one of the two instructional groups with
certainty. Under these circumstances, the child is first
assigned to one of the two instructional groups and his or
her progress monitored. Based on the outcome of such
trial teaching, the child either continues in the same group
or is transferred to the other group.
Instructional Procedures
Word Recognition Training
As a starting point, children in the READ program who
were placed in the word recognition instruction group
received phonemic awareness training. During the initial
stages, the instructional procedure followed the steps
recommended in the Phoneme Sequencing Program for
Reading, Spelling, and Speech (Lindamood & Lindamood,
1998). Following the procedures recommended in this pro-
gram, the consonant sounds and vowel sounds were intro-
duced using the multisensory approach, followed by the
introduction of graphemes and simple words.
After the phonemic awareness training, The Writing
Road to Reading (Spalding & Spalding, 1990) was used
for further training in word recognition. Following the
instructions provided in this program, phonograms
(graphemes) such as /b/, /c/, and /ough/ were introduced
gradually, with the children copying the phonograms and
pronouncing them. There are 70 phonograms in this pro-
gram; about 4 phonograms were introduced in every ses-
sion. The instructor sounded out a phonogram first and
then introduced the word in which the phonogram
appeared. The children were asked to say the phonogram
and to copy the word down and then pronounce it. The
instructor drew the attention of the children to the word
and its meaning. The following day, the words that had
been learned the previous day were pronounced again by
the instructor, and the children were asked to spell the
words by writing them down. The rate of introducing
new phonograms depended on how fast the group had
learned the previous phonograms. When the phonograms
segment was completed, simple sentences using the pre-
viously learned words were constructed, and children
were asked to read them and copy them. Depending on
the progress each child had made, he or she was intro-
duced to simple decodable storybooks.
The READ program is based on the belief that all
children, regardless of their IQ, can make progress, albeit
to a small degree, and that all poor readers can become
better readers. As a result of this policy, we have had a
few children who had borderline cognitive abilities. In
this category, there were two children with Down syn-
drome and three others the etiology of whose low cogni-
tive ability was not known. These children experienced
difficulty even in learning the letters of the alphabet. The
instructional program was, therefore, modified to suit
their abilities. First, these children were taught to associ-
ate a few pictures with their names. Subsequently, they
were taught to associate written words with these pic-
tures “by sight.” Once the children had learned the names
of about 10 words by sight, they were taught the names
of the constituent letters in these words. The method fol-
lowed could be described as analytic phonics. This is in
contrast to the synthetic phonics approach used to teach
the rest of the children in the word recognition group. In
the present article, the data of the children who were taught
through the analytic phonics method were not used in
data analysis.
Comprehension Training
Children in the comprehension training group were
taught to use seven strategies that many researchers have
found to be useful in promoting reading comprehension
skills. The instructional method drew heavily from pro-
cedures recommended by researchers such as Brown,
Pressley, Van Meter, and Schuder (1996) and Palincsar
and Brown (1984). Passages within the reading level of
the children were used for comprehension training exer-
cises. Stories that were fewer than two pages long and
could be completed within the 45-min session were pre-
ferred. Such short stories were helpful in keeping the
children motivated, which served as a factor in maintain-
ing attention.
Aaron et al. / Component Model of Reading 77
by guest on June 2, 2013ldx.sagepub.comDownloaded from
The steps followed in comprehension strategy instruc-
tion were as follows:
1. Schema activation (“What do I already know about
2. Purpose of reading (“What am I reading this for?”)
3. Stop and think (“Am I understanding what I have
4. Visualize (“Are there maps and pictures that I should
look at?”)
5. Seek help (“If I do not know a word, I should raise my
6. Ask a question (“I should ask the teacher or the author
a question”)
7. Summarize (“Can I tell the story in one or two
These seven comprehension strategies were written on
blank sheets of paper, and the children were told that they
should memorize them and use them as they read.
First, the instructor modeled the process of reading
using all the seven strategies.
Strategy 1: Schema activation. Each child in the group
was provided with a copy of the story to be read that day.
Children were then asked to scan the story, look at the
pictures, and highlight the words they did not know. Each
child then highlighted or pointed to the words he or she
did not know, and the instructor pronounced the words
and explained the meaning of these words by embedding
them in simple sentences. This initial activity was used as
a means of vocabulary development. Following this, the
instructor asked the children a few questions and made
some remarks that enabled activation of the appropriate
Strategy 2: Purpose of reading. The reader, who was
one of the children in the group, was instructed to be
aware of the reason for reading the story. Was it about
knowing something, for answering questions, or was it
for enjoying the story?
Strategy 3: Stop and think. As the reader progressed
along, he or she was asked to occasionally stop and think
if he or she was understanding what he or she had been
Strategy 4: Visualization. The reader was encouraged
to look at the pictures or the maps in the passage and
try to relate them to the text at several points as he or
she read.
Strategy 5: Seek help. If the child did not know how to
pronounce a word or did not know what the word meant,
he or she was asked to raise a hand and ask the instruc-
tor for help.
Strategy 6: Ask a question. If the reader had a ques-
tion, what would he or she ask the author of the story, or
the instructor?
Strategy 7: Summarize. The reader was reminded at the
beginning, and again periodically, that he or she should
be able to tell what he or she had read in one or two sen-
tences. The child had to recall the main idea from mem-
ory without looking at the passage or the story. Stressing
the summarization requirement is important because it
helps to keep the reader’s attention from wandering.
During strategy instruction, one child in the group was
the focus of instruction, and other children were encour-
aged to be actively involved in making comments and
seeking help. For instance, any child in the group could
remind the target child to look at the map or pictures in
addition to asking the meaning of a word in the text. A
fact that was learned over the years is that children can
learn to recite the seven comprehension strategies but fail
to incorporate them when they read a story or a passage.
In the READ program, modeling by the instructor has
been found to be the most effective way of overcoming
this difficulty. This requires the instructor to model the
use of the seven strategies before every new lesson is
In the READ program, there was no special vocabulary
instruction group. Vocabulary instruction was introduced
through morphemic patterns, prefixes, suffixes, and root
words that were used in the comprehension group by ask-
ing the children to look at the story and highlight the
words they did not know. The instructor pronounced these
words and gave their meaning with ample illustrations.
Thus, vocabulary instruction was provided in context,
with an emphasis on the meaning of the words.
Throughout the 7-year period, the basic format of
training remained unchanged. However, in the light of
experience gained every year, small changes in instruc-
tion were made. For instance, we started the 1st year with
the instruction of 10 comprehension strategies. Soon we
found that was too many and reduced it to the present
7 strategies. Similarly, during the first 2 years, we spent
almost 8 weeks with the Phoneme Sequencing program,
but we found that this did not leave much time for read-
ing sentences. As a result, we reduced it to 4 weeks and
moved on to The Writing Road to Reading immediately
after that. We also allotted more time to reading decod-
able texts after the first 2 years.
78 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
Comparison Group Instruction
Whereas children in the treatment group received dif-
ferential training according to the nature of their deficit,
such instruction had not been provided in the resource
rooms. The 15 teachers of the children in the LD com-
parison group used a variety of procedures to teach the
children. Information obtained through informal inter-
views of the teachers indicates that most of the instruc-
tional time in the resource room was used for completing
homework assignments. This was so because many of
the children with LD were lagging behind in their assign-
ments. There is no evidence that instruction was individ-
ualized or that specific instruction focusing on phonology
or comprehension was used by most of the teachers.
Three teachers reported that they provided phonics train-
ing, but it was not clear if the children they taught had
weak decoding skills or what specific procedure was fol-
lowed. In general, teachers were reluctant to discuss their
instructional strategy but reported that they provided
individual help as needed.
Depending on their weak reading component, children
in the treatment group received either word recognition
training or comprehension training. Children in the com-
parison group did not receive such differential instruction
in their LD resource rooms. It may be recalled that 125
children from the treatment group received word recogni-
tion training, and 46 children from the same group received
comprehension strategy training. Of the 159 children in the
comparison group, word recognition scores were available
for 62 children and comprehension scores for 97 children.
Pre- and posttest scores of the children in the treatment
group and the comparison group are shown in Table 6.
Reading scores gained or lost by the two groups are shown
in Figure 2.
Four questions were raised to evaluate the relative
outcomes of reading instruction based on the CMR and
instruction based on the LD discrepancy model. Repeated
measures analysis of variance was used to answer these
1. Is the word attack gain score of the children in the
treatment group who received word recognition training
under the CMR significantly greater than the word attack
gain score of the children in the traditional LD compar-
ison group who did not receive differentiated instruc-
tion? Of a total of 171 children in the treatment group,
Aaron et al. / Component Model of Reading 79
Figure 2
Gains in Word Attack and Comprehension Scores
by Treatment and Comparison Groups
Table 6
Pre- and Posttest Word Attack and Comprehension Scores by Instructional Group
Treatment Group Comparison Group
Pretest Posttest Pretest Posttest
Measure n M SD M SD n M SD M SD
WLPB Word Attack 62 86.19 12.55 87.08 11.08
Word recognition training 125 84.66 9.96 90.05 11.41
Comprehension training 46 98.55 12.08 98.74 12.81
WLPB Comprehension 97 86.67 14.12 84.90 12.97
Word recognition training
125 88.14 12.40 91.79 12.48
Comprehension training
46 88.50 9.47 93.54 10.61
Note: WLPB = Woodcock Language Proficiency Battery (Woodcock, 1991).
a. The mean listening comprehension score of this group of children was 102.3. Their reading comprehension score was much lower because
of their poor word recognition skills.
b. The mean pretest coprehension score of these children on the Stanford Diagnostic Reading Test (SDRT; Karlsen & Gardner, 1996) was 88.21
(SD = 10.11). They did not complete SDRT posttests.
Note: LD = learning disabilities.
by guest on June 2, 2013ldx.sagepub.comDownloaded from
125 received training in word recognition skills. Of the
total of 159 children in the comparison group, word attack
scores were available for 62 children. The repeated mea-
sures statistical analysis showed that the word attack
gain scores of the treatment group who received word
recognition training were significantly higher than the
word recognition gain scores of the comparison group,
F(1, 185) = 7.793, p < .006, η
= 0.4. This finding is
interpreted to mean that word recognition training pro-
vided to children with a deficit in the word recognition
component is more effective than the undifferentiated
instruction provided to children in resource rooms.
2. Is the word attack gain score of the children in the
treatment group who received comprehension training
under the CMR greater than the word attack gain score of
the children in the traditional LD comparison group who
did not receive differentiated instruction? Of a total of 171
children in the treatment group, 46 received comprehen-
sion strategy instruction. Of the 159 children in the com-
parison group, word attack scores were available for 62
children. Statistical analysis of data showed no significant
differences between the comprehension gain scores of the
two groups. That is, both groups did not show any signif-
icant improvement in decoding skills, and, therefore, there
was no significant difference between the gain scores of
the two groups, F(1, 106) = 0.147, p < .7, η
= 0.01. This
means that the comprehension training provided to
children who were not identified as having a word attack
deficit in the READ program did not improve their word
attack scores. Children in LD resource rooms likewise did
not significantly improve their word attack scores.
3. Is the comprehension gain score of children in the
treatment group who received word recognition training
under the CMR greater than the comprehension gain
score of children in the traditional LD comparison group
who did not receive differentiated instruction? There
were 125 children in the treatment group who received
word recognition training. Their reading comprehension
scores were compared with the reading comprehension
scores of the 97 LD comparison children for whom these
scores were available. Statistical analysis showed that the
treatment group who received the word recognition train-
ing made significant gains in their comprehension scores,
which were greater than the gains in comprehension attained
by the comparison group, F(1, 220) = 13.05, p < .001,
= 0.56. Assignment to the comprehension training group
was based on their performance on two tests of comprehen-
sion, the WLPB and the SDRT. The fact that the children in
the READ group who received training in word recognition
skills showed significant gains in comprehension could be
seen as an unexpected outcome. However, the compre-
hension gains of children who received word recognition
training are not surprising because the reading compre-
hension of these children was held down by poor word
recognition skills. In other words, poor word recognition
skills might have functioned as a factor that limited the
reading comprehension of these children. It has to be
remembered that these children were placed in the word
recognition training group because their listening com-
prehension was in the typical range but their word attack
skill was well below average. Once the constraints of
poor word recognition skill were removed, a concomitant
improvement in reading comprehension occurred.
4. Is the comprehension gain score of children in the
treatment group who received comprehension training
under the CMR greater than the comprehension gain score
of the children in the traditional LD comparison group who
did not receive differentiated instruction? There were 46
children in the treatment group who received comprehen-
sion strategy instruction. Reading comprehension scores
were available for 97 children in the LD comparison group.
Statistical analysis showed that the treatment group who
received the comprehension strategy training made gains in
their comprehension scores that were significantly higher
than the comprehension “gain” scores of the comparison
group, F(1, 141) = 3.855, p < .05, η
= 0.26. Actually, the
children in the LD resource rooms lost 1.77 comprehen-
sion standard scores, the difference between their pretest
and posttest scores being marginally significant, t(96) =
1.43, p = .07, an instance of the Matthew effect. The data
suggest that the reading instruction provided in the READ
program, which focused on the weak comprehension com-
ponent, improved comprehension skills, whereas the LD
resource room instruction did not.
The outcomes of the present study can be summarized
as follows:
1. Word recognition training provided to children with a
deficit in that component is more effective than the undif-
ferentiated resource room instruction provided to
children with LD.
2. Children in the READ group who received training in
word recognition skills showed significant gains in com-
prehension. This is so because poor word recognition skill
usually functions as a factor that limits reading compre-
hension. However, children in the READ group who
received comprehension training did not improve in their
word recognition skills.
80 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
3. The data also suggest that reading instruction pro-
vided in the READ program that focused on the weak
comprehension component improved comprehension
On the whole, the results indicate that instruction
provided under the framework of the CMR is more effec-
tive than undifferentiated resource room instruction. As
noted earlier, there could be several reasons for the fail-
ure of the LD comparison children to improve, but the
data show that the CMR-based diagnosis and treatment
of children increased the word recognition and compre-
hension skills of children in the treatment group. Several
studies have shown that phonological awareness instruc-
tion improves children’s reading skills. Similarly, many
studies also have reported success in improving children’s
comprehension when these children are provided with
comprehension strategy instruction. As far as the present
authors know, differential diagnosis and instruction in
the format described in this article has not been previ-
ously reported.
General Discussion
Study 1 was designed to assess the relative contribution
made by word recognition and comprehension skills to the
reading performance of children in Grades 2 through 5.
Because fluency has been receiving much attention recently,
its contribution to reading was also examined. Multiple
regression analysis of the reading-related scores of 204
children showed that listening comprehension and
decoding can account for between 38% and 41% of read-
ing comprehension. Speed of processing accounted for
11% of the variance at the second grade and 2.5% at the
fifth grade. This steady decline of the contribution made
by speed of processing may be due to the fact that by the
time children reach Grade 5, particularly in the case of
typical readers, as in the present case, they have mastered
word recognition skills. This indicates that fluency—or
the lack of it—may be a by-product of word recognition
skill. Repeated reading has been shown to improve read-
ing performance, maybe because repeated reading
improves word recognition skills and thereby increases
fluency. As noted earlier, according to Vukovic and
Siegel (2006), speed deficits are unlikely to occur in iso-
lation in individuals with reading impairments and,
therefore, slow naming is unlikely to be a core deficit in
reading disabilities. This is not to say that striving for
fluency is not productive, but that fluency should be treated
as a symptom and not as a cause. This can be accom-
plished through guided repeated reading (National Reading
Panel, 2006) using a variety of reading materials, which
affords an opportunity for repeated exposure to different
words and also teaches students to read with expression.
Study 2 showed that when the weak reading compo-
nent that is responsible for poor reading performance is
identified, and instruction is targeted at that weak com-
ponent, the word recognition and reading comprehension
scores of poor readers improve significantly more than
those of children with LD who are provided with instruc-
tion in resource rooms. Thus, the present report not only
validates the component nature of the reading process
but also shows that instruction based on the component
model of reading could be used with greater success for
teaching poor readers. That is, to be successful, remedial
instruction should focus on the source of the reading
problem and address it. In addition to being direct and
effective, this CMR-based procedure also eliminates the
need for the administration of numerous tests and ties
assessment to intervention. It also obviates the need for
labeling poor readers into categories. The results of the
present study further confirm the limited utilitarian value
of the discrepancy model in dealing with the reading
problems of children.
The observed limitations of the LD practice have led
researchers and educators to look for better methods of
teaching students who experience difficulty in learning
to read and write. One such approach that has received
much attention in recent years is RTI (Bradley et al.,
2005; D. Fuchs et al., 2003; L. S. Fuchs, Fuchs, &
Speece, 2002). RTI is primarily an alternate process for
the identification of and intervention in LD. It is embed-
ded in a multitiered model of assessment, intervention,
and progress monitoring (Kovaleski, 2004). RTI comes
in more than one model, but they all have more than one
tier or phase. During the initial phase, a determination is
made whether effective instruction is in place in the
classroom. During the next phase, intense instruction is
provided to at-risk students, and its effect is measured.
During the next phase, individualized instruction is pro-
vided for those students whose progress is below aver-
age. During the final phase, the students who still failed
to make progress are further tested, or their previous
history is reviewed. Based on the outcome of this eval-
uation, the students may be referred to special education
Although the concept of RTI was introduced in the
opening section of this article, the present study did not
make a direct comparison between the CMR and RTI. As
an alternative to the discrepancy model, RTI is believed
to be promising and is being tried out in several school
systems. Some unanswered questions, however, remain.
Aaron et al. / Component Model of Reading 81
by guest on June 2, 2013ldx.sagepub.comDownloaded from
One of them is, How is RTI different from what teachers
usually do with their poor readers in the general educa-
tion classroom? According to Gerber (2005), RTI is
simply urging the educational system to try harder and to
invest more effort in students who are difficult to teach
and manage. According to Kavale (2005), RTI simply reop-
erationalizes LD instead of redefining it. Furthermore,
there are several variants or approaches to RTI (D. Fuchs
& Fuchs, 2006), indicating that a consensus about the
method of implementation of RTI has not yet emerged.
A review of the titles of published reports also indicates
that the focus of RTI is still on the identification of LD and
not on the specific method of instruction. If RTI is imple-
mented as presented in the literature, it still relegates the
unresponsive children to special education without recom-
mending what should be done with them. In this respect,
RTI differs from the CMR-based remedial instruction.
Limitations of the Present Study
One of the obvious limitations of the present study is
that children in the LD programs were tested after 3
years of instruction in the resource rooms, whereas the
children instructed using the CMR were tested after one
semester. The interval between pre- and posttests may be
a confounding factor. Nevertheless, if long-term instruc-
tion bestows any advantage, it should favor the children
in the LD program.
Another limitation is that the resource room instruc-
tion was not uniform in the three regions from where
data were collected. Generally speaking, the lack of uni-
formity of assessment and instruction is also a weakness
of LD programs. On the other hand, the instruction given
to the children taught under the CMR was fairly uniform,
though short. Attempts to replicate the present study
should, therefore, provide CMR-based instruction for a
longer period.
The educational strategy described under the CMR has
several advantages over the discrepancy model. The
assessment procedures used by the CMR are simple and
straightforward, and the instructional strategies that
emanate from them are uniform, logical, and effective.
Implementation of a CMR-based educational policy does
not require IQ testing—a procedure that typically con-
sumes a good part of the psychologist’s time. Most of the
tests used by the CMR can be administered by classroom
teachers or trained teacher’s aides. Most important, a diag-
nosis based on the CMR tells the teacher what component
of reading should be the focus of instruction and how that
instruction is to be delivered. Furthermore, the present
study indicates that children who experience reading diffi-
culties can be made to become better readers regardless of
their current level of reading skills. Finally, the component
model of reading leaves no child behind uninstructed.
Aaron, P. G. (1997a). A component-based approach to the diagnosis and
treatment of reading disabilities. In B. Ericson & J. Ronnberg (Eds.),
Reading disability and its treatment (pp. 3766). Norrköping,
Sweden: Läspedagogiska Institutet, Linköping University.
Aaron, P. G. (1997b). The impending demise of the discrepancy for-
mula. Review of Educational Research, 67, 461502.
Aaron, P. G., Franz, S., & Manges, A. (1990). Dissociation between
pronunciation and comprehension in reading disabilities. Reading
and Writing: An Interdisciplinary Journal, 3,122.
Aaron, P. G., Joshi, R. M., Ayotollah, M., Ellsberry, A., Henderson,
J., & Lindsey, K. (1999). Decoding and sight-word naming: Are
they independent components of word recognition skill? Reading
and Writing: An Interdisciplinary Journal, 11,89127.
Aaron, P. G., Joshi, R. M., Palmer, H., Smith, N., & Kirby, E. (2002).
Separating genuine cases of reading disability from reading
deficits caused by predominantly inattentive ADHD behavior.
Journal of Learning Disabilities, 35, 425447.
Aaron, P. G., Joshi, R. M., & Williams, K. A. (1999). Not all reading dis-
abilities are alike. Journal of Learning Disabilities, 32, 120127.
Aaron, P. G., & Kotva, H. (1999). Component model–based remedial
treatment of reading disabilities. In I. Lundberg, E. Tonnessen, &
I. Austad (Eds.), Dyslexia: Advances in theory and practice
(pp. 221244). Boston: Kluwer.
Adlof, S. M., Catts, H. W., Hogan, T. P., & Little, D. T. (2005, June).
The role of fluency in reading comprehension: Should fluency be
included in the simple view? Poster presented at the annual meet-
ing of the Society for the Scientific Study of Reading, Toronto,
ON, Canada.
Bentum, K., & Aaron, P. G. (2003). Does reading instruction in learn-
ing disability resource rooms really work? A longitudinal study.
Reading Psychology, 24, 361382.
Berninger, V. W., Dunn, A., Lin, S. C., & Shimada, S. (2004). School
evolution: Scientist–practitioner educators creating optimal learn-
ing environments for all students. Journal of Learning
Disabilities, 37, 500508.
Bradley, R., Danielson, L., & Doolittle, J. (2005). Response to inter-
vention. Journal of Learning Disabilities, 38, 485–486.
Brown, R., Pressley, M., Van Meter, P., & Schuder, T. (1996). A quasi-
experimental validation of transactional strategies instruction with
low-achieving second grader readers. Journal of Educational
Psychology, 88,1837.
Carlisle, J. F., & Katz, L. A. (2005). Word learning and vocabulary
instruction. In J. R. Birsh (Ed.), Multisensory teaching of basic
language skills (pp. 345375). Baltimore: Brookes.
Carlson, E. (1997). In-school and post-school outcomes of students
declassified from special education. Unpublished doctoral disser-
tation, College of William and Mary, Williamsburg, VA.
Carver, R. P. (1998). Predicting reading level in Grades 1 to 6 from
listening level and decoding level: Testing theory relevant to
simple vie
w of reading. Reading and Writing: An Interdisciplinary
Journal, 10, 121154.
82 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
Catts, H. W., Hogan, T. P., & Fey, M. E. (2003). Subgrouping poor
readers on the basis of individual differences in reading-related
abilities. Journal of Learning Disabilities, 36, 151165.
Catts, H. W., & Kamhi, A. G. (Eds.). (1999). Language and reading
disabilities. Needham Heights, MA: Allyn & Bacon.
Cho, J. R., & McBride-Chang, C. (2005). Correlates of Korean
Hangul acquisition among kindergartners and second graders.
Scientific Studies of Reading, 9,316.
Coltheart, M., Patterson, K. A., & Marshall, J. C. (1980). Deep
dyslexia. London: Routledge & Kegan Paul.
Crain, S. (1989). Why poor readers misunderstand spoken sentences.
In D. Shankweiler & I. Y. Liberman (Eds.), Phonology and reading
disability (pp. 3042). Ann Arbor: University of Michigan Press.
de Jong, P. F., & van der Leij, A. (2002). Effects of phonological abil-
ities and linguistic comprehension on the development of reading.
Scientific Studies of Reading, 6,5177.
Dudley-Marling, C. (2004). The social construction of learning dis-
abilities. Journal of Learning Disabilities, 37, 482–489.
Duke, N., Pressley, M., Fingeret, L., Golos, D., Halladay, J., Hilden, K.,
et al. (2006, July). Revisiting the simple view of reading. Paper
presented at the annual meeting of the Society for the Scientific
Study of Reading, Vancouver, BC, Canada.
Ehri, L. C. (1992). Reconceptualizing the development of sight word
reading and its relationship to recoding. In P. B. Gough, L. C.
Ehri, & R. Treiman (Eds.), Reading acquisition (pp. 107143).
Hillsdale, NJ: Lawrence Erlbaum.
Frith, U., & Snowling, M. (1983) . Reading for meaning and reading
for sound in autistic and dyslexic children. British Journal of
Developmental Psychology, 1, 320342.
Fuchs, D., & Fuchs, L. (1995). What’s so “special” about special edu-
cation? Phi Delta Kappan, 76, 522530.
Fuchs, D., & Fuchs, L. S. (2006). Introduction to response to inter-
vention: What, why, and how valid is it? Reading Research
Quarterly, 41,9399.
Fuchs, D., Mock, D., Morgan, P. L., & Young, C. L. (2003).
Responsiveness to intervention: Definitions, evidence, and impli-
cations for the learning disabilities construct. Learning
Disabilities Research and Practice, 18, 151171.
Fuchs, L. S., Fuchs, D., & Speece, D. L. (2002). Treatment validity as
a unifying construct for identifying learning disabilities. Learning
Disability Quarterly, 25,3345.
Gerber, M. M. (2005). Teachers are still the test: Limitations of
response to instruction strategies for identifying children with
learning disabilities. J
ournal of Learning Disabilities, 38, 485
486, 516524.
Gough, P., & Tunmer, W. (1986). Decoding, reading, and reading dis-
ability. Remedial and Special Education, 7(2), 610.
Hannon, B., & Daneman, M. (2001). A new tool for measuring and
understanding individual differences in reading comprehension.
Journal of Educational Psychology, 93, 103128.
Hawelka, S., & Wimmer, H. (2005). Impaired visual processing of
multi-element arrays is associated with increased number of eye
movements in dyslexic reading. Vision Research, 45, 855863.
Haynes, M. C., & Jenkins, J. R. (1986). Reading instruction in spe-
cial education resource rooms. American Educational Research
Journal, 23, 161190.
Healy, J. (1982). The enigma of hyperlexia. Reading Research
Quarterly, 17, 319338.
Hinshelwood, J. (1895). Word blindness and visual memory. Lancet,
21(1), 15061508.
Hoover, W., & Gough, P. (1990). The simple view of reading.
Reading and Writing: An Interdisciplinary Journal, 2, 127160.
Jackson, M. D., & McClelland, J. L. (1979). Visual factors in word
perception. Perception and Psychophysics, 14, 365370.
Joshi, R. M., & Aaron, P. G. (2000). The component model of read-
ing: Simple view of reading made a little more complex. Reading
Psychology, 21,8597.
Joshi, R. M., Williams, K., & Wood, J. (1998). Predicting reading
comprehension from listening comprehension: Is this the answer
to the IQ debate? In C. Hulme & R. M. Joshi (Eds.), Reading and
spelling: Development and disorders (pp. 319327). Mahwah, NJ:
Lawrence Erlbaum.
Karlsen, B., & Gardner, E. F. (1996). Stanford Diagnostic Reading
Test (4th ed.). San Antonio, TX: Harcourt Brace.
Kavale, K. A. (2005). Identifying specific learning disability: Is
responsiveness to intervention the answer? Journal of Learning
Disabilities, 38, 553562.
Kovaleski, J. F. (2004). Response to instruction, the identification of
learning disabilities:
A guide to school teams. NASP
Communique, 32(5), 25.
Kintsch, W., & Kozminsky, E. (1977). Summarizing stories and reading
and listening. Journal of Educational Psychology, 69, 491499.
Kirk, S. A. (1963, April). Behavioral diagnosis and remediation of
learning disabilities. Paper presented at the conference “Exploration
Into the Problems of the Perceptually Handicapped Child,” Chicago.
Lindamood, P., & Lindamood, P. (1998). Phoneme Sequencing
Program for Reading, Spelling, and Speech. Austin, TX: PRO-ED.
Lyon, R., Fletcher, J., Shaywitz, S., Shaywitz, B., Torgesen, J., Wood,
F. B., et al. (2001). Rethinking learning disabilities. In C. E. Finn,
A. J. Rotherham, & C. R. Hokanson (Eds.), Rethinking special
education for a new century (pp. 259287). Washington, DC:
Fordham Foundation.
MacGinitie, W. H., & MacGinitie, R. K. (1989). Gates-MacGinitie
Reading Tests. Chicago: Riverside.
McDermott, K. B., Watson, J. M., & Ojemann, J. G. (2006).
Presurgical language mapping. Current Trends in Psychological
Science, 14, 291294.
Mirak, L. J., Scarborough, H. S., & Rescorla, L. (2003). Late-emerg-
ing disabilities. Journal of Educational Psychology, 95, 211224.
Moody, S. W., Vaughn, S., Hughes, M. T., & Fischer, M. (1998).
Broken promises: Reading instruction in resource rooms.
Learning Disability Quarterly, 64, 211215.
Morgan, W. P. (1896). A case of congenital word blindness. British
Medical Journal, 2, 1368.
National Center for Education Statistics. (1998). Report card for the
nation and the states. Washington, DC: U.S. Department of Education.
National Reading Panel. (2000). Teaching children to read (NIH Pub.
No. 00-4754). Washington, DC: National Institute of Health and
Human Performance. Retrieved from
Orton, S. T. (1937). Reading, writing, and speech problems in children.
New York: Norton.
Palincsar, D., & Brown, A. (1984). Reciprocal teaching of compre-
hension fostering and comprehension monitoring activities.
Cognition and Instruction, 1, 117175.
Palmer, J., McCleod, C., Hunt, E., & Davidson, J. (1985). Information
processing correlates of reading. Journal of Memory and
Language, 24,5988.
Patterson, K. E., Marshall, J. C., & Coltheart, M. (1985). Surface
dyslexia. London: Lawrence Erlbaum.
Aaron et al. / Component Model of Reading 83
by guest on June 2, 2013ldx.sagepub.comDownloaded from
Perfetti, C. A. (1988). Verbal efficiency theory in reading ability. In
M. Daneman, G. E. MacKinnnon, & T. G. Waller (Eds.), Reading
research: Advances in theory and practice (pp. 109143). New
York: Academic Press.
Poldrack, R. A., & Wagner, A. D. (2004). What can neuroimaging tell
us about the mind? Current Directions in Psychological Science,
13, 177181.
Pressley, M. (2000). What should comprehension instruction be the
instruction of? In M. L. Kamil, P. B. Mosenthal, D. Pearson, &
R. Barr (Eds.), Handbook of reading research (Vol. 3, pp. 545
561). Mahwah, NJ: Lawrence Erlbaum.
Shankweiler, D., Crain, S., Katz, L., Fowler, A. E., Liberman, A.,
Brady, S., et al. (1995). Cognitive profiles of reading-disabled
children: Comparison of language skills in phonology, morphol-
ogy, and syntax. Psychological Science, 6, 149156.
Spalding, R. B., & Spalding, W. T. (1990). The writing road to read-
ing. New York: Morrow.
Stanovich, K. E., Cunningham, A. E., & Feeman, D. J. (1984).
Intelligence, cognitive skills, and early reading progress. Reading
Research Quarterly, 19, 278303.
Sternberg, R. J. (1985). Beyond IQ: A triarchic theory of human intel-
ligence. New York: Cambridge University Press.
Stothard, S. (1994). The nature and treatment of reading comprehen-
sion difficulties. In C. Hulme & M. Snowling (Eds.), Reading
development and dyslexia (pp. 85102). London: Whurr.
Swanson, H. L. (1999). Instructional components that predict treat-
ment outcomes for students with learning disabilities: Support for
a combined strategy and direct instructional model. Learning
Disabilities Research and Practice, 14, 129140.
Swanson, H. L., Howard, C. B., & Saez, L. (2006). Do different com-
ponents of working memory underlie different subgroups of read-
ing disabilities? Journal of Learning Disabilities, 39, 252269.
Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1999). Test of Word
Reading Efficiency. Austin, TX: PRO-ED.
Vaughn, S., Levy, S., Coleman, M., & Bos, C. (2002). Reading
instruction for students with LD and EBD. Journal of Special
Education, 36,213.
Vukovic, R. K., & Siegel, L. S. (2006). The double-deficit hypothe-
sis: A comprehensive analysis of evidence. Journal of Learning
Disabilities, 39,2547.
Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (1999). Comprehensive
Test Of Phonological Proficiency. Austin, TX: PRO-ED.
Wleklenski, M. H. (1993). Effectiveness of special education pro-
grams on the reading achievement of students with learning dis-
abilities. Unpublished doctoral dissertation, Indiana State
University, Terre Haute.
olf, M., & Bowers, P. G. (1999). The double-deficit hypothesis
for developmental dyslexia. Journal of Educational Psychology,
91, 415438.
Woodcock, R. W. (1989). Woodcock-Johnson Psychoeducational
Battery. Allen, TX: DLM.
Woodcock, R. W. (1991). Woodcock Language Proficiency
Battery–Revised. Chicago: Riverside.
Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-
Johnson Tests Of Achievement (3rd ed.). Itasca, IL: Riverside.
Yuill, N., & Oakhill, J. (1991). Children’s problems in text compre-
hension: An experimental investigation. New York: Cambridge
University Press.
P. G. Aaron, PhD, is the Coffman Distinguished Professor in the
Department of Educational Psychology at Indiana State University.
His areas of research interest are reading disability and inconsistent
attention in children. His publications focus on differential diagnosis
and treatment of these learning-related problems.
R. Malatesha Joshi, PhD, is a professor of literacy education at Texas
A&M University, where he teaches courses and conducts research in
the areas of assessment and intervention of reading problems.
Regina Gooden, PhD, is director of reading at St. Thomas University
in Houston, Texas. Her current interests are identifying early reading
failure, oral language comprehension, and teaching reading to
English language–learning children.
Kwesi E. Bentum, PhD, is a school psychologist in the Bi-County
Special Education Cooperative in Sterling, Illinois. His area of pro-
fessional and research interests are intervention and intervention-
related diagnosis of reading and behavioral problems in children.
84 Journal of Learning Disabilities
by guest on June 2, 2013ldx.sagepub.comDownloaded from
... Students demonstrate reading profiles based on individual differences in these reading components (Aaron et al., 2008;Catts et al., 2003;Hoover & Gough, 1990). SVR research suggests readers can be classified into four groups related to the strength (i.e., poor vs. good) of their linguistic comprehension and decoding abilities (Spencer et al., 2019). ...
... These factors may be important to consider when constructing student profiles if reading difficulties in each subgroup relate to different underlying mechanisms. The Component Model of Reading (Aaron et al., 2008) and the Direct and Indirect model of reading (DIER; Kim, 2020) contextualize the SVR within a framework that considers additional components (e.g., cognitive, psychological, and ecological) that influence heterogeneity in reading achievement and the underlying mechanisms of RD. Mindset is one factor related to these additional components that may influence the variation in students' reading profiles. ...
Full-text available
Recent research has focused on evaluating the relation between mindset and reading achievement. We used exploratory factor mixture models (E-FMMs) to examine the heterogeneity in reading achievement and mindset of 650 fourth graders with reading difficulties. To build E-FMMs, we conducted confirmatory factor analyses to examine the factor structure of scores of (a) mindset, (b) reading, and (c) mindset/reading combined. Our results indicated (a) a 2-factor model for mindset (General Mindset vs. Reading Mindset), (b) a 2-factor model for reading (Word Reading vs. Comprehension; four covariances), and (c) a combined model with significant correlations across mindset and reading factors. We ran E-FMMs on the combined model. Overall, we found three classes of students. We situate these results within the existing literature and discuss implications for practice and research.
... In this study, the investigation of the interplay between reading anxiety and working memory is grounded on three theoretical frameworks. The Component Model of Reading emphasizes both affective and cognitive factors play an important role in reading ability (Aaron et al., 2008). Investigating the contributions of affective and cognitive factors simultaneously provides a more thorough picture of reading comprehension development. ...
This study investigated the relationships between affective and cognitive factors and reading comprehension in English as a second language (ESL). Specifically, we evaluated the contributions of reading anxiety and verbal working memory to ESL reading comprehension in Chinese students. A total of 105 Chinese ESL undergraduates were included. Structural equation modeling results showed that reading anxiety, represented by reading trait and state anxiety, and verbal working memory were unique predictors of ESL reading comprehension. In addition, there was no significant reading anxiety × working memory interaction effect. Mediation analyses revealed that reading anxiety partially mediated the relationship between verbal working memory and ESL reading comprehension. These results highlight the importance of affective and cognitive factors in predicting ESL reading comprehension and shed light on the methods in enhancing ESL learning.
... In the context of the Simple View of Reading, non-proficient readers may show deficiencies in word identification, language processing, or both skill areas (Aaron et al., 2008). In a large-scale urban study, Hock et al. (2009) reported that 61% of at-risk students demonstrated low performance on tasks assessing both word-level and language processing skills. ...
Full-text available
Unfortunately, far too many American adolescents are unable to read proficiently. The science of reading suggests explicit instruction in both word identification and language processing skills should bolster reading proficiency, but most commercial reading interventions for secondary students focus exclusively on the latter skill area. This study explored the effectiveness of the Lexia PowerUp Literacy program (PowerUp), a digital reading intervention that provides explicit instruction in word identification, grammar, and comprehension. There were 122 sixth-grade students attending low-SES schools who participated in this year-long study. Students who used PowerUp showed significantly greater gains on an assessment of word identification, syntactic processing, and basic reading comprehension skills compared to students using an alternative program that offered opportunities to apply comprehension strategies in the absence of explicit and skills-based instruction. Results demonstrate the value of instruction extending beyond comprehension strategies to incorporate the full complement of skills needed for reading proficiency.
... However, we could not further analyze potential individual-level factors that might have influenced the representations of radicals embedded in different configurational structures, due to a lack of relevant measurements on the participants (e.g., the time spent on daily Chinese reading, handwriting, and typing on electronic devices, the instruction received when acquiring Chinese literacy, educational level). Moreover, researchers have shown increasing attention to the importance of psychological and ecological factors to literacy development (e.g., see the component model of reading in Aaron et al., 2008). Teachers' knowledge and methods of instruction (e.g., Piasta et al., 2009), learners' attitude and environmental influences (e.g., Chiu & McBride-Chang, 2006;Chiu et al., 2012), as well as other non-cognitive factors may contribute substantially to literacy outcomes, which could potentially account for the individual differences observed in the current study. ...
Full-text available
Most Chinese characters comprise radicals that are embedded in a specific structure (e.g., left–right structure like 相, or top–bottom structure like 杏). We investigated the representations of word-form units (i.e., radicals) in planning Chinese handwritten production. Adopting the picture-word interference paradigm (Experiment 1, N = 29), we manipulated picture names and their corresponding distractors to share one radical. Radicals embedded in the same character structure as the target significantly shortened the participants’ writing latency, no matter whether the radicals appeared at the same or a different position in the distractor. This finding suggests that radicals appearing at different positions in characters with the same structure may share the same representations in planning Chinese written word production. On the other hand, when the shared radical appeared in the distractor that possessed a different structure than the target, a non-significant facilitation trend was observed. Experiment 2 adopted the form-preparation paradigm and obtained a similar non-significant trend of facilitation, in a larger sample (N = 53). Radicals embedded in different character structures might be represented as distinct units among (a proportion of) Chinese writers, suggesting the important role of character structure in planning Chinese handwritten production. The potential influence of language-specific features on the representations of word-form units should be taken into account when building word production models. Practical implications for literacy instruction are also discussed.
... Whereas, children who exhibit unaffected decoding ability exhibit the SCD profile. These children will likely benefit from different instructional approaches because the root cause of their reading difficulties varies (Aaron et al., 2008;Carson et al., 2013;Clarke et al., 2010;Gillon et al., 2019). ...
Full-text available
Traditionally, the New Zealand Ministry of Education opposed the recognition of dyslexia. However, since 2007, the Ministry of Education’s position has started to change, evidenced by the development of a working definition. In 2021 the Ministry of Education released Three Steps in Screening for Dyslexia (TSSD), an assessment protocol designed to support teachers to screen for dyslexia. The current research evaluated the TSSD with a sample of 209 children in Years 4 to 6 (8–10 years-of-age) from New Zealand. The research investigated whether children could be accurately classified using tests from the TSSD, whether the three-step protocol described in the TSSD was a valid assessment approach, and what effect operationalising the term average at different cut-off points had on dyslexia screening. Children were classified using two cluster analyses. The first analysis was based on tests from the Woodcock Johnson IV and the second analysis was based on tests from the TSSD. Subsequent analyses investigated specific aspects of the TSSD protocol, including its sequential design and the placement of cut-off points. Results revealed a number of limitations to the TSSD approach. The authors discuss three changes that could be made to improve the validity and reliability of the TSSD, including a broader assessment of the decoding and language comprehension constructs; directing teachers to assess both decoding and language comprehension, irrespective of a child’s language comprehension ability; and placing a greater emphasis on discrepancy bands over cut-off points.
The present study tested the postulation that “knowledge begets reading, which begets knowledge.” Using Random Intercepts Cross‐Lagged Panel Models (RI‐CLPM), we analyzed a U.S. nationally representative data set to examine the directionality and magnitude of the longitudinal relation between domain knowledge (operationalized as science domain knowledge) and reading throughout the elementary years (from kindergarten to fifth grade), while accounting for important covariates, such as working memory, cognitive flexibility, English language proficiency, basic literacy skills, and demographic information. Moreover, we conducted multi‐group RI‐CLPM analyses to examine whether language status (being bilingual or monolingual) moderates the longitudinal relation between domain knowledge and reading. The results showed that the relation between domain knowledge and reading is bidirectional and positive throughout the elementary years, providing empirical evidence that domain knowledge and reading may mutually enhance with each other. In addition, language status did not moderate the relation between domain knowledge and reading, suggesting that the directionality and magnitude of the relation were similar between bilingual and monolingual students. Taken together, the results have important implications for integrating content knowledge and English language arts core instruction in elementary grades. The present study tested the postulation that “knowledge begets reading, which begets knowledge.” Using Random Intercepts Cross‐Lagged Panel Models (RI‐CLPM), we analyzed a U.S. nationally representative data set to examine the directionality and magnitude of the longitudinal relation between domain knowledge (operationalized as science domain knowledge) and reading throughout the elementary years (from kindergarten to fifth grade), while accounting for important covariates, such as working memory, cognitive flexibility, English language proficiency, basic literacy skills, and demographic information.
Full-text available
The present study explored the relationships between English as a foreign language (EFL) reading anxiety and learning motivation, attitudes, and strategies in two groups of EFL learners with different first language (L1) backgrounds. Thirty-five Chinese EFL undergraduates and thirty-seven Spanish EFL undergraduates participated in this study via a questionnaire. Quantitative analyses indicated different types of relationships between EFL reading anxiety and the abovementioned factors between these two groups of learners. While learning attitudes had a predictive effect on EFL reading anxiety for both Chinese and Spanish EFL undergraduates, the predictive effect of learning motivation and strategies were significant only among Spanish undergraduates. In addition, with learning attitudes as a mediator, learning motivation had an indirect effect on Chinese undergraduates’ reading anxiety. These findings expand previous studies by exploring the relationships between EFL reading anxiety and these correlates among EFL learners with different L1 backgrounds and provide reference and basis for the intervention of FLRA throughout the language learning process.
This paper presents a vulnerability framework as a means to contextualize inequities in reading achievement among children who are vulnerable to poor reading outcomes. Models to understand vulnerability have been applied in the social sciences and public health to identify population disparities and design interventions to improve outcomes. Vulnerability is multifaceted and governed by context. Using a vulnerability framework for the science of reading provides an innovative approach for acknowledging multilevel factors contributing to disparities. The ecological considerations of both individual differences in learners and conditions within and outside of schools ensures that scientific advances are realized for learners who are more vulnerable to experiencing reading difficulty in school.
This paper investigates the implementation of a reading program designed for third grade elementary school classrooms in Austria. Using a mixed-methods approach, lesson types were identified, respective class compositions analyzed, and the effects on students’ reading gains examined. The results show that the lesson types seem to reflect learner group needs as students in the different lesson types profited similarly from the reading program. This supports the hypothesis that reading is a highly individualized process and leads to the conclusion that the instructional approach of the reading program investigated supports teachers in conducting reading lessons tailored to their students’ needs.
Full-text available
Background: The term learning disability applied to explain the problems of a person with average intelligence in basic academic skills and can affect a person's ability to understand and use spoken or written language, mathematical calculations, coordinate movements, or direct attention. Learning disability can be divided into three difficulties in reading, writing, and arithmetic. The prevalence of learning disability in recent years significantly increased and in several studies, it has been various estimates. Conclusion: One of the main reasons for the significant differences in the prevalence of this disorder in several studies is different diagnostic approaches. Three approaches in diagnosis learning disabilities are inconsistencies between achievement-ability, response to intervention, and cognitive processing approach. Each of these approaches has their own supporters and opponents. Thus, in this paper, in addition to the overview of definitions and descriptions of Learning disability, various diagnostic approaches are discussed in detail and the final conclusions are done.
Full-text available
Psychologists interested in the workings of the mind may wonder whether brain-imaging data can provide insight regarding cognitive mechanisms. Here we consider one means through which imaging can inform cognitive theory: reverse inference, wherein activations in well-characterized neural structures serve as markers for the engagement of particular cognitive processes. To illustrate this approach, we review brain-imaging evidence regarding the organization of cognitive and linguistic processes in the prefrontal cortex, which indicates that phonological (speech-sound-based) and semantic (meaningbased) processing are consistently associated with topographically distinct patterns of activity in the left inferior prefrontal cortex. We then illustrate how this finding of regional differentiation has provided useful guidance for understanding the cognitive processes supporting memory encoding and retrieval. We conclude with caveats that highlight some of the limitations of the reverse-inference approach.
Full-text available
Fifty-six first-grade children were administered measures of general intelligence, decoding speed, phonological awareness, and listening comprehension. All four types of measures were moderately related to end-of-year reading comprehension. Decoding speed accounted for the largest amount of unique variance. The hypothesis that reading is strongly related to general intelligence once differences in decoding ability have been accounted for was not supported. Other relationships among the variables were explored via multiple regression, factor analysis, and path analysis. Developmental comparisons were made with groups of third- and fifth-grade children. The relationships between decoding, intelligence, and reading comprehension found in the first-grade sample were replicated in the fifth-grade sample but were somewhat different in the third-grade sample. The interrelationships between the various subskills of reading and intelligence increased with age, probably due to mutual facilitation. /// [French] On a administré à cinquante six enfants de cours primaire des mesures d'intelligence générale, de vitesse de décodage, de conscience phonologique et de compréhension d'écoute. Les quatre facteurs étaient modérément reliés à une mesure de compréhension de lecture de fin d'année. La vitesse de décodage rendait compte de la plus large quantité d'écart unique. L'hypothèse qui veut que la lecture soit fortement reliée à l'intelligence générale après avoir tenu compte des différences de compétence de décodage, n'a pas été soutenue. On a exploré d'autres rapports parmi les écarts à travers une régression multiple, une analyse de facteur et une analyse de parcours. On a établi des comparaisons de développement avec des groupes d'enfants de neuvième et septième. Les rapports entre décodage, intelligence, et compréhension de lecture trouvés dans l'échantillon de cours primaire sont répétés dans l'échantillon de la septième mais ont été quelque peu différents dans celui de la neuvième. Les rapports étroits entre les différentes sous-compétences de lecture et intelligence ont augmenté avec l'âge, ceci étant probablement dû à une facilitation réciproque. /// [Spanish] Se administraron medidas de inteligencia general, velocidad de descifre, conocimiento fonológico y escuchar y comprender. Los 4 factores estaban moderadamente relacionados a una medida de comprensión de lectura de fin de año escolar. La velocidad de descifre dio cuenta del mayor número de variabilidad singular. No recibió apoyo la hipótesis de que la destreza de lectura está fuertemente relacionada a inteligencia general, una vez que se han considerado las diferencias de habilidad de descifre. Otras relaciones entre variables fueron exploradas por medio de regresión múltiple, análisis factorial y análisis de trayecto. Se hicieron comparaciones de desarrollo con grupos de alumnos de tercer y quinto grado. La relación entre descifre, inteligencia y comprensión de lectura encontrada en la muestra del primer grado, fue repetida en el quinto grado, pero resultó algo diferente en la muestra del tercer grado. La correlación entre las varias subdestrezas de lectura e inteligencia aumentó con la edad, probablemente debido a facilitación mútua.
In the US, children with average or higher intellectual ability but have difficulty in learning to read are described as having Learning Disability (LD, hereafter). The assumption that underlies the classification of poor readers into the LD and non-LD categories is that there are qualitative differences in the cognitive skills of these categories of poor readers. In addition to the questionable validity of this assumption (e.g., Fletcher, 1992), implementation of the discrepancy formula has run into several conceptual and practical problems. One of the difficulties arises from certain statistical problems and computational difficulties associated with the formula itself (c.f., Reynolds, 1985). During the course of the past few years, some of these statistical bugs have been removed and the formula patched up. Nevertheless, a major problem of a practical nature remains. That is, the correlation between IQ and reading achievement seldom exceeds.50 indicating that IQ scores do not account for more than 25% of variance seen in the reading performance of children from elementary grades (Stanovich, Cunningham, & Feeman, 1984). This means IQ is not a potent predictor of reading potential. Another conceptual problem arises from the likelihood that IQ does not influence reading ability unidirectionally, but that the relationship between IQ and reading is reciprocal in the sense that not only the level of IQ can influence the reaing score but reading experience also can affect IQ.
Hyperlexia has been described as a unique disorder in which children notably delayed in language and cognitive development begin spontaneously to recognize words at a very early age and continue to develop surprisingly advanced word recognition skills. Although such divergent development is provocative for both theory and practice in reading, little research attention has been directed toward hyperlexia, and a systematic description of the reading behaviors of such children has never appeared. This study describes 12 such children in terms of cognitive, linguistic, and reading competencies. Extensive interviews and individual testing with each subject confirmed the presence of an identifiable syndrome of hyperlexia, characterized by word recognition before age 5, disordered cognitive and linguistic development, and word recognition skills notably superior to other tested abilities. Comprehension, while present for literal units, broke down when abstract or relational thinking beyond a perceptual level were required. Comparison of listening and reading comprehension indicated generalized comprehension deficits, and results of testing suggested that hyperlexic children are deficient in formation and organization of schemata for incoming experience. It is proposed that, despite its unique nature, the study of this syndrome may contribute to our understanding of both word recognition and comprehension and their relationship in the total reading process./// [Spanish] Se ha descrito hiperlexia como un trastorno singular en el que niños con retraso de lenguaje y desarrollo cognoscitivo, comienzan súbitamente a reconocer palabras a muy temprana edad y continúan a desarrollar reconocimiento de palabras avanzadas de una manera sorprendente. Aunque tal desarrollo divergente pone a prueba a ambas la teoría y la práctica de lectura, poco interés de investigación ha sido dirigido hacia hiperlexia, y una descripción sistemática de hábitos de lectura de estos niños nunca ha aparecido. Este estudio describe la destreza cognoscitiva, lingüística y de lectura de 12 niños en esta categoría especial. Entrevistas extensas y exámenes individuales con cada alumno confirmaron la presencia del sindrome identificable de hiperlexia, caracterizado por reconocimiento de palabras antes de los 5 años, desarrollo cognoscitivo y lingüístico divergente, y destreza de reconocimiento de palabras considerablemente superior a otras destrezas analizadas. Comprensión, aunque existente en unidades literales, cesó cuando se requería análisis abstracto o de relación más allá de un nivel perceptivo. La comparación de la habilidad de escuchar y entender y de la comprensión de lectura, indicó déficits generalizados de comprensión, y los resultados de los exámenes sugirieron que niños con hiperlexia sufren deficiencias en la formación y organización de esquemas para procesar experiencias nuevas. Se propone, que a pesar de su naturaleza única, el estudio de este síndrome puede contribuir hacia nuestra compensión de ambos, reconocimiento de palabras y de comprensión, y de su relación en el proceso total de lectura./// [French] L'hyperlexie a été décrite comme désordre unique au cours duquel les enfants notamment retardés dans le domaine du langage et du développement cognitif, commencent spontanément à reconnaitre les mots à un très jeune âge et continuent à développer des compétences de reconnaissance de mots étonnamment avancés. Bien qu'un tel développement divergent soit stimulant pour la théorie et la pratique en lecture, on a dirigé une attention de recherche moindre vers l'hyperlexie, et par conséquent une description systématique des comportements de lecture chez de tels enfants n'a jamais pris jour. Cette étude décrit 12 enfants en termes de compétences de lecture, de cognition, et de linguistique. Des entrevues considérables et des séries de tests individuels avec chaque sujet ont confirmé la présence d'un syndrome identifiable d'hyperlexie, caractérisé par la reconnaissance de mots avant l'âge de 5 ans, un développement linguistique et cognitif désordonné, et des compétences de reconnaissance de mots notamment supérieures à d'autres capacités testées. Tandis que la compréhension est présente dans les unités littérales, elle diminue lorsque une reflexion abstraite et de rapport au delà d'un niveau perceptif est requise. Une comparaison de compréhension d'écoute et de lecture a indiqué des déficits de compréhension généralisée et des résultats de tests ont suggéré que les enfants hyperlexiques manquent de formation et d'organisation de schémas pour toute expérience reçue. On a proposé qu'en dépit de sa nature unique, l'étude de ce syndrome peut contribuer à un meilleur entendement de la reconnaissance de mots et de leur compréhension et leurs rapports dans le procédé total de lecture.
In two separate studies, 100 South Korean kindergartners and 100 second graders were administered tests of speed of processing and phonological-processing skills as well as a Korean Hangul reading test. Speed of processing tasks were significantly correlated with most of the reading-related tasks but not with Hangul reading itself. Across studies, only tests of syllable and phoneme awareness (either onset or both onset and coda), as well as naming speed, were significantly associated with Korean word-reading. In both studies, in regression equations including all measured skills and age, only syllable and phoneme awareness uniquely predicted Hangul word recognition. Results underscore the uniqueness of the Korean Hangul orthography, which requires children to be sensitive to both syllable- and phoneme-level linguistic and orthographic units.