INT. J. LANG. COMM. DIS.

2010, Early Online Article, 112

Research Report
Follow-up study on reading comprehension in Downs syndrome: the role
of reading skills and listening comprehension
Maja Roch, Elena Florit and Chiara Levorato
Department of Developmental Psychology, University of Padua, Padua, Italy
(Received 8 June 2009; accepted 15 April 2010 )

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Abstract
Background: According to the Simple View of Reading, reading comprehension requires some abilities such as reading
skill and listening comprehension. Individuals with Downs syndrome show relative strengths in reading skills, mainly
in word recognition, where they attain a reading age of about 78 years. Compared with word recognition, their
reading comprehension is usually delayed by at least 6 months. Poor reading comprehension is paralleled by weak
listening comprehension. It is claimed that poor listening comprehension might constrain the development of reading
comprehension and, therefore, be a cause for the asynchrony between reading skills and reading comprehension.
Aims: A follow-up study was carried out in order to analyse the improvements in reading skills, listening and reading text
comprehension, and to support the hypothesis of a causal relationship between listening and reading comprehension.
Methods & Procedures: Ten children and adolescents with Downs syndrome, aged between 11 years 3 months and
19 years 10 months, were assessed twice over a one-year period as to their reading skills, listening and reading text
comprehension.
Outcomes & Results: Three main findings emerged: (1) reading skills, on the one hand, and comprehension (both
listening and reading), on the other hand, are independent; (2) reading comprehension development is determined
mainly by listening comprehension, which in the present study proved to be very poor; and (3) an improvement after a
one-year period, even though limited, occurred for all examined abilities except for listening comprehension.
Conclusions & Implications: The results are discussed in the light of the theoretical framework of the Simple View of
Reading and of their relevance for practical and educational issues.
Keywords: Downs syndrome, Simple View of Reading, reading skills, listening comprehension, reading
comprehension, follow-up study.

What this paper adds

What is already known on this subject
Recent studies have shown that most individuals with Downs syndrome who attend mainstream schools develop at
least some degree of reading skills, despite their impaired linguistic competence. For reading comprehension to
occur, both reading skills and language comprehension are necessary, and they both give a unique and specific
contribution to reading comprehension. The role played by these components in the reading comprehension of
individuals with Downs syndrome is atypical and similar to that exhibited by typically developing children who are
poor comprehenders. Their ability to understand written texts is more strongly related to their listening
comprehension level rather than to their reading skills.
What this study adds
The current study supported the hypothesis that the nature of the relationship between listening and reading
comprehension in Downs syndrome is causal: the level of listening comprehension is the strongest predictor of the
level of reading comprehension one year later. On the other hand, the relative contribution of reading skills to
reading comprehension one year later is weak. The educational implication of this study is that training based on
abilities related to listening comprehension might be appropriate for the improvement of reading comprehension as
are training in reading skills.

Address correspondence to: Maja Roch, Developmental Psychology, University of Padua, via Venezia, 8, Padua I-35133, Italy;
e-mail: maja.roch@unipd.it
International Journal of Language & Communication Disorders
ISSN 1368-2822 print/ISSN 1460-6984 online q 2010 Royal College of Speech & Language Therapists
http://www.informahealthcare.com
DOI: 10.3109/13682822.2010.487882

Maja Roch et al.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Introduction
The current study analysed the development of reading
comprehension in individuals with Downs syndrome;
specifically, a follow-up investigation analysed the role
played by reading skills and listening comprehension in
the development of reading comprehension.
Learning to read is an essential competence not only
for typically developing children, but also for people
with cognitive impairments, such as individuals with
Downs syndrome. In fact, literacy may broaden
vocational opportunities as well as contribute to increase
personal autonomy and independence in community
activities and living (Boudreau 2002). In Italy, as in
various other European countries, individuals with
Downs syndrome attend mainstream educational
institutions together with typically developing children
of a similar age. This educational integration allows
most children with Downs syndrome to develop at least
some degree of reading competence.
All the studies conducted to date on reading skills1
agree that, despite their severe language difficulties,
individuals with Downs syndrome show relative
strengths in word identification, where they can attain
a reading age of about 7 8 years (Byrne et al. 2002).
On the other hand, they show a selective disadvantage
in non-word reading (Cupples and Iacono 2002). This
pattern of strengths and weaknesses can be attributed to
a relatively preserved visual approach in word
recognition, which contrasts with an impairment in
phonology (Roch and Jarrold 2008). Compared with
word and non-word reading, which are skills that have
been extensively investigated, reading comprehension
has been less studied in individuals with Downs
syndrome. Some recent studies suggest that the
advantage in word identification is not a guarantee
that what is read is also understood (Roch and Levorato
2009). In fact, the comprehension of written sentences
or short texts seems to be delayed at least 6 months
with respect to word identification (Boudreau 2002,
Laws and Gunn 2002, Verucci et al. 2006).
Reading comprehension difficulties are likely to
reflect weaknesses in oral language abilities. Compared
with non-verbal cognition, language acquisition in
Downs syndrome is particularly impaired (Chapman
1995, Abbeduto et al. 2007). Since the comprehension
of what is read is the ultimate aim of learning to read, a
detailed investigation of reading comprehension and its
components is essential both to identify the difficulties
that individuals with Downs syndrome may encounter
in this process and, subsequently, to plan educational
programmes.
In the current study we used a follow-up design to
investigate the extent to which reading skills and
listening comprehension are causally linked to the

development of reading comprehension in a group of

children and adolescents with Downs syndrome.
The Simple View of Reading (Hoover and Gough
1990)
The theoretical background provided by the Simple
View of Reading (Gough and Tunmer 1986, Hoover
and Gough 1990) indicates that reading skills and
listening comprehension are components of reading
comprehension. According to this model, reading
comprehension is the product of two distinct
components, each of which makes its own unique and
independent contribution (Savage 2001): decoding and
listening comprehension. Both of these abilities are
necessary, but neither is sufficient in itself to explain
reading comprehension (Gough et al. 1996). The model
may be expressed by the formula:
r dc
where r is reading comprehension; d is decoding, and c
is listening comprehension. Each of these variables can
assume a value between zero and one, and for reading
comprehension to occur, both values must be greater
than zero. According to this model, reading comprehension correlates with both of these components,
which in turn are independent of each other and are
based on different mechanisms (Savage 2001). Decoding refers to the ability to read printed symbols by
applying the grapheme phoneme conversion rules and
is based mainly on phonological skills (Kirby and
Savage 2008). In the original definition, Gough and
Tunmer (1986) conceptualized listening comprehension as the ability to take lexical information (word
level) and derive sentence and discourse interpretations.
Hoover and Gough (1990) stated that measures of
listening and reading comprehension should assess the
ability to answer questions about the contents of aurally
and written presented text. Additionally, de Jong and
van der Leij (2002) found in first to third grades of
typically developing children that the effect of listening
comprehension was more important than the effect of
vocabulary in predicting reading comprehension.
While the former explained unique variance, the
vocabulary did not add a unique contribution over and
above the listening comprehension. For this reason, the
Simple View of Reading has been tested mainly by
taking into account the relationship between the
comprehension of oral and written texts. In different
phases of the acquisition of the ability to read and
understand written texts, the two components
decoding and listening comprehensionhave different
levels of impact: Decoding is a strong predictor of
reading comprehension mainly in the early stages of
reading development because most of the necessary
resources are involved in the processing of graphic

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Reading comprehension in Downs syndrome

symbols. On the other hand, when reading becomes an
automatized activity, during middle and high school
listening comprehension is increasingly strongly related
to reading comprehension (Diakidoy et al. 2005).
The validity of this model has been demonstrated
in typically developing children, both English speakers
(for example, Catts et al. 2006, and Joshi and Aaron
2000) and speakers of languages, such as Dutch and
French, which have a more transparent orthography than
does English (Aro and Wimmer 2003, de Jong and
van der Leij 2002). These studies have demonstrated
that the degree of orthographic transparency of the
language that has to be mastered affects the development
of reading comprehension. First of all, in orthographically
transparent languages word recognition is more predictive of reading comprehension than decoding; since
reading is in general accurate, speed is a more reliable
measure than accuracy. Moreover, learning to decode is
a more rapid process and, as a consequence, listening
comprehension becomes a powerful predictor of reading
comprehension since the early phases of literacy
acquisition (Megherbi et al. 2006).
Italian has a highly regular orthography. A study
carried out with Italian third- and fifth-grade typically
developing children confirmed the predictions of the
Simple View of Reading. Both word-reading skills and
listening comprehension were related to reading
comprehension; the latter accounted for more variance
in reading comprehension in older children than
reading skills, while in third graders word recognition
and listening comprehension accounted for a similar
amount of variance (Florit et al. 2008).
The validity of this model has also been confirmed
in some populations characterized by difficulties in the
development of language and reading skills. Studies
that compared children as to their ability to read, to
understand oral language, and to understand written
language have identified different profiles. Children
with deficits in reading comprehension, often
accompanied by deficits in oral language comprehension, but with age-appropriate word reading, have been
defined as poor comprehenders (Cain and Oakhill
2006). On the other hand, poor decoders are
characterized by an opposite pattern: They show
average language abilities but poor decoding skills and
phonological processing (Catts et al. 2006). These
studies indirectly support the claim that reading skills
and language comprehension are both involved in
reading comprehension and that they are independent
of each other, even though some children can be
impaired in both (garden variety poor readers).

3
The Simple View of Reading in Downs syndrome
Recently, the Simple View of Reading provided the
basis for an investigation regarding individuals with
Downs syndrome (Roch and Levorato 2009). In this
study, a group of individuals with Downs syndrome
(mean age 15 years and 5 months) and a group of
typically developing children (mean age 6 years and
8 months) were matched for their level of reading
comprehension and then compared as to their reading
skills and listening comprehension (both word and
non-word reading). Some differences emerged in the
profiles of the two groups. When matched with
typically developing children with the same level of
reading comprehension, participants with Downs
syndrome were shown to be weaker in listening
comprehension and more advanced in word recognition. The authors concluded that individuals with
Downs syndrome show an uneven profile when
compared with typically developing children, and that
their profile might be considered to be similar to that
exhibited by poor comprehenders. The analysis of the
relationships between the components considered and
reading comprehension showed that: (1) in typically
developing children, both reading skills and listening
comprehension accounted to a similar extent for
differences in reading comprehension, as expected on
the basis of the Simple View of Reading; and (2) in the
group with Downs syndrome, listening comprehension was a stronger predictor of reading comprehension
than reading skills. Even though the reading comprehension level was similar in the two groups, the
underlying abilities proved to be different. These results
call for a deeper analysis of a possible causal link
between listening and reading comprehension, on the
one hand, and reading skill and reading comprehension, on the other hand. The follow-up procedure is a
privileged method for tracing causal relations between
reading comprehension and its components (Cain and
Oakhill 2007). For this reason, we carried out a study
on the development of reading comprehension in a
follow-up investigation, and, in the light of the results
reported by Roch and Levorato (2009), it was expected
that poor listening comprehension might be causally
related to the difficulties that individuals with Downs
syndrome exhibit in reading comprehension.
Development of reading skills in Downs syndrome
Several studies have undertaken a longitudinal analysis
of the development of reading and reading-related skills
in Downs syndrome. In a 5-year-longitudinal study,
Laws and Gunn (2002) tested the hypothesis that
learning to read improves language comprehension in a
group of 30 individuals with Downs syndrome aged
between 10 and 24 years. Most of the participants who

Maja Roch et al.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

4
were able to read made progress over the 5-year period
covered by the study, although they showed an uneven
profile. Compared with their word-reading skills, which
were relatively preserved, they showed a disadvantage in
reading comprehension, which was 6 21 months
delayed in comparison with word-reading skills. Several
measures at Time 1 were significant predictors of the
level reached in both reading skills and reading
comprehension at Time 2: non-verbal ability, language
comprehension and production, phonological memory
and phonological awareness, but no evidence was found
that learning to read produced benefits in the
development of language comprehension.
Byrne et al. (2002) also conducted a longitudinal
study that analysed the development of reading skills
and the cognitive abilities associated with reading
progress in children with Downs syndrome. Twentyfour children with Downs syndrome aged between 4
and 12 years were followed over a two-year period. They
were administered annually a large battery of tests
evaluating the progress of their reading, language and
cognition. Their profiles were compared with that of
two control groups of children with typical development: one consisting of slow readers who had the same
reading age as children with Downs syndrome; and one
having age-appropriate reading skills. The results
indicated that individuals with Downs syndrome
performed well in word recognition in comparison
with their cognitive abilities, and that they made
significant progress over a one-year period in single
word-reading skills. On the other hand, significant
improvement in reading comprehension was observed
only after two years, and the gains were very small:
reading comprehension scores did not rise above the
reading comprehension age of 6 years compared with
average readers. When compared with the control
group of younger reading-age children, participants
with Downs syndrome did not differ in single-word
reading from them, even though they had poorer
language memory and general intelligence. Overall, this
result suggests that reading may be relatively independent from other cognitive and linguistic skills: in all
three groups there was no evidence of any strong
and direct relation between reading ability, reading
comprehension, and other measures of language
acquisition. This study supported the view that
reading skills and reading comprehension
develop independently in Downs syndrome, but it
did not provide evidence about the factors that
determine improvements in reading comprehension.
Aims and predictions
The aim of the current study was to identify the causal
contribution made by the components of reading

comprehension in a one-year period. In accordance

with the Simple View of Reading, both reading skills
and listening comprehension were included as
possible predictors of reading comprehension. Taking
the theoretical background of the Simple View of
Reading and the above-mentioned evidence together,
the following aims were formulated:
. To trace the trend of improvement in all the abilities

examined (listening and reading comprehension, word

and non-word reading) after a one-year period.
Consistently with previous longitudinal studies
(Byrne et al. 2002, Laws and Gunn 2002), we
expected greater gains in reading skills than in
comprehension, both listening and reading.
. To verify whether listening comprehension was a
stronger predictor of the development of reading
comprehension than reading skills. In accordance
with Roch and Levorato (2009) and coherently with
the profile described for poor comprehenders
(Cain and Oakhill 2006), it was hypothesized that
listening comprehension might be causally related to
reading comprehension while reading skills, although
more preserved, might have a less strong influence
on the development of reading comprehension.

Method
Participants
Ten individuals with Downs syndrome took part in
every phase of the present study. An additional five
participants dropped out between Time 1 and Time 2
and were not considered as participants. They were
selected from a larger group of 20 children and
adolescents who were attending elementary, middle and
high school, were younger than 20 years, and were able
to read. They were all native Italian speakers (four males
and six females). Only those who did not have any
physical or neurological problems, or who were not
being treated for either physical or neurological diseases
at the time of the assessments, were included as
participants of the study. They were assessed twice in a
one-year period. At Time 1, participants were aged
between 11 years and 3 months and 19 years and 10
months (mean 15 years and 6 months, standard
deviation (SD) 2 years and 8 months). All these
participants had full trisomy 21, and had normal or
corrected vision and hearing. The National Health
Service (NHS) provides them with regular medical
examinations, psychological evaluation, and monitoring once or twice a year. In accordance with Italian law,
all of them were attending mainstream schools with
children and adolescents who had approximately the
same chronological age (mean years of schooling 9
years, SD 2 years); these children were supported for
normal class activities by a specially trained teacher.
The NHS supplied the data regarding their full-scale

Reading comprehension in Downs syndrome

intelligence quotient (IQ), as measured by the WISC-R
test (adapted for Italian speakers by Orsini 1993): their
mean IQ was 52 (SD 9), ranging from 40 to 65.
Materials

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Evaluation of reading comprehension

In order to assess reading comprehension, a testMT
testvalidated on Italian school-age children
(n 5700) sampled in different areas of Italy was used
(Cornoldi and Colpo 1998). The test for the first grade
was chosen in accordance with results from the literature,
indicating that reading comprehension in Downs
syndrome individuals is usually comparable with that
of 67-year-olds (Verucci et al. 2006). The test requires
each participant to read a one-page story followed by ten
multiple-choice questions, which are answered by
choosing one out of three alternatives. The story is
formed of syntactically simple sentences, the text
structure respects the story grammar, and the words
used are appropriate for children of 67 years of age.
The comprehension questions concern information that
may be either explicitly or implicitly stated in the text.
There are no time limits, and to minimize the memory
load children are allowed to return to the text whenever
they want while answering the multiple-choice questions.
Scores correspond to the number of correct answers
(zero to 10) and are transformed into the percentage of
correct answers. The reliability of the test measured by an
item total correlation ranges between 0.41 and 0.69.
Evaluation of listening comprehension
Listening comprehension was evaluated by using the
test TOR 3 8, which was standardized on 1700 Italian
children aged between 3 and 8 years (Levorato and
Roch 2007), and it measures listening text comprehension without involving language-production skills.
The test is similar to the reading comprehension test
in terms of story structure, types of questions (that is,
literal and inferential) and type of task (that is, multiple
choice), and it was chosen in order to have a parallel
measure of listening and reading comprehension, as
suggested by previous studies (Hoover and Gough
1990, de Jong and van der Leij 2002).
In this study, participants were asked to listen to
two stories which were of the same level of difficulty as
those used for the reading comprehension, namely for
6 8-year-old Italian children. The two stories are
characterized by sentences with a simple syntax and by a
lexicon appropriate for Italian children of the age
considered. The test procedure calls for the stories to be
read aloud by the experimenter who, in order to
minimize the cognitive and the memory load,
interrupts the reading of the story in two predetermined

5
points and verifies comprehension up to the point of
interruption. Comprehension is evaluated by multiplechoice questions having four alternative answers, each
of which is accompanied by an image. The alternatives
are expressed verbally by the tester, who also points to
the corresponding picture and participants are asked to
point to the correct picture. Half the questions concern
information explicitly stated in the story and half
require an inference to be made. The score consists of
the sum of correct answers, twelve for each story, for a
total of 24. The raw scores were transformed into
percentages of correct answers. The reliability of the test
measured by Cronbachs alpha is 0.60 and the test
retest reliability is 0.85.

Evaluation of reading skills: fluency and accuracy

Given the uneven profile of participants with respect to
their reading skills, both word and non-word reading
were assessed. We used two tasks from the Battery for
Dyslexia Evaluation (Sartori et al. 1995) standardized
on 929 Italian-speaking children: the tasks evaluate
word and non-word reading for 7-year-old typically
developing children.
The test requires participants to read a list of items
consisting of 112 words and 48 non-words as fast and as
accurately as they can. The words are bi-, tri-, and quadrisyllabic items which differ for the age of acquisition and
frequency (ranging from high to moderately low). All the
items have a regular pronunciation. The non-words are
similar to the word items for length (two, three of four
syllables) and are constructed respecting the phonotactic
rules of Italian.
The average time (seconds) needed to read the list
of words and non-words (total time/number of items)
and the proportion of word and non-word-reading
errors (total errors/number of items) produced four
variables:
. Word-reading fluency: speed (seconds) of word

reading.2

. Word-reading accuracy: percentage of word-reading

errors.
. Non-word-reading fluency: speed (seconds) of non-

word reading.
. Non-word-reading accuracy: percentage of non-word-

reading errors.

The test retest reliability is 0.80 for speed and 0.71 for
accuracy.
To evaluate the role of reading skills, we used the
measure of word-reading fluency, which is the most
powerful predictor of reading comprehension in
languages with transparent orthographies (cf. Aro and
Wimmer 2003, Florit et al. 2008, Roch and Levorato
2009).

Maja Roch et al.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Procedure
The tests were administered to all participants twice by
the same person, whom they already knew in advance,
with a one-year period of distance between the first
(Time 1) and the second (Time 2) assessments. The
same tests were administered at both time points.
Testing took place over two visits, and the participants
were assessed individually in a quiet room. The usual
order of assessments was word reading then listening
comprehension in the first session and non-word
reading then reading comprehension in the second
session. Each session lasted from 30 to 40 min. Schools
and parents provided written consent for participants to
be assessed at school or at home.
Results
In order to analyse the role played by the individual
characteristics of the participants with Downs
syndrome, their performance on each of the tasks
(reading and listening comprehension, word and nonword reading) was related at both time points to each of
the individual characteristics: chronological age, years
of schooling and IQ. No significant correlation
emerged between the variables (r , 0.35 in all cases).
See table 1 for details.

These results suggest that: (1) in the age range of the

participants, the impact of life experience on reading
and reading related skills might be marginal; and (2) no
important changes in literacy experience occurred
during the year of the study. In the light of these
considerations, age, years of schooling and mental
development were not taken into consideration in any
subsequent analysis.
Table 2 presents the performances of the group of
participants at Time 1 and Time 2.
In reading comprehension, seven out of ten
participants (at both points in time) achieved a level
of comprehension which, according to the guidelines of
the test, is considered appropriate for first graders.
Three participants achieved the highest score, which
means that their comprehension level might be even
higher than that expected in first graders. On the other
hand, listening comprehension was an area of weakness
in this group of participants, in particular in
comparison with their ability to understand written
texts. In fact, only four participants answered more than
50% of the comprehension questions correctly, which
means that listening comprehension is poorer than
reading comprehension. Word recognition seems to
constitute a relatively preserved ability in which the
participants performed well. They were fluent in word

Table 1. One-tailed zero-order correlations between individual characteristics (age, schooling and intelligence quotient (IQ)) and the
tasks adopted in the study (word and non-word reading speed and accuracy, listening and text comprehension) at Time 1 and at Time 2
Time 1

Reading comprehension
Listening comprehension
Speed (words)
Speed (non-words)
Errors (words)
Errors (non-words)

Time 2

Age

Schooling

IQ

Age

Schooling

IQ

0.129
2 0.261
0.119
0.217
2 0.010
0.005

0.180
2 0.221
0.192
0.189
2 0.034
0.040

0.032
20.029
20.156
0.128
0.324
0.297

2 0.295
2 0.233
2 0.263
2 0.287
0.271
0.344

20.246
20.266
20.295
20.269
0.120
0.219

0.256
0.212
0.162
0.056
0.058
0.064

Table 2. Description of participants at Time 1 and Time 2: proportions of correct answers, standard deviations, and ranges on
reading and listening comprehension, and on word and non-word reading (speed and accuracy)

Reading comprehension
Range
Listening comprehension
Range
Speed words
Range
Speed non-words
Range
Errors words
Range
Errors non-words
Range

Time 1

Time 2

0.59 (0.25)
0.20 1.0
0.41 (0.14)
0.17 0.63
1.4 (0.23)
1.04 5.73
3.3 (1.6)
1.17 5.73
0.10 (0.12)
0 0.41
0.31 (0.26)
0.06 0.92

0.64 (0.26)
0.20 1.0
0.41 (0.13)
0.25 0.63
1.5 (0.34)
1.12 2.05
2.5 (0.37)
1.77 3.15
0.07 (0.07)
0.01 0.20
0.15 (0.13)
0.04 0.42

0.89 d 0.19

Notes: The comparison of the performance on each task between Time 1 and Time 2 is reported (t-test and effect size). *p , 0.05.

0.12 d 0
1.6 d 0.68
1.7 d 0.69
1.4 d 0.30
2.4* d 0.80

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Reading comprehension in Downs syndrome

recognition comparably with second-grade typically

developing children, and most of them were even more
accurate in word recognition than 7 8-year-old
children. Two participants made no reading errors in
a list of 112 words and three of them made just three or
four errors. Different results were obtained regarding
non-word decoding: six participants read more than
30% of the items incorrectly. Most of the errors
consisted in reading non-word items as if they were real
words, suggesting that some of the participants
persisted in using a visual route for reading even
though it was inappropriate and they were informed
that they had to read also nonsense words.
In one years time the participants did not undergo
any specific intervention, but participated in normal
school activities, among which literacy, text reading
and listening. Comparing the mean scores obtained in
the two sessions, little improvement could be observed.
A t-test compared the performance of participants at
Time 1 and Time 2 in order to evaluate the amount
of improvement occurring between the two times.
As far as language comprehension is concerned, an
improvement occurred in reading comprehension, even
though not statistically significant, whereas no improvement occurred in the one-year period for listening
comprehension. As far as reading skills are concerned,
participants improved both in speed and in accuracy of
words and non-words reading (see effect sizes), but there
was a significant decrease in the number of errors only for
non-words. This improvement had a fairly large effect
size, namely 0.80.
In order to evaluate whether the same pattern of
relationships between the tasks emerged at the two time
points, Pearsons zero-order correlations were carried
out. Table 3 reports the results.
At both points in time, reading comprehension was
correlated with listening comprehension, but not with
any of the reading measures. Also, listening comprehension correlated only with reading comprehension,
but not with any of the reading measures. On the
other hand, the measures of reading ability showed
a strong pattern of relationships between each other.
This pattern of results is very similar to that reported by
Roch and Levorato (2009), and suggests a dissociation
between language comprehension and reading skill.
Table 3.

The correlations observed at Time 1 and one year later

are very similar, showing that the relations among the
abilities measured are stable over a one-year period.
In order to verify whether the pattern of correlations
with reading comprehension might be interpreted as
causal, a hierarchical linear regression analysis was
carried out. The dependent variable was reading
comprehension at Time 2 and the predictors were
performance at Time 1 in reading comprehension,
reading skill and listening comprehension. The order of
entrance of the predictors was established a priori. At the
first step, the reading comprehension score at Time 1
had the function of autoregressor and it is particularly
important for testing a causal hypothesis. In fact, this
procedure allowed for a control of the variance not only
explained by the autoregressor itself, as measured at time
1, but also of its correlations with the other components.
According to de Jong and van der Leij (2002), any
additional variance after the inclusion of the autoregressive effect can be taken as support for a causal
relation between the predictors and the dependent
variable. At the second step, word recognition speed
at Time 1 was inserted. This variable was inserted
because we wanted to test the role of listening
comprehension on reading comprehension excluding
the role of word recognition. Finally, at the third step,
listening comprehension at Time 1 was entered. Table 4
illustrates the results of the regression analysis.
Reading comprehension at Time 1 explained
57.4% variance in the performance of the same task
one year later. Word recognition added 8% of the
variance over and above the variance explained by the
autoregressor, which did not yield statistical significance. Listening comprehension explained a further
31.6% of unique variance. This variance emerged after
controlling for the relation between reading and
listening comprehension at Time 1, and the possible
role of reading skills. Therefore, it may be argued that
reading comprehension is strongly related to listening
comprehension, which seems to be a causal factor
in predicting reading comprehension one year later.
In fact, we are able to rule out the possibility that the
relation between listening comprehension at Time 1
and reading comprehension at Time 2 was simply due

Correlations between performances at Time 1 (above the diagonal) and at Time 2 (below the diagonal)

1. Reading comprehension
2. Listening comprehension
3. Speed (words)
4. Speed (non-words)
5. Errors (words)
6. Errors (non-words)
Note: *p , 0.05; and **p , 0.01.

1.

2.

3.

4.

5.

6.

1
0.717*
0.105
2 0.143
2 0.354
2 0.149

0.721**
1
0.165
0.091
2 0.493
2 0.419

0.158
20.132
1
0.662*
0.254
0.150

0.362
2 0.133
0.757**
1
0.180
0.147

0.223
0.018
0.376
0.587*
1
0.937**

0.279
0.066
0.215
0.416
0.958**
1

Maja Roch et al.

8
Table 4.

Regression analysis on the reading comprehension score at Time 2: R 2 5 0.971 [F (3, 9) 5 65.8, p < 0.001]
R 2 change

Predictors
Step 1: Reading comprehension (T1)

0.574a

Step 2: Word reading speed (T1)

0.081b

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Step 3: Listening comprehension (T1)

0.316c

Reading comprehension (T1)

0.757

3.3

,0.01

Reading comprehension (T1)

Word reading speed (T1)

0.803
20.288

3.6
2 1.3

,0.01
n.s.

Reading comprehension (T1)

Word reading speed (T1)
Listening comprehension (T1)

0.141
20.069
0.869

1.3
2 0.903
8.1

n.s.
n.s.
,0.001

Notes: aF change (1, 8) 10.8, p , 0.01.

b
F change (1, 7) 1.6, not significant (n.s.).
c
F change (1, 6) 64.4, p , 0.001.

to its association with reading comprehension at

Time 1.
The next analysis was aimed at testing the causal
hypothesis, excluding the possibility that such a relation
might simply be due to the fact that the correlation
between reading comprehension and listening comprehension was stable over time. An additional hierarchical
regression analysis was carried out in order to compute
the variance explained by the two predictors as they
were evaluated at Time 2. In this analysis, therefore,
reading comprehension at Time 2 was the dependent
variable whereas the predictors were, in addition to
reading comprehension at Time 1, reading speed and
listening comprehension at Time 2. The results,
summarized in table 5, indicate that beside the 54.7%
of variance explained by the autoregressor, word
reading speed at Time 2 explained 4% variance, a
proportion that was not significant. The important
result was that listening comprehension added 14.4%
variance in explaining reading comprehension, and this
proportion was significant. Since this proportion is
smaller than the proportion of variance explained by
listening comprehension at time 1 (see the previous
analysis and the table 4), it can be concluded that the
relation between listening and reading comprehension
is causal more than simply correlational.
Table 5.

Discussion
The aim of the current study was to examine, in a
group of individuals with Downs syndrome, the
development over a one-year period of reading
comprehension in relation to two component skills
that, according to the Simple View of Reading, are
necessary for reading comprehension: reading skills and
listening comprehension.
Before analysing these main findings, it is worth
considering the absence of any statistically significant
correlation between age, IQ and years of schooling, on
the one hand, and any measures adopted in the current
work, on the other hand. This lack of statistically
significant correlations, which has already been
documented in a number of previous studies on people
with Downs syndrome (Byrne et al. 2002, Roch and
Levorato 2009, Levorato et al. 2009), suggests that the
variability as to age, school experience and mental
development may not influence reading skills and
comprehension level, at least not directly and not to the
same extent as in typical development. In Italy, for part
of the time individuals with Downs syndrome attend
classes with their peers of similar chronological age; for
the rest of the time they receive individualized teaching
from a specially trained support teacher. Certainly, part

Regression analysis on the reading comprehension score at Time 2: R 2 5 0.761 [F(3, 9) 5 6.4, p < 0.05]

Predictors

R 2 change

Step 1: Reading comprehension (T1)

0.574

Step 2: Word reading speed (T2)

0.043b

Step 3: Listening comprehension (T1)

Notes: aF change(1, 8) 10.8, p , 0.01.

b
F change (1, 7) 0.786, not significant (n.s.).
c
F change (1, 6) 3.6, n.s.

0.144c

Reading comprehension (T1)

0.757

3.3

,0.01

Reading comprehension (T2)

Word reading speed (T2)

0.787
0.210

3.3
0.887

,0.05
n.s.

Reading comprehension (T2)

Word reading speed (T2)
Listening comprehension (T2)

0.542
0.100
0.476

2.3
0.476
1.9

n.s.
n.s.
n.s.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Reading comprehension in Downs syndrome

of the daily school activities consist of reading and
comprehending what is read. Although it is likely that
the contents of literacy instruction and the quality
of teaching itself influence reading and related skills, at
the moment we can only speculate that in the age range
considered in this study, and given the skills already
achieved by the participants, the quantity of instruction
received, not necessarily reflecting the quality of the
teaching received, does not have any statistical influence
on the skills investigated in the current study.
The main results obtained in the current study
confirm that reading comprehension is supported by
the two component skills that contribute to its
development in a complex way. Three main findings
emerged: (1) reading skills, on the one hand, and
comprehension, both listening and reading, on the
other hand, are independent; (2) the development of
reading comprehension is determined mainly by
listening comprehension, which in this study proved
to be very poor; and (3) participants improved their
performances in all the abilities considered, except for
listening comprehension which remained at the same
level after a one-year period.
Independence of reading skills and language
comprehension
The results show the independence of reading skills
from both listening and reading comprehension. This
result is consistent with the model of the Simple View
of Reading, according to which each of the two
components provides a unique and specific contribution to reading comprehension. In individuals with
Downs syndrome, reading skills and text comprehension seem to develop asynchronically, and differently
from children with typical development. This conclusion, which is coherent with previous research
(Byrne et al. 2002, Laws and Gunn 2002, Roch and
Levorato 2009), was supported by three findings in the
current study. First, the correlations between comprehension of a text, whether read or listened to by
participants, and reading skills were quite low. Second,
in comparison with what is expected in typically
developing children, individuals with Downs syndrome were more advanced in word recognition than in
reading and listening comprehension. Finally, reading
comprehension development over a one-year period
was only marginally affected by reading skills.
In typical development, the mutual independence
of reading skills and reading comprehension is
particularly apparent in children characterized by
dyslexia, on the one hand, and in children defined as
poor comprehenders, on the other hand (Cain and
Oakhill 2006). The latter, like the individuals with
Downs syndrome in the present study, show poor

9
reading comprehension and good reading skills.
For individuals with Downs syndrome, as well as
for poor comprehenders, it seems that fluent and
accurate word recognition is not per se a guarantee
of adequate comprehension, indicating that access to
the meaning of what is read is not as automatic as it
seems on the basis of fluency and accuracy in reading.
This description leaves the question of the nature of
the independence of reading skills from reading
comprehension unanswered. We can conclude that
the two abilities, reading and understanding,
develop independently and even asynchronically.
Presumably reading and understanding are not
independent as processes while occurring during text
reading: it might be hypothesized that there is some
point where the two processes interact and share the
results of their processing. We speculate that during text
comprehension, the activities of decoding and getting
access to the meaning occur in parallel and interact:
Word recognition is also driven by the construction of
the meaning of the previous word string (from meaning
to decoding); alternately, successful word recognition
triggers the access to word meaning (from decoding to
meaning). Studies using on-line procedures might shed
light on the interconnections between these two
processes.
Causal role of listening comprehension on reading
comprehension
The current study showed a close relationship between
listening and reading comprehension. This result was
also documented in a previous study (Roch and
Levorato 2009), but in the current study new evidence
was added, namely that a causal relationship exists
between the two: listening comprehension at Time 1
predicted reading comprehension one year later. This
result may be explained by the Simple View of Reading:
When reading skills are automatized, listening
comprehension becomes the stronger predictor of
reading comprehension (Catts et al. 2006). This strong
relationship, and the amount of shared variance
between the two, suggests the importance of the
comprehension process per se independently of how the
text is presented. This conclusion stems from our
regression analysis and in particular from the use of the
autoregressor: this method allowed us to control both
the correlation between reading comprehension with
itself, and the correlation between reading and listening
comprehension at Time 1. This analysis rules out
the intervention of mediating variables in the
relationship between listening comprehension at
Time 1 and reading comprehension one year later:
30% of shared variance can be attributed uniquely to
the level of text comprehension, namely to the processes

Maja Roch et al.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

10
which lead to the construction of the mental
representation of the text meaning.
Comprehension of a text, whether written or oral, is
a complex ability that involves both cognitive and
linguistic components, and occurs through various
processes that have been defined as lower and higher
level processes (Cain and Oakhill 2007). Among the
lower level processes there are the recognition of word
meanings and the construction of sentence meanings
through the use of morpho-syntactic knowledge. These
processes are necessary but not sufficient because the
linguistic information contained in the text has to be
integrated into a coherent semantic representation
through inferential processes and the use of previously
acquired world knowledge (Cain and Oakhill 2001).
All these processes are common to reading and listening
comprehension (Cain and Oakhill 2007). In a recent
study it was demonstrated that also in Downs
syndrome, both lower level abilities, such as receptive
vocabulary and sentence comprehension, and higher
level components, such as the ability to use the context,
accounted for individual differences in listening
comprehension (Levorato et al. 2009). What has still
to be established is which factors, among those that
underlie text comprehension in individuals with
Downs syndrome, are shared by reading and listening
comprehension, and which ones are, on the contrary,
specific to each modality of text understanding.
Speculatively, we could argue that short term and
working memory may, at least in part, account for the
unshared variance between listening and reading
comprehension, and, therefore, be modality dependent.
It is likely that in a reading comprehension task, where
linguistic information is available for the time necessary
to process it, short term memory plays a minor role; on
the other hand, listening comprehension might rely
more heavily on the ability to remember verbal
information, and may be, therefore, a cause for a
poorer performance in listening than in reading
comprehension. Future investigation should control
the effect of the modalitywritten versus oralon the
involvement of short term memory in text comprehension, in particular in Downs syndrome individuals
where short term memory represents an area of
particular weakness (Jarrold and Baddeley 1997).
The substantial influence of poor listening
comprehension explains, at least in part, the fact that
reading comprehension lags behind reading skills in
Downs syndrome. Listening comprehension is poor
and develops very slowly, and this might also slow
down the development of reading comprehension.
The automatization of word recognition is a necessary
prerequisite, but is not sufficient for the comprehension
of written texts. The strong relationship between
listening and reading comprehension, on one hand, and

the relative independence of reading skills from reading

comprehension, on the other, have important significance for intervention: Training based on abilities
related to comprehension of texts, whether written
or oral, might be as useful as training in reading skills.
Actually, the former has proven fruitful for children
with typical development who have poor text
comprehension abilities (Oakhill and Yuill 1996) and
therefore exhibit a similar profile to that of our
participants with Downs syndrome, for which the
efficacy of such trainings has not yet been established.
Asynchronies in improvements
One of the advantages of follow-up studies is that they
offer an opportunity to trace the improvements
occurring over a given time period. In the current
study the pattern of changes in reading comprehension
and related skills during the one-year period of the
study showed an improvement in all the abilities we
examined, with the exception of listening comprehension, which remained stable. Reading comprehension
showed some improvement although the effect size was
low (0.19). Reading skills showed improvement with
moderately high effect sizes (from 0.30 to 0.70), but
only non-word-reading accuracy showed an increase
that yielded statistical significance and a high effect size
(0.80). A slow improvement of reading and reading
related skills was documented also by previous
longitudinal studies (Byrne et al. 2002, Laws and
Gunn 2002), in which it was demonstrated that, while
reading skills show steady annual progress, two years are
necessary in order to observe substantial changes in
reading comprehension. The current evidence is in line
with this result.
We note that participants showed high performance
on word reading at Time 1. We cannot say whether they
would improve if the words were more difficult, or if the
performance we observed mirrors the reading automatization level that the participants are able to attain.
Actually, our participants, who did not perform at
ceiling on non-word reading, during the year of the
study, improved in accuracy. Decoding, namely the
reading of non-words, is considered an area of particular
weakness for individuals with Downs syndrome; this
difficulty is determined by very poor phonological
awareness skills (Roch and Jarrold 2008). Therefore, if
an improvement occurs in decoding during a one-year
period, it could be suggested that boosting this ability
and its underlying skills, may prove fruitful.
With regards to text comprehension, some
improvement in reading comprehension was documented, while listening comprehension remained
stable and poor, over a one-year period. Therefore,
it seems that the two abilities do not show the same

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

Reading comprehension in Downs syndrome

rate of improvement in the same period, although
they are highly inter-dependent. Tracing the developmental trajectories of reading and listening
comprehension could contribute to a better understanding of how their development differs diachronically. Thomas et al. (2009) argue that there are at
least three ways in which different improvements
might be explained. The first is that one ability
develops at a slower rate than the other; the second
possibility is that the two abilities follow different
developmental trajectories and therefore do not show
the same changes over the same period of time; the
third is that both a slower rate of development and
different developmental trajectories characterize the
two abilities. Future studies, including participants
within an appropriate age range, will have to test
these different hypotheses with regards to listening
and reading comprehension.

Conclusions
The current study shed light on important issues
related to reading comprehension and its components
in Downs syndrome. Nonetheless, the great portion
of unexplained variance in reading comprehension
suggests that while the Simple View of Reading
represents a valid theoretical model of reading
competence, it is not sufficient to capture the
complexity of the abilities and processes involved in
reading comprehension and literacy acquisition,
especially in atypical development. For instance, this
model does not take contextual and experience factors
into consideration, and does not give any suggestion
concerning the influence of the quantity and quality of
literacy teaching received. This matter is also neglected
by previous studies on reading development in
Downs syndrome, with the exception of very few
studies that show better reading skills in children with
Downs syndrome who attend mainstream schools
compared with those attending special schools
( Laws et al. 1995).
Finally, some limits to the present study must be
acknowledged, the most relevant being the number of
participants. Therefore, for the time being, our
conclusions should be treated with caution. Nonetheless, we are confident that the pattern of results
reported in the current work is quite reliable. In fact, the
level reached in each ability, the pattern of relationships,
the trend of improvements, and the portion of variance
explained in reading comprehension are consistent with
the results of a number of previous studies (Boudreau
2002, Byrne et al. 2002, Laws and Gunn 2002, Roch
and Levorato 2009, Verucci et al. 2006), which, overall,
support our conclusions.

11
Acknowledgements
The authors are grateful to all the individuals with Downs
syndrome who took part in this study; and to the schools for the
collaboration in this work. They would also like to thank the Editor
and two anonymous reviewers for their helpful suggestions.
This research was financed by a grant to the Chiara Levorato
(Grant Number PRIN 2005 11 2005119758_005).

Notes
1. When we speak of reading skills, we refer to the reading of
both words and non-words. To refer specifically to non-word
reading, we use the term decoding, whereas to refer to word
reading, we use the expression word recognition.
2. Participants are provided with the whole list of words and
non-words. Therefore, total reading time also includes the
latencies between one item and another.

References
ABBEDUTO, L., WARREN, S. F. and CONNERS, F. A., 2007,
Language development in Down syndrome: from the
prelinguistic period to the acquisition of literacy. Mental
Retardation and Developmental Disabilities, 13, 247261.
ARO, M. and WIMMER, H., 2003, Learning to read: English in
comparison to six more regular orthographies. Applied
Psycholinguistics, 24, 621 635.
BOUDREAU, D., 2002, Literacy skills in children and adolescents
with Down syndrome. Reading and Writing, 15, 497525.
BYRNE, A., MACDONALD, J. and BUCKLEY, S., 2002, Reading,
language and memory skills: a comparative longitudinal
study of children with Down syndrome and their mainstream peers. British Journal of Educational Psychology, 72,
513 529.
CAIN, K. and OAKHILL, J. V., 2001, Comprehension skill,
inference-making ability, and their relation to knowledge.
Memory and Cognition, 29, 850 859.
CAIN, K. and OAKHILL, J. V., 2006, Profiles of children with specific
reading comprehension difficulties. British Journal of
Educational Psychology, 76, 683 696.
CAIN, K. and OAKHILL, J. (Eds.), 2007, Childrens Comprehension
Problems in Oral and Written Language. A Cognitive
Perspective (New York: The Guilford Press).
CATTS, H. W., ADLOF, S. M. and WEISMER, S. E., 2006, Language
deficits in poor comprehenders: a case for the Simple view of
Reading. Journal of speech, Language and Hearing Research,
49, 278 293.
CHAPMAN, R. S., 1995, Language development in children
and adolescents with Down syndrome. In P. Fletcher and
B. Mac Whinney (eds), The Handbook of Child Language
(Oxford: Blackwell), pp. 641 663.
CORNOLDI, C. and COLPO, A., 1998, Prove di lettura MT per la
scuola elementare [MT Reading Tests for Elementary School]
(Florence: O. S. Organizzazioni Speciali).
CUPPLES, L. and IACONO, T., 2002, The efficacy of whole word
versus analytic reading instruction for children with Down
syndrome. Reading and Writing, 15, 549 574.
DE JONG, P. F. and VAN DER LEIJ, A., 2002, Effects of phonological
abilities and linguistic comprehension on the development of
reading. Scientific Studies of Reading, 6, 51 77.
DIAKIDOY, I. N., STYLIANOU, P., KAREFILLIDOU, C. and
PAPAGEORGIOU, P., 2005, The relationship between listening
and reading comprehension of different types of text at
increasing grade levels. Reading Psychology, 26, 55 80.

Int J Lang Commun Disord Downloaded from informahealthcare.com by University of Padova on 07/15/10
For personal use only.

12
FLORIT, E., LEVORATO, M. C. and ROCH, M., 2008, Verba Volant,
scripta manent. Cambiamenti evolutivi nella comprensione del
testo scritto e orale. [Verba Volant, scripta manent.
Developmental changes in listening and reading comprehension]. Giornale Italiano di Psicologia, 35, 639660.
GOUGH, P. B., HOOVER, W. A. and PETERSON, C. L., 1996, Some
observations on a Simple View of Reading. In C. Cornoldi
and J. Oakhill (eds), Reading Comprehension Difficulties.
Processes and Intervention (Mahwah, NJ: Lawrence Erlbaum
Associates), pp. 1 3.
GOUGH, P. B. and TUNMER, W., 1986, Decoding, reading and
reading disability. Remedial and Special Education, 7, 6 10.
HOOVER, W. A. and GOUGH, P. B., 1990, The Simple View of
Reading. Reading and Writing, 2, 127 160.
JARROLD, C. and BADDELEY, A. D., 1997, Short-term memory for
verbal and visuospatial information in Downs syndrome.
Cognitive Neuropsychiatry, 2, 101 122.
JOSHI, R. M. and AARON, P. G., 2000, The component model of
reading: Simple View of Reading made little more complex.
Reading Psychology, 21, 85 97.
KIRBY, R. J. and SAVAGE, R. S., 2008, Can the simple view deal with
the complexities of reading? Literacy, 42, 75 82.
LAWS, G., BUCKLEY, S. J., BIRD, G., MACDONALD, J. and
BROADLEY, I., 1995, The influence of reading instruction on
language and memory development in children with Downs
syndrome. Down Syndrome Research and Practice, 3, 59 64.
LAWS, G. and GUNN, D., 2002, Relationships between reading,
phonological skills and language development in individuals
with Down syndrome: a five year follow-up study. Reading
and Writing, 15, 527 548.
LEVORATO, M. C. and ROCH, M., 2007, TOR 3 8: Test di
valutazione della comprensione del testo orale [TOR 3 8: Test
for the Evaluation of Listening Text Comprehension]
(Florence: O. S. Organizzazioni Speciali).
LEVORATO, M. C., ROCH, M. and BELTRAME, R., 2009, Text
comprehension in Down syndrome: the role of lower and
higher level abilities. Clinical Linguistics and Phonetics, 23,
285 300.

Maja Roch et al.

MEGHERBI, H., SEIGNEURIC, A. and EHRLICH, M. F., 2006,
Reading comprehension in French 1st and 2nd grade
children: contribution of decoding and language comprehension. European Journal of Psychology of Education, 21,
135 147.
OAKHILL, J. and YUILL, N., 1996, Higher order factors in
comprehension disability: processes and remediation.
In C. Cornoldi and J. Oakhill (eds), Reading Comprehension
Difficulties. Processes and Intervention (Mahwah, NJ:
Lawrence Erlbaum Associates), pp. 69 92.
ORSINI, A., 1993, WISC-R. Contributo alla taratura italiana
[WISC-R. A Contribution to Italian Standardisation]
(Florence: O. S. Organizzazioni Speciali).
ROCH, M. and JARROLD, C., 2008, A comparison between word
and nonword reading in Down syndrome: the role of
phonological awareness. Journal of Communication Disorders,
41, 305 318.
ROCH, M. and LEVORATO, M. C., 2009, Simple View of Reading in
Down syndrome: the role of listening comprehension and
reading skills. International Journal of Communication
Disorders, 2, 206 223.
SARTORI, G., JOB, R. and TRESSOLDI, P., 1995, Batteria per la
valutazione della dislessia e della disortografia evolutiva
[Battery of Tests for the Evaluation of Developmental
Dyslexia and Dysorthography] (Florence: O. S. Organizzazioni Speciali).
SAVAGE, R., 2001, The Simple view of reading: some evidence and
possible implications. Educational Psychology in Practice, 17,
17 33.
THOMAS, M. S. C., ANNAZ, D., ANSARI, D., SCERIF, G.,
JARROLD, C. and KARMILOFF-SMITH, A., 2009, Using
developmental trajectories to understand developmental
disorders. Journal of Speech Language and Hearing Research,
52, 336 358.
VERUCCI, L., MENGHINI, D. and VICARI, S., 2006, Reading skills
and phonological awareness acquisition in Down syndrome.
Journal of Intellectual Disability Research, 50, 477 491.