Long-term effects of synthetic versus analytic phonics teaching on the reading and spelling ability of 10 year old boys and girls

Article (PDF Available)inReading and Writing 25(6) · July 2011with 3,423 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
DOI: 10.1007/s11145-011-9323-x
Cite this publication
Abstract
A comparison was made of 10-year-old boys and girls who had learnt to read by analytic or synthetic phonics methods as part of their early literacy programmes. The boys taught by the synthetic phonics method had better word reading than the girls in their classes, and their spelling and reading comprehension was as good. In contrast, with analytic phonics teaching, although the boys performed as well as the girls in word reading, they had inferior spelling and reading comprehension. Overall, the group taught by synthetic phonics had better word reading, spelling, and reading comprehension. There was no evidence that the synthetic phonics approach, which early on teaches children to blend letter sounds in order to read unfamiliar words, led to any impairment in the reading of irregular words.
Figures - uploaded by Sarah P. Mcgeown
Author content
All content in this area was uploaded by Sarah P. Mcgeown
Content may be subject to copyright.
Long-term effects of synthetic versus analytic phonics
teaching on the reading and spelling ability of 10 year
old boys and girls
Rhona S. Johnston
Sarah McGeown
Joyce E. Watson
Ó Springer Science+Business Media B.V. 2011
Abstract A comparison was made of 10-year-old boys and girls who had learnt to
read by analytic or synthetic phonics methods as part of their early literacy pro-
grammes. The boys taught by the synthetic phonics method had better word reading
than the girls in their classes, and their spelling and reading comprehension was as
good. In contrast, with analytic phonics teaching, although the boys performed as
well as the girls in word reading, they had inferior spelling and reading compre-
hension. Overall, the group taught by synthetic phonics had better word reading,
spelling, and reading comprehension. There was no evidence that the synthetic
phonics approach, which early on teaches children to blend letter sounds in order to
read unfamiliar words, led to any impairment in the reading of irregular words.
Keywords Synthetic phonics Analytic phonics Opaque orthography
The English spelling system has an opaque orthography; although it is an alphabetic
system, some spellings have inconsistent grapheme-phoneme connections, e.g.,
‘aisle’. This inconsistency in English spelling has led to models of adult reading such
as the dual route model, where the pathways envisaged for the reading of words with
irregular versus regular spelling-sound correspondences are seen as largely indepen-
dent (Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001). A substantial literature has
examined whether individuals take a phonological approach to reading English,
determined by whether their responses to irregular words are slower and less accurate
than to regular words; it has been found that these effects are shown in both children
and adults (e.g., Waters, Seidenberg, & Bruck, 1984). There is some evidence,
however, that boys take a more phonological approach to reading than girls.
R. S. Johnston (&) S. McGeown J. E. Watson
Department of Psychology, University of Hull, Hull HU6 7RX, England, UK
e-mail: R.S.Johnston@hull.ac.uk
123
Read Writ
DOI 10.1007/s11145-011-9323-x
Baron (1979) found, in a post-hoc analysis of a study of 9–10 year olds, that the boys
were slower to read lists of orthographically similar words where words with
inconsistent spelling-sound correspondences were included (e.g., maid, said)
compared to control lists (e.g., made, said). In a study of 7 year olds learning to
read by a non-phonic method, Thompson (1987) found that boys had inferior reading
of exception words (e.g., great), their performance on regular words being very
similar to that of girls. It is possible, therefore, that a phonic approach to teaching
reading would be especially beneficial for boys, who generally do not fare well in
international comparisons of reading attainment (e.g., Mullis, Martin, Kennedy, &
Foy, 2007), as it would suit their natural approach to word reading. On the other hand,
their problems with reading irregular words might be exacerbated by using a phonics
approach to reading, as it might increase their errors on these words.
Concerns about the irregularities in English orthography have led to the wide-spread
adoption of methods of teaching reading using whole word methods, and the use of
phonics approaches with a substantial whole-word element, such as analytic phonics.
Indeed, it has been proposed that rigorous early grapheme-phoneme based phonics
programmes, such as synthetic phonics ones, cannot be effective in English (Dombey,
2006). The so-called ‘reading wars’ have led to a number of reviews of the efficacy of
whole word versus phonic teaching methods in English speaking countries. Recently,
the US National Reading Panel (NRP) concluded that phonic teaching approaches
were more effective than whole word approaches (National Institute of Child Health
and Development (NICHD), 2000; Ehri, Nunes, Stahl, & Willows, 2001).
This paper is concerned with examining what type of phonics programme is
effective in English, and whether one approach is particularly effective for boys.
Many countries in Europe use a synthetic phonics approach (e.g., Austria, see
Feitelson, 1988), where children learn very early on how to blend letter sounds in
order to decode unfamiliar words. However, where phonics is taught in English
speaking countries, it generally starts with an analytic phonics approach, where
children initially learn to recognise words by sight, that is, it is a mixed methods
approach. Alongside this, children learn to recognise letter sounds at the beginning,
the end, and then the middle position of printed words. At this point, usually at the
end of the first year at school or at the start of the second, they may then be taught
how to decode printed words by blending the letter sounds all through the word
(usually after hearing how it is pronounced). Johnston and Watson (2004) found in
two studies (Experiments 1 and 2) that 5 year old children taught by this analytic
phonics method read and spelt less well than those taught by the synthetic phonics
method, even when speed of letter learning was equated. In following the children
from Experiment 1 through to the age of 11, Johnston and Watson (2005) found that
the synthetic phonics taught children gained in word reading ability relative to
chronological age year after year. This is in contrast to the NRP finding that in most
training studies the gains were lost a few years after the end of the programme
(NICHD, 2000). It is also of interest that children from areas of disadvantage
performed as well as those from advantaged areas until close to the end of primary
schooling, although other UK studies have shown them to fall behind right from the
start of schooling (Duncan & Seymour, 2000; Stuart, Dixon, Masterson, & Quinlan,
1998). Finally, it was found that from the end of the third year of school the boys
R. S. Johnston et al.
123
had better word reading than the girls (Johnston & Watson, 2005), and were still
better at the end of the study.
The analytic phonics approach used in Johnston and Watson’s (2004) studies was
taken from the approach widely used in Scottish schools (Watson, 1998). It is also
very similar to the approach that was used until recently in England (Progression in
Phonics, DfEE, 1999). In the larger of their two studies (Johnston & Watson, 2004,
Experiment 1), the children taught by the analytic phonics method switched to the
synthetic phonics method as soon as the first post-test was carried out, as the benefits
of the synthetic phonics method were clearly evident to the education authority.
Therefore, it was unclear from this particular study whether children taught by the
analytic phonics method would ultimately perform as well as those taught by a
synthetic phonics approach. On the other hand, if the early use of a grapheme to
phoneme conversion approach is problematical because of the English spelling
system, then it is also possible that the synthetic phonics children would fall behind
their analytic phonics counterparts, or at least have specific problems with reading
irregularly spelt words (Dombey, 2006). Such a view fits with the dual route model of
reading, but not with one route connectionist models (e.g., Seidenberg & McClelland,
1989), which emphasise that even irregular words such as ‘yacht’ contain information
about pronunciation. Although Harm and Seidenberg (2004) have developed a two-
pathway connectionist model (the orthographic-phonological-semantic and the
orthographic-semantic pathways), irregular words are not said to be processed purely
by the latter, more visual orthographic, pathway, as is proposed by the dual route
model. As connectionist models propose that a reader may capitalise on the fact that
irregular words contain regular elements that will assist pronunciation, it can be
hypothesised that the synthetic phonic approach to reading may not lead to so severe
an impairment in reading irregular words as that predicted by the dual route model.
A number of cross-national studies have suggested that learning to read in an
opaque language such as English has costs for attainment. For example, Wimmer
(1995) reported that 7 year old Austrian children who learnt to read in a transparent
orthography (German) had better pseudoword reading skills after 1 year at school
than 9 year old English children after around 4 years at school, which might suggest
quite a penalty for having an opaque orthography. However, Wimmer also pointed
out that the two groups also differed according to teaching method, the Austrian
group following a rigorous synthetic phonics programme, and the English children
learning by an eclectic approach including a whole word ‘look say’ approach and a
phonics programme that used a ‘word families’ approach (typical of analytic
phonics teaching). He suggested that the synthetic phonics teaching approach may
have contributed to the better phonological recoding skills of the Austrian children,
in addition to the benefits of a consistent orthography, and suggests that it would be
interesting to explore how well such an approach would work in an opaque
orthography such as English. Although Johnston and Watson (2004) have shown
that a rigorous synthetic phonics teaching programme works well in English, there
is a need for evidence as to whether children taught this way outperform, in the
longer term, children who receive analytic phonics teaching.
Johnston and Watson’s (2004) studies were primarily concerned with the
development of word reading and spelling skills, but the question arises as to what
Long-term effects of synthetic versus analytic phonics
123
impact synthetic phonics teaching has on reading comprehension. Stannard (2006)
has argued that the Clackmannanshire Study (Johnston & Watson, 2004, Exper-
iment 1) showed that good phonics teaching delivered good word reading but had
little impact on reading comprehension. As the analytic phonics groups also carried
out the synthetic phonics programme before the end of their first year at school, no
comparative data are available on this issue, even though the sample was studied for
a further 6 years. However, it was found that the synthetic phonics taught children
had reading comprehension skills significantly ahead of what was expected for their
chronological age right through to the end of the study, although the gains were
smaller than those for word reading and spelling (Johnston & Watson, 2005).
According to the Simple View of Reading, boosting decoding or word reading skills
may by itself lead to an increase in reading comprehension (Gough & Tunmer,
1986). If the synthetic phonics taught children were found to have superior word
reading skills, this raises the issue as to whether they would also have better reading
comprehension skills than children taught by the analytic phonics approach.
The aim of this study was to compare the literacy skills of boys and girls from the
synthetic phonics taught Clackmannanshire sample (Johnston & Watson, 2004,
Experiment 1) at the age of 10 with boys and girls who had learnt by an analytic
phonics approach. This study capitalises on the fact that the synthetic phonics group were
virtually all non-readers at the start of the study; this meant that they could be matched
for time at school and socio-economic background to a group taught by analytic phonics
who, if they differed at all in pre-tuition reading skills, could only be at an advantage.
Firstly, it was hypothesised that if the synthetic phonics method is unsuited to the
irregularities of the English language, then children learning to read by this method
would have lower levels of word reading ability compared with the mixed-method
analytic phonics approach; alternatively, it was predicted that because even irregular
words contain some letter sounds that give a guide to pronunciation, those taught by the
synthetic phonics method would have better word reading skills. It was also hypothesised
that if the irregularities in English spelling are problematical for a synthetic phonics
approach, then reading exception and strange words would be particularly problematical
for boys, for whom there is evidence that they take a more phonological approach to
reading. If so, they might have more difficulty in reading irregular words than boys
taught by a mixed-method analytic phonics approach, whereas girls might be less
affected by teaching method. Finally, it was hypothesised that whichever group had an
advantage in word reading would also have better reading comprehension skills.
Study 1: comparison of synthetic phonics and analytic phonics taught children
on word reading, spelling and reading comprehension
Method
Participants
Data from the Clackmannanshire Study, which was carried out in Scotland, were
compared with data from schools in England. It was not possible to collect
R. S. Johnston et al.
123
comparative data in Scotland as the influence of the study there was such that many
schools had adopted early sounding and blending, and this was directly confirmed in
responses to questionnaires by teachers from outside the region.
Starting with the Primary 6 data collected in Clackmannanshire in Scotland when
the children were 10 years old (Johnston & Watson, 2005), a subset of the original
synthetic phonics sample (n = 190) was matched on time at school and
socioeconomic background with a sample (n = 203) from a city in England taught
by Progression in Phonics (Department for Education and Employment (DfEE),
1999), a mixed method approach that includes analytic phonics. Very few of the
children in this Clackmannanshire subsample could read or spell at the start of
schooling in August 1997; only five children gained a reading score, and only two
gained a spelling score. This was because in Scotland at that time, literacy skills
(i.e., letter sounds and word reading) were not taught in Nursery schools.
The synthetic phonics children were tested in March of their sixth year at school,
but the analytic phonics sample was tested in June of their sixth year, or November
of their seventh year, to equate for time at school (in the latter area, children
attended school for between one and three terms in their first year of school,
whereas in Scotland all of the children attended school for three terms in their first
year). A term is on average around 64 school days. Whilst time at school was
equated across groups, the children in the analytic phonics sample were on average
3 months younger; this difference was dealt with by comparing literacy skills using
age-standardised scores. The samples were also matched on socioeconomic status
and were categorised as moderately advantaged, moderately disadvantaged,
disadvantaged and greatly disadvantaged using Clackmannanshire Council’s Index
of Disadvantage (synthetic phonics sample) and the Department for Education and
Skills (DfES) Panda system (analytic phonics sample).
The schools for the analytic phonics sample were matched to the schools in
Clackmannanshire. Although there were no data available on the performance of this
sample before reading tuition started in primary school, literacy skills could scarcely
be lower than those of the synthetic phonics sample. The schools were matched on
SES as follows. In the Synthetic Phonics (SP) condition, seven schools were involved,
contributing 11 classes to the sample. In the Analytic Phonics (AP) condition, there
were eight schools, contributing one class each to the sample. In order to effect a
match on SES, the two samples were stratified by bands of SES; whole classes were
selected, and all pupils were included. This led to the following match: very low SES
(SP, n = 34; AP, n = 29), low SES (SP n = 33; AP n = 28), moderately low SES
(SP n = 20 AP n = 38), and medium to moderately high SES (SP n = 103, AP
n = 108). The greater numbers for moderately low SES AP taught children did not
skew the mean scores on the WRAT downwards for the analytic phonics sample, as
these 38 children had a mean WRAT of 99.68, whereas the means for the medium to
moderately high SES classes were lower in all except one case (95.79, 97.78, 99.15,
and 101.17). School inspections carried out around the time of testing show that all of
the English schools were considered satisfactory; this assessment takes into account
performance on tests carried out in all schools in England for English and Math.
The analytic phonics sample was composed of 46.6% from moderately to
severely disadvantaged areas and 53.4% from moderately advantaged areas.
Long-term effects of synthetic versus analytic phonics
123
The synthetic phonics sample was composed of 45.6% from moderate to severely
disadvantaged areas and 54.4% from moderately advantaged areas. In terms of
percentage of free schools meals, the resulting groups were well matched. For the
advantaged groups, the figures were 14.5 and 14.2% for the synthetic phonics and
analytic phonics samples, respectively. For the disadvantaged groups, the figures
were 41.6 and 45.25% for the synthetic phonics and analytic phonics samples,
respectively.
Teaching methods
Both groups of children had learnt to read by a phonic method early on in their
schooling, and ultimately covered the same ground. However, these phonic methods
differed in important respects. The children in England (mixed-method analytic
phonics group) had learnt to read by the National Literacy Strategy programme
Progression in Phonics (DfEE, 1999), which provides considerable detail on how
the programme should be delivered. It is designed for use with children in the first
3 years of school. Children are initially taught phonological awareness, that is, they
are trained to hear rhymes and phonemes in spoken words. Following this, they are
taught letter sounds at the beginning of words, then at the end, and then in the
middle. This stage is usually reached at the end of the first year of school. After this
point, children see printed words, hear them spoken, and then sound and blend them.
This phonics programme closely resembles what is done in traditional analytic
phonics programmes (Harris & Smith, 1976; Johnston & Watson, 2007, Chap. 1),
the only difference being that the Progression in Phonics (PiPs) children are also
taught to segment spoken words for spelling. The children proceed to learning to
read and spell words with consonant digraphs, (e.g.,
thin), initial and final consonant
blends (e.g.,
swim, tent), vowel digraphs (e.g., coat), and split digraphs (e.g., cake).
At this stage, children may work with words with similar rime spellings.
Throughout, children are encouraged to guess unfamiliar words from context, and
so sounding and blending is not the primary approach to identifying unfamiliar
words. High frequency words in general, and irregular words in particular, were
taught by sight without phonic analysis. In the city in which the study was carried
out, literacy consultants were trained and paid for by the central government, to give
in-service teaching in the method and to monitor the satisfactory teaching of this
programme, and for literacy throughout the primary school years. Observations by
the authors in classes throughout the city showed that the programme was uniformly
being observed.
There has been some confusion about what constitutes a synthetic phonics
programme. For example, Brooks (2003) has argued that PiPs should be categorised
as a synthetic phonics programme, on the grounds that it contains some sounding
and blending. This is a statement that needs some consideration, as it underlines the
fact that there are subtle but important differences between different types of
phonics programmes. Firstly, in synthetic phonics programmes, sounding and
blending is introduced at the beginning of reading tuition, whereas in PiPs it is
introduced towards the end of the first year at school, or even later. Thus children
spend most of their first year at school reading words largely by sight; even when
R. S. Johnston et al.
123
introduced, sounding and blending is not the predominant approach to learning to
recognise words, as children are taught to guess unknown words from context.
Secondly, in PiPs children are told how the words are pronounced before sounding
and blending them, which means that they do not have to synthesise the sounds in
order to pronounce the words; studying the letter sounds in known words is a core
feature of the analytic phonics approach. Interestingly, in 2002, Brooks correctly
described the synthetic phonics method as being where the child sounds and blends
the letters in unknown printed words, in order to discover how to pronounce them,
so he has changed his position on this issue.
The children in the Clackmannanshire sub-sample learnt to read by the synthetic
phonics approach in their first year of school, and also learnt how to segment spoken
words for spelling (Johnston & Watson, 2004, Experiment 1). In synthetic phonics
programmes, at the start of reading tuition children learn a few letter sounds, e.g.,
‘s’, ‘a’, ‘t’, ‘p’ (see Johnston & Watson, 2007) and then see whole words made up
from those letters, e.g., tap, pat, and sat. They are not told what these words are,
however, but have to sound and blend the letter-sound sequences to read the words
independently. They are also not taught to guess unfamiliar words from context. The
class synthetic phonics programmes in Clackmannanshire were closely monitored
for the 16 weeks of the experimental programme by the region’s Senior Quality
Assurance Officer. The programme was handed out on a weekly basis in order to
keep up the momentum; the quality assurance officer frequently monitored the class
teaching of the programme. Some teachers were unhappy with the method,
particularly expressing concern about the speed of learning, but the quality
assurance officer ensured that they complied with the method. The Clackmannan-
shire teachers were therefore less familiar with the programme they implemented
than the teachers in England, having had only half a day in-service training before
the study commenced, whereas in England, the government scheme Progression in
Phonics, had been in place for several years. Therefore, given the lack of experience
of the teachers in Clackmannanshire, one might predict a less efficacious
implementation of the method, and therefore poorer results. For the second year
of school, the teachers in the region were given an outline of the phonics to teach,
covering more complex digraphs, syllables, and morphemes, but this teaching was
not closely monitored.
The two phonics programmes ultimately covered the same ground in the teaching
of orthographic patterns. However, the phonics programme in Clackmannanshire
was largely completed by the end of the second year of school, whereas England’s
Progression in Phonics (DfEE, 1999) programme was largely completed by the end
of the third year of school.
There was considerable similarity in the rest of the reading curriculum, both
schemes fostering the ability to read for information and for pleasure (in Scotland,
Scottish Executive Education Department (SEED), 2000; in England, DfEE, 1998).
In Scotland, according to the 5–14 Guidelines (SEED), by Primary 6 children
should reach Level C in literacy attainment. When reading for information, they
should be able to read a variety of texts, and in discussion and writing show that
they understand the main and supporting ideas, and draw conclusions from the text
where appropriate. They should also be able to find and use information from a
Long-term effects of synthetic versus analytic phonics
123
range of sources. Similarly, in England (DfEE, 1998) a progression is outlined such
that children by the end of Year 5 (i.e., sixth year at school) should be able to read
information passages and identify the main points of the text, summarise a sentence
or paragraph by identifying the most important elements and rewording them in a
limited number of words, and read and evaluate a range of instructional text in terms
of their purposes, organisation, layout, clarity and usefulness. Both schemes also
develop children’s reading for enjoyment, fostering their awareness of genre and of
the ideas and the techniques used by authors.
The children in Clackmannanshire had undertaken a much more accelerated
reading and spelling programme than had hitherto been the case, and Primary 2
teachers were faced with classes that were very much further advanced than they
had been used to. Although the intervention had ended, advice was given on further
aspects of phonics for reading and spelling that should be covered. The teachers also
spent time on developing the children’s reading comprehension skills. The teachers
were additionally encouraged to spend some time on developing thinking skills. In
order to assist the teachers with this, they were offered a programme lasting for one
half hour session per week. For example, using the stimulus of a picture, the teacher
would ask open-ended questions and invite oral responses. The children would also
be introduced to the idea of defining, classifying and comparing; later on they would
also cover ambiguity, inference, and comparison. Towards the end of Primary 2 the
children would be encouraged to answer questions with questions, using paragraphs
of printed text as stimuli. These sessions continued in Primary 3, where children
were introduced to discussing in pairs concepts such as being brave, the meaning of
precious, or being poor, again using text as the stimuli. By the end of the year, they
would be matching paragraphs to relevant cartoon frames, as well as continuing to
develop their discussion skills. These half-hour per week programmes were not
separately monitored by the authors, nor were any outcomes assessed. Feedback
suggested that this area of work was not popular with the teachers.
Materials
The following tests were used with the children at the age of 10. Word reading was
tested using the WRAT Reading Test (Wilkinson, 1993). This test was adopted in
the 6th year of the Clackmannanshire study as so many children had reached ceiling
on the British Ability Scales Word Reading Test (Elliott et al., 1977), which has an
upper reading age of 14 years and 5 months. The BAS test contains both regular and
irregular words. In the sixth year, both the BAS and WRAT tests were administered;
the results were very comparable when children at ceiling on the BAS were
excluded. Reading comprehension was tested using the Group Reading Test
(Macmillan Unit, 2000a); this is a cloze procedure test, but for 10 year olds it has
been found to correlate well with the Neale Analysis of Reading ability (Neale,
1989), a test of passage comprehension (r = 0.76 on a sample of 54 10 year old
children; McGeown, unpublished results). The GRT II test manual also shows that
the Group Reading Test has very good reliability and validity (K-R 21 is .88 for
Form C and .84 for Form D). See manual guidelines for the GRT II for further
details of reliability and validity (Macmillan Unit, 2000b). Spelling was tested using
R. S. Johnston et al.
123
the Schonell Spelling Test (Schonell & Schonell, 1952), and vocabulary knowledge
was assessed by the English Picture Vocabulary Test (Brimer & Dunn, 1984).
Results
Vocabulary knowledge
In order to assess whether the two samples were matched on verbal ability, an
analysis was made of vocabulary knowledge using the English Picture Vocabulary
Test (Brimer & Dunn, 1984) (see Table 1 for means and standard deviations). A
three-way analysis of variance was carried out, with type of teaching (analytic
versus synthetic phonics), level of disadvantage (disadvantaged versus advantaged),
and sex (boys versus girls) as the between subjects factors. There was no main effect
of type of teaching, F(1, 385) = 1.80, p [ .05, (M = 91.2, SD = 12.0 for synthetic
phonics, M = 89.4, SD = 11.3 for analytic phonics). However, there was a main
effect of sex, F(1, 385) = 10.10, p \ .002, g
p
2
= 0.03, with males gaining higher
scores than females (M = 91.9, SD = 11.6 for boys, M = 88.2, SD = 11.5 for
girls). There was also a main effect of disadvantage, F(1, 385) = 12.40, p \ .001,
g
p
2
= 0.03, with children from advantaged areas performing better (M = 92.1,
SD = 11.0 for advantaged areas, M = 88.1, SD = 12.1 for disadvantaged areas).
There were no significant interactions between any of these factors, F \ 1 in all
cases except for type of teaching and sex, F(1, 385) = 1.53, p [ .05. Thus it is clear
that this test was sensitive to the indices of disadvantage used, and that the samples
were well matched on this variable.
Word reading
A three-way analysis of variance was carried out, with type of teaching (analytic
versus synthetic phonics), level of disadvantage (disadvantaged versus advantaged),
and sex (boys versus girls) as the between subjects factors (see Table 1 for means
and standard deviations). There was no main effect of level of disadvantage, F(1,
385) = 2.37, p [ .05 (M = 104.0, SD = 13.9 for advantaged areas, M = 101.9,
SD = 15.3 for disadvantaged areas). The main effect of sex just failed to reach
significance, F(1, 385) = 3.54, p = .061, (M = 104.1, SD = 15.1 for boys,
M = 101.5, SD = 13.9 for girls). There was a main effect of type of teaching,
F(1, 385) = 46.95, p \ .001, g
p
2
= .11, performance being better in the synthetic
phonics group (M = 108.2, SD = 13.9 for synthetic phonics, M = 98.1, SD = 13.6
for analytic phonics). There were no interactions between levels of disadvantage
and sex, F(1, 385) \ 1, and levels of disadvantage and type of teaching, F(1,
385) \ 1. However, there was a two-way interaction between sex and type of
teaching, F(1, 385) = 4.29, p \ 0 .04, g
p
2
= .01. Newman Keuls tests showed that,
regardless of sex, the synthetic phonics group read better than analytic phonics
group (p \ .01 in both cases). Synthetic phonics boys read better than the girls in
their classes (p \ .01), whereas there was no sex difference with analytic phonics.
There was no three-way interaction between sex, levels of disadvantage, and type of
teaching, F(1, 385) \ 1.
Long-term effects of synthetic versus analytic phonics
123
Reading comprehension
A three-way analysis of variance was carried out, with type of teaching (analytic
versus synthetic phonics), level of disadvantage (disadvantaged versus advantaged),
and sex (boys versus girls) as the between subjects factors (see Table 1 for means
and standard deviations). There was a main effect of levels of disadvantage, F(1,
385) = 14.59, p \.001, g
p
2
= .04. The children in advantaged areas performed
better (M = 99.8, SD = 11.9 for advantaged areas, M = 95.4, SD = 12.9 for
disadvantaged areas). There was no main effect of sex, F(1, 385) \ 1, (M = 97.5,
SD = 13.5 for boys, M = 97.9, SD = 11.4 for girls), but there was a main effect of
the type of teaching, F(1, 385) = 10.37, p \ .001, g
p
2
= .03, performance being
better with synthetic phonics (M = 100.0, SD = 11.7 for synthetic phonics,
M = 95.7, SD = 13.0 for analytic phonics). Type of teaching interacted with sex,
F(1, 385) = 8.01, p \ .005, g
p
2
= .02. Newman Keuls tests showed that girls
comprehended equally well regardless of teaching method, but that boys did better
if taught by synthetic phonics; it was also the case that boys taught by analytic
phonics had poorer reading comprehension than girls (p \ .01). There was no sex
difference with synthetic phonics teaching, but with analytic phonics teaching girls
had better reading comprehension (p \ .05). There was no interaction between
levels of disadvantage and sex, F(1, 385) \ 1, and no interaction between levels of
disadvantage and type of teaching, F(1, 385) = 1 .06, p [ .05. There was no three-
way interaction between levels of disadvantage, sex and type of teaching, F(1,
385) \ 1.
Table 1 Study 1: mean standardised scores on vocabulary knowledge, word reading, reading compre-
hension and spelling, comparing teaching method, boys versus girls, and advantage versus disadvantage
N Age Vocabulary
knowledge
Word reading Reading comp Spelling
Analytic phonics
Advantaged
Boys 64 10.49 (0.34) 92.20 (10.83) 97.84 (12.66) 95.73 (12.59) 97.12 (14.83)
Girls 44 10.49 (0.32) 89.57 (11.79) 99.25 (11.56) 101.50 (12.18) 101.27 (14.36)
Disadvantaged
Boys 45 10.43 (0.35) 88.24 (11.91) 98.16 (15.96) 91.44 (13.79) 91.46 (15.70)
Girls 50 10.51 (0.35) 86.34 (10.52) 97.28 (14.27) 93.78 (11.75) 95.00 (15.99)
Synthetic phonics
Advantaged
Boys 53 10.72 (0.28) 95.17 (10.48) 112.72 (14.24) 102.51 (12.10) 108.34 (12.95)
Girls 50 10.72 (0.29) 90.78 (10.44) 106.14 (11.63) 100.16 (9.32) 104.93 (10.53)
Disadvantaged
Boys 53 10.78 (0.31) 91.30 (12.56) 108.19 (12.39) 99.66 (13.40) 103.14 (11.79)
Girls 34 10.69 (0.31) 85.38 (13.30) 103.74 (16.84) 96.09 (10.55) 100.42 (11.76)
R. S. Johnston et al.
123
Spelling
A three-way analysis of variance was carried out, with type of teaching (analytic
versus synthetic phonics), level of disadvantage (disadvantaged versus advantaged),
and sex (boys versus girls) as the between subjects factors (see Table 1 for means
and standard deviations). There was a main effect of levels of disadvantage, F(1,
385) = 14.93, p \ .001, g
p
2
= .04 (M = 102.8, SD = 14.0 for advantaged areas,
M = 97.5, SD = 14.70 for disadvantaged areas). There was also a main effect of
type of teaching, F(1, 385) = 32.58, p \ .001, g
p
2
= .08, with better spelling in the
synthetic phonics group (M = 104.6, SD = 12.0 for synthetic phonics, M = 96.4,
SD = 15.6 for analytic phonics). There was no main effect of sex, F(1, 385) \ 1,
(M = 100.2, SD = 15.1 for boys, M = 100.4, SD = 13.8 for girls), but sex
interacted with type of teaching, F(1, 385) = 6.09, p \ .02, g
p
2
= .02. Newman
Keuls tests showed that both boys (p \ .01) and girls (p \ .05) spelt better with
synthetic than analytic phonics teaching. Girls spelt better than boys with analytic
phonics (p \ .05), but there was a non-significant trend towards boys spelling better
than girls with synthetic phonics teaching. There were no interactions between level
of disadvantage and sex, F(1, 385) \ 1, level of disadvantage and type of teaching,
F(1, 385) \ 1, or level of disadvantage, type of teaching, and sex, F(1, 385) \ 1.
Study 2: comparison of regular and irregular word reading in analytic
versus synthetic phonics classes
Although the word reading test used in Study 1 was composed of both regular and
irregular words, it is possible that the gains found for word reading with synthetic
phonics teaching were due to an enhanced ability to read only the regular words.
Study 2 was designed to examine the reading of regular and irregular words.
Method
Participants
Sixty-four children from two classes, one taught by analytic phonics and one taught
by synthetic phonics, took part in this study. The analytic phonics taught class was
one of the classes included in Study 1; there were 33 (20 male) children (M = 10;6,
SD = .40) in this class. The school was in a moderately high SES area; in 2005,
85% of the pupils met the required standard in English, where the average for the
region was 73%. The synthetic phonics taught class was taken from a school
matched on socioeconomic status to the analytic phonics class. The children were
not in the original experimental study (Johnston and Watson, 2004, Experiment 1),
having started school a year later. There were 31 (14 male) children in this class
(M = 10; 8, SD = .28).
Long-term effects of synthetic versus analytic phonics
123
Materials
Word reading, reading comprehension and vocabulary knowledge were tested using
the same tests as in Study 1.
Regularity test
High and low frequency regular (e.g., back, ramp), strange (e.g., eight, ache) and
exception words (e.g., bear, bald) were presented individually on a computer in a
quasi-random order. In total there were 95 words (5 practice words and 15 examples
of each word type). All children were tested individually and were instructed to
pronounce each word as accurately and quickly as possible.
Results
Word reading, reading comprehension, and vocabulary knowledge
Two way analyses of variance, with two between-subjects factors, teaching
programme and sex, were carried out. The synthetic phonics taught children had
better word reading than the analytic phonics group; F(1, 67) = 5.96, p \ .02,
g
p
2
= .08, and there were no sex differences, F(1, 67) = 1.09, p [ .05. They also
had better reading comprehension, F(1, 66) = 16.22, p \ .001, g
p
2
= .20, with no
sex differences being found, F(1, 66) = 2.34, p [ .05. Finally, they also had better
vocabulary knowledge; F(1, 62) = 14.75, p \.001, g
p
2
= .21, with no sex
differences being found, F(1, 62) \ 1.
Regularity task
A29 3 9 2 9 2 (frequency 9 regularity 9 teaching programme 9 sex) analysis
of variance was carried out on the accuracy data, see Table 2 for means and
standard deviations. There was a main effect of frequency, F(1, 60) = 146.91,
p \ .001, g
p
2
= .71, with high frequency words being read better than low frequency
words. In addition, there was a main effect of teaching programme, F(1, 60) = 4.91,
p \ .03, g
p
2
= .076, favouring the synthetic phonics group. There was also a main
effect of regularity, F(2, 120) = 111.81, p \ .001, g
p
2
= .65, but there was an
interaction between frequency and regularity, F(2, 120) = 120.92, p \ .001,
g
p
2
= .67; Newman Keuls tests showed that there was no regularity effect for high
frequency words, but for low frequency items, regular words were read better than
exception and strange words, and exception words were read better than strange
words (p \ .01 in all cases). There was an interaction between teaching programme
and frequency, F(1, 60) = 7.14, p \ .01, g
p
2
= .11; Newman Keuls tests showed
that the synthetic phonics group read low frequency words better than the analytic
phonics group (p \ .01). There was no interaction between teaching programme and
regularity, F(2, 120) = 1.93, p [ .05. There was no sex difference, F(1, 60) [ 1,
and there were no interactions with sex: regularity by sex, F(2, 120) = 1.51,
p [ .05, frequency by sex, F(1, 60) = 1.56, p [ .05, frequency by regularity by sex,
R. S. Johnston et al.
123
Table 2 Study 2: mean word reading, reading comprehension and vocabulary standardised scores; mean accuracy (out of 15) of reading high and low frequency regular,
strange and exception words (standard deviations in brackets)
Word reading Reading
comprehension
Vocabulary
knowledge
High
frequency
regular
High frequency
exception
High
frequency
strange
Low frequency
regular
Low frequency
exception
Low frequency
strange
Synthetic phonics
Boys 104.43 (14.08) 105.79 (14.15) 102.14 (12.94) 15.00 (0.00) 14.64 (0.75) 14.57 (0.51) 14.71 (0.61) 11.50 (1.79) 9.86 (3.59)
Girls 109.12 (11.37) 109.12 (9.70) 97.71 (13.32) 14.88 (0.33) 14.47 (0.80) 14.59 (1.00) 14.65 (0.79) 12.12 (2.29) 10.94 (3.67)
Analytic phonics
Boys 97.44 (14.27) 92.60 (11.27) 89.35 (8.96) 14.80 (0.52) 14.45 (0.83) 14.55 (0.69) 14.05 (1.61) 9.95 (2.80) 7.60 (3.79)
Girls 99.73 (15.04) 98.36 (13.86) 86.67 (12.35) 14.85 (0.38) 14.46 (0.88) 14.38 (0.96) 13.85 (1.63) 10.69 (3.50) 8.85 (4.36)
Long-term effects of synthetic versus analytic phonics
123
F(2, 120) = 2.53, p [ .05. Finally, there was no interaction between frequency,
regularity and teaching programme, F(2, 120) = 2.89, p [ .05, or between
frequency, regularity, teaching programme and sex, F(2, 120) [ 1. An analysis of
covariance controlling for the differences in overall word reading and ability and
vocabulary knowledge between the groups removed the main effect of teaching
programme, F(1, 54) = 1.63, p [ .05, and the interaction between teaching
programme and frequency was no longer significant, F(1, 54) = 1.97, p [ .05.
Discussion
It was found in Study 1 that, after 6 years at school, children taught by the synthetic
phonics approach read words, spelt words and had reading comprehension skills
significantly in advance of those taught by the analytic phonics method. This shows
that despite English being an opaque orthography, children are not impaired when
taught by an approach to reading that is common in transparent orthographies.
However, interactions were found between teaching methods and sex. Boys
benefited the most from synthetic phonics teaching, as they had word reading scores
better than those of the girls in their classes, and had equivalent spelling and reading
comprehension (in the latter case, contrary to the findings of international surveys
e.g., Mullis et al., 2007). However, the analytic phonics taught boys had the
typically observed pattern of inferior performance compared with the girls in their
classes in all except word reading. Interestingly, the synthetic and analytic phonics
taught girls had equivalent reading comprehension scores, although the former
group had better word reading (and spelling) skill. Although children from areas of
disadvantage had lower levels of reading comprehension and spelling than those
from advantaged areas, it was found that word reading ability was not affected by
differences in socio-economic background. In Study 2, the synthetic phonics taught
group showed no impairment in reading irregular words compared with the analytic
phonics taught sample, and boys did not make significantly more errors on irregular
words than girls.
The analytic phonics group in Study1 did not have low levels of word reading
ability, mean performance on the WRAT being 98.1, despite the fact that 46.6% of
the sample came from areas of moderate to severe socio-economic disadvantage.
The significant group difference in word reading ability reflects the fact that the
synthetic phonics group had a mean score of 108.2 on the WRAT. There is no
indication, therefore, that the analytic phonics children were selected from schools
having inadequate teaching standards. Furthermore, the groups were well matched
in verbal ability, and had a similar SES profile. The evidence rather supports the
view that synthetic phonics teaching led to above average levels of word reading
ability; the effect size comparing the two group, using Cohen’s d, was quite large at
0.73.
In the Clackmannanshire Study (Johnston & Watson, 2004, Experiment 1), at the
age of 5 the synthetic and analytic phonics groups were well matched at pre-test on
literacy measures, but as the former group came from a much lower SES
background lower levels of attainment would be expected (Duncan & Seymour,
R. S. Johnston et al.
123
2000; Stuart et al., 1998). At the end of the 16 week programme, however, the
synthetic phonics group’s word reading was 7 months ahead of that of analytic
phonics taught children, who were reading appropriately for their chronological age
(the effect size between the two groups being 0.91). Subsequently, the total sample
of synthetic phonics taught children (including those initially taught by the analytic
phonics method) showed increasing gains for word reading over age, ending up
reading 3.6 years ahead of chronological age at the age of 11 (Johnston & Watson,
2005). For the girls, the effect size for word reading age versus chronological age
was 1.12 at the end of the second year of school, and 1.36 at the end of the seventh
year of school; the boys were on a steeper trajectory, the effect sizes being 1.24 and
1.71, respectively, reflecting the fact that from the third year of school their word
reading was significantly better than that of the girls.
The increasing gains in word reading ability for the synthetic phonics group
needs to be viewed in the light of evidence that, in general, gains with phonics
programmes diminish over time. Ehri et al.’s (2001) meta-analysis showed that
treatment gains for phonics over non-phonics programmes declined from a
moderate effect size of 0.51 at immediate post-test to a small effect size of 0.27
at follow up. However, a study by Torgesen et al. (1999) showed increasing gains
over time with synthetic phonics teaching. Groups of children at risk of reading
failure were taught by a synthetic phonics approach and compared with those taught
by embedded phonics (which more closely resembles analytic phonics). In
kindergarten, the latter group actually started out ahead of the synthetic phonics
group, showing an effect size of -0.61. However, at the end of the first grade, when
the synthetic phonics group had started to sound and blend with letters, there was an
effect size of 0.36 favouring the synthetic phonics group; after second grade this had
risen to 0.45. Thus the synthetic phonics taught groups in both Torgesen et al.
(1999) and Johnston and Watson’s (2004, 2005) studies atypically showed an
upward trajectory of gains in word reading ability across time. As to the present
study, it would have been beneficial to have had an analytic phonics comparison
group that was also assessed when it started school, but the higher levels of reading
ability found for the synthetic phonics group do fit with the existing literature.
The advantage found for children learning to read English using the synthetic
phonics method in Johnston and Watson’s (2004) study led to the method being
advocated for use in all schools in England (Rose Review, 2006), and a government
programme was provided for schools to use (Letters and Sounds, DfES, 2007).
However, a meta-analysis, funded by England’s Department for Education and
Skills (DfES), claimed that there was no clear outcome as to whether synthetic or
analytic phonics was the most effective method (Torgerson, Brooks, & Hall, 2006),
which may seem surprising in the context of the research by Torgesen et al. (1999)
and Johnston and Watson (2004). There are various reasons for this null result. One
of the three studies included in the meta-analysis was an unpublished study of
kindergarten children, where the children were inappropriately trained on complex
vowels, such as tape and rode (Skailand, 1971); these sorts of words are not suitable
for early sounding and blending. An advantage was found for the analytic phonics
group on the trained items, but not on the untrained words. However, the data on the
reading of the trained words were used in the meta-analysis, whereas the National
Long-term effects of synthetic versus analytic phonics
123
Reading Panel only analysed examined performance on untrained items. Torgesen
et al.’s (1999) study was also included. This showed in the long term that the
synthetic phonic method was more effective than embedded phonics but Torgerson
et al. (2006) used data from a few months into this two and a half year study, when
the embedded phonics group was briefly ahead in reading. This was because the
synthetic phonics group was mostly learning phoneme awareness at this stage rather
than phonics. The third study included was Johnston and Watson’s (2004)
Experiment 2, and this also showed that synthetic phonics teaching led to much
better reading skills than the analytic phonics method.
Stannard (2006) and Wyse and Styles (2007) have argued that synthetic phonics
teaching is not as effective in developing reading comprehension as PiPs, although
they present no data to support this belief. However, there is no evidence that
systematic phonics tuition retards reading comprehension. Ehri et al.’s (2001) meta-
analysis found that children taught by a systematic phonics method made gains in
text comprehension as well as decoding, word reading, and spelling. As analytic and
synthetic phonics methods are both systematic, one might predict at the very least
that the two methods would produce equivalent results. However, in the present
study it was found that synthetic phonics teaching led to the boys showing
significantly better reading comprehension compared with those taught by an
analytic phonics approach. This raises questions as to whether this advantage was
caused by the higher levels of word reading skill shown by the synthetic phonics
boys. According to the Simple View of Reading (Gough & Tunmer, 1986), word
reading and oral language comprehension ability together give a good prediction of
reading comprehension skills. Furthermore, Vellutino, Tunmer, Jaccard and Chen
(2007), using structural modelling, have shown that word recognition ability does
play a significant role in reading comprehension for children aged around 7–8 years.
This suggests a direction of causation for the early years of the Clackmannanshire
Study, with the accelerated development of word reading leading to reading
comprehension being significantly above what was expected for age. However,
whereas word reading ability had an upward trajectory, reading comprehension
showed the reverse pattern. In Primary 2, the gain for reading comprehension over
age was 7 months, but by Primary 7 it was 3.5 months, a difference which was
statistically significant (Johnston & Watson, 2005). Interestingly, Vellutino et al.
(2007) found that for children aged around 11–12 years, word reading did not play
such a large role in reading comprehension, and listening (i.e., oral language)
comprehension was found to play a more significant role. An increasing reliance on
oral language skills is likely to be disadvantageous for children from areas of
deprivation, as their general language skills may not be as well developed as those
for children from more advantaged areas. This may also be disadvantageous for
boys, who generally do less well in verbal tests than girls (Hyde & Linn, 1988).
However, if boys can boost their reading comprehension by having very good word
reading ability, this may explain why the synthetic phonics taught boys were as
good as the girls in reading comprehension. With analytic phonics teaching, the
boys had equivalent word reading skills to the girls, but were behind them in reading
comprehension; boys may need higher level of word reading ability to achieve the
same level of reading comprehension as girls. It is interesting that the girls’ reading
R. S. Johnston et al.
123
comprehension was as good as that of the synthetic phonics taught girls, despite
having inferior word reading skills; this may indicate that they were better able to
use general language skills to support their reading comprehension. Indeed, the
programme Progression in Phonics (DfEE, 1999) encouraged the strategy of
guessing unknown words from context and the girls may have been better able to do
this, whereas the boys may have needed to be able to read the individual words more
accurately in order to get meaning from text.
The gain that boys experienced in word reading when learning to read by a
synthetic phonics approach may have a neural substrate underlying it. Burman,
Bitan, and Booth (2008) found, in a study of 9–15 year old children, that boys’
processing of printed words was associated with the activation of areas of the brain
concerned with visual processing, and spoken words were processed in areas
concerned with auditory and phonological processing. That is, their pattern of
activation was modality specific, which may imply a lack of integration of visual
and phonological information. Girls’ performance, on the other hand, was correlated
with activation in supramodal areas of the brain during the reading and spelling
tasks. Boys did also show activation in these areas, but at a lower level, and it was
not associated with task performance. Burman et al. (2008) concluded that language
processing was more abstract in girls and more sensory in boys. Synthetic phonics
teaching may aid boys in learning to integrate visual and phonological information,
thus bringing up their spelling levels to those of girls, and also boosting their word
recognition skills. Mixed methods/analytic phonics approaches may not be so
effective at overcoming boys’ problems in making these links.
This fits with Ehri’s (2005) conclusion that in English the route to skilled sight
word reading is paved with phonology, good readers developing a sight word
recognition of words that is well-underpinned by phonological information in
memory. As word reading involves the integration of visual and phonological
information even in an opaque orthography, synthetic phonics may be more
effective because early on it develops the integration of information from these two
modalities, and this may be particularly beneficial for boys. The analytic phonics
approach, having an early sight word element and late teaching of sounding and
blending, may lead to some children reading largely by a form of sight word reading
underpinned only by superficial connections between print and sounds. However,
there is a view, stemming from the dual route model of reading, that there are two
separate processes in reading, sight word and phonically based decoding (Stuart
et al., 2008). This idea overlooks the fact that once a child has sounded and blended
a word a few times, it will be able to store it in memory and access it without pre-
lexical segmentation (Reitsma, 1983). If children taught by synthetic phonics did
not develop sight word reading, they would become stuck in the full alphabetic
phase outlined by Ehri, making a lot of regularisation errors when reading irregular
words. However, the synthetic phonics group in Study 2 showed no impairment in
reading irregular words compared with the analytic phonics group; in fact, with low
frequency words there was a clear trend towards superior reading of exception and
strange words (they were from around 11 to 16% better on these items). This is
compatible with the connectionist view that even strange words contain some
regular elements that can be accessed by using phonological information. Thus it is
Long-term effects of synthetic versus analytic phonics
123
likely that children taught by the synthetic phonics approach form connections
between the regularly spelled elements and sounds in memory for irregular words;
indeed, the synthetic phonics taught children in the present study were taught to read
such items in this way.
It is interesting that synthetic phonics was found to be very effective, given that
English has an opaque orthography. Certainly, the boost that synthetic phonics
teaching gave to word reading skills in this study suggests that reading development
need not be as slow in English as has been suggested (e.g., Seymour, Aro, &
Erskine, 2003). Indeed, the advantage found for Austrian children in cross-linguistic
studies (e.g., Landerl, 2000; Wimmer & Goswami, 1994) may in part be accounted
for by differences in teaching methods between Britain and Austria, as Wimmer
(1995) has suggested. Another factor to consider is that the children in the studies
were often tested on the reading of polysyllabic nonwords, which might have been
advantageous for the German speaking children, as it is a more polysyllabic
language. The English and Austrian school systems do differ in a number of ways,
not least of which is the age of commencing school. Therefore, it is of interest that a
comparison of children learning to read in countries with similar educational
systems, that is, England versus Wales (the language in the latter country having a
transparent orthography), found that the Welsh children did indeed read better when
aged 5–7 than children in England, but interestingly there were no differences at the
age of 10 (Hanley, Masterson, Spencer & Evans, 2004; Spencer & Hanley, 2003).
This present study makes an important contribution to documenting the long-
term effects of synthetic phonics teaching. Maintaining the gain in word reading for
age would have been noteworthy, but in fact it increased over time, leading to a high
level of attainment at the age of 10. This study has confirmed that the synthetic
phonics approach is effective in English, even though it is an opaque orthography,
and that boys do very much better with this method than the analytic phonics
approach. It is suggested that boys may be slower to develop the integration
between visual and phonological information that underpins word reading due to
sex differences in brain activation when carrying out reading tasks. The early
teaching of synthetic phonics may be more effective in developing these
interconnections for boys in particular, whereas a method like analytic phonics,
which begins with sight words and has a late introduction of sounding and blending,
may not foster this integration so well.
Acknowledgments The authors would like to thank the pupils and teachers who took part in this study.
They would also like to gratefully acknowledge funding from the Scottish Executive Education
Department and the University of Hull; however, the views expressed here are not necessarily those of
these bodies.
References
Baron, J. (1979). Orthographic and word-specific mechanisms in children’s reading of words. Child
Development, 50, 60–72.
Brimer, M. A., & Dunn, L. M. (1984). English picture vocabulary test. Awre, England: Educational
Evaluation Enterprises.
R. S. Johnston et al.
123
Brooks, G. (2002). Phonemic awareness is a key factor in learning to be literate: How best should it be
taught. In M. Cook (Ed.), Perspectives on the teaching and learning of phonics. Royston, England:
UKRA.
Brooks, G. (2003). Sound sense: The phonics element of the National Literacy Strategy. A report to the
Department for Education and Skills. Retrieved from http://www.standards.dfes.gov.uk/pdf/literacy/
gbrooks_phonics.pdf.
Burman, D. D., Bitan, T., & Booth, J. R. (2008). Sex differences in neural processing of language among
children. Neuropsychologia, 46, 1349–1362.
Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001). DRC: A dual route cascaded model
of visual word recognition and reading aloud. Psychological Review, 108, 204–256.
DfEE. (1998). The national literacy strategy framework for teaching. London, England: DfEE.
DfEE. (1999). Progression in phonics. London, England: DfEE. Retrieved from www.standards.dfes.
gov.uk/primary/publications/literacy/63309/.
DfES. (2007). Letters and sounds. London, England: DfES. http://www.standards.dfes.gov.uk/local/clld/
las.html.
Dombey, H. (2006). Phonics and English orthography. In M. Lewis & S. Ellis (Eds.), Phonics: Practice,
research and policy (pp. 95–104). London, England: Paul Chapman Publishing.
Duncan, L. G., & Seymour, P. H. K. (2000). Socio-economic differences in foundation level literacy.
British Journal of Psychology, 91, 145–166.
Ehri, L. (2005). Development of sight word reading: Phases and findings. In M. J. Snowling & C. Hulme
(Eds.), The science of reading: A handbook (pp. 135–154). Blackwell: Oxford.
Ehri, L., Nunes, S., Stahl, S., & Willows, D. (2001). Systematic phonics instruction helps students learn to
read: Evidence from the National Reading Panel’s meta-analysis. Review of Educational Research,
71, 393–447.
Elliott, C. D., Murray, D. J., & Pearson, L. S. (1977). The British abilities scales. Windsor, England:
NFER Nelson.
Feitelson, D. (1988). Facts and fads in beginning reading: A cross-language perspective. Norwood, NJ:
Ablex.
Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading and reading disability. Remedial Special
Education, 7, 6–10.
Hanley, J. R., Masterson, J., Spencer, L. H., & Evans, D. (2004). How long do the advantages of learning
a transparent orthography last? An investigation of the reading skills and incidence of dyslexia in
Welsh children at the age of 10. Quarterly Journal of Experimental Psychology, 37, 1393–1410.
Harm, M. W., & Seidenberg, M. S. (2004). Computing the meanings of words in reading: Cooperative
division of labor between visual and phonological processes. Psychological Review, 111, 662–720.
Harris, L. A., & Smith, C. B. (1976) Reading instruction: Diagnostic teaching in the classroom (2nd ed.,
p. 181). London, England: Holt, Rinehart and Winston.
Hyde, J. S., & Linn, M. C. (1988). Gender differences in verbal ability: A meta-analysis. Psychological
Bulletin, 104(1), 53–69.
Johnston, R. S., & Watson, J. (2004). Accelerating the development of reading, spelling and phonemic
awareness. Reading and Writing: An Interdisciplinary Journal, 17, 327–357.
Johnston, R. S., & Watson, J. (2005). The effects of synthetic phonics teaching on reading and spelling
attainment, a seven year longitudinal study. Ediburgh, Scotland: Scottish Executive Education
Department. Retrieved from http://www.scotland.gov.uk/library5/education/sptrs-00.asp.
Johnston, R. S., & Watson, J. (2007). Teaching synthetic phonics. Exeter, England: Learning Matters.
Landerl, K. (2000). Influences of orthographic consistency and reading instruction on the development of
nonword reading skills. European Journal of Psychology of Education, 15, 239–257.
Macmillan Unit. (2000a). The group reading test II. Windsor, England: NFER-Nelson.
Macmillan Unit. (2000b).
Group reading test II teacher’s guide. Windsor, England: NFER Nelson.
Mullis, V. S., Martin, M. O., Kennedy, A. M., & Foy, P. (2007). IEA’s progress in international reading
literacy study in primary school in 40 countries. Chestnut Hill, MA: TIMSS and PIRLS International
Study Center, Boston College. http://timss.bc.edu/pirls2006/intl_rpt.html.
National Institute of Child Health and Human Development (NICHD). (2000). Report of the national
reading panel: Teaching children to read. Washington, DC: NICDH. Retrieved from http://
www.nichd.nih.gov/publications/nrp/smallbook.cfm.
Neale, M. D. (1989). The neale analysis of reading ability—revised British edition. Windsor, England:
NFER-Nelson.
Long-term effects of synthetic versus analytic phonics
123
Reitsma, P. (1983). Printed word learning in beginning readers. Journal of Experimental Child
Psychology, 36, 321–339.
Rose, J. (2006). Independent review of the early teaching of reading. Retrieved from http://
www.standards.dfes.goc.uk/rosereview/report.pdf.
Schonell, F. J., & Schonell, F. E. (1952). Diagnostic and attainment testing (2nd ed.). Edinburgh,
Scotland: Oliver and Boyd.
Scottish Executive Education Department (SEED). (2000). 5 to 14 Guidelines. Retrieved from http://
www.ltscotland.org.uk/5to14/guidelines/.
Seidenberg, M. S., & McClelland, J. L. (1989). A distributed, developmental model of visual word
recognition and naming. Psychological Review, 96, 523–568.
Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in European
orthographies. British Journal of Psychology, 94, 143–174.
Skailand, D. B. (1971, February). A year comparison of four language units in teaching beginning
reading. New York: Paper presented at annual meetings of the American educational research
Association.
Spencer, L. H., & Hanley, J. R. (2003). Effects of orthographic transparency on reading and phoneme
awareness in children learning to read in Wales. British Journal of Psychology, 94, 1–28.
Stannard, J. (2006). Keeping Phonics in Perspective. In M. Lewis & S. Ellis (Eds.), Phonics. Practice,
research, policy (pp. 120–121). London, England: Paul Chapman Publishing.
Stuart, M., Dixon, M., Masterson, J., & Quinlan, P. (1998). Learning to read at home and at school.
British Journal of Educational Psychology, 68, 3–14.
Stuart, M., Stainthorp, R., & Snowling, M. (2008). Literacy as a complex activity: deconstructing the
simple view of reading. Literacy, 42, 59–66.
Thompson, G. B. (1987). Three studies of predicted gender differences in processes of word reading.
Journal of Educational Research, 80, 212–219.
Torgerson, C. J., Brooks, G., & Hall, J. (2006). A systematic review of the research literature and use of
phonics in the teaching of reading and spelling. DfES research reports(RR711). Retrieved from
http://www.dcsf.gov.uk/research/programmeofresearch/projectinformation.cfm?projectid=14389&
resultspage=1.
Torgesen, J. K., Wagner, R. K., Rose, E., Lindamood, P., Conway, T., & Garvan, C. (1999). Preventing
reading failure in children with phonological processing disabilities: group and individual responses
to instruction. Journal of Educational Psychology, 91, 579–593.
Vellutino, F. R., Tunmer, W. E., Jaccard, J. J., & Chen, R. (2007). Components of reading ability:
Multivariate evidence for a convergent skills model of reading development. Scientific Studies of
Reading, 11, 3–32.
Waters, G. S., Seidenberg, M. S., & Bruck, M. (1984). Childrens’ and adults’ use of spelling-sound
information in three reading tasks. Memory and Cognition, 12, 293–305.
Watson, J. E. (1998). An investigation of the effects of phonics teaching on children’s progress in reading
and spelling. Doctoral Dissertation. University of St Andrews, St Andrews, Scotland.
Wilkinson, G. (1993). Wide range achievement test—3rd edition (WRAT-3) (3rd ed.). Wilmington,
Delaware: Wide Range.
Wimmer, H. (1995). From the perspective of a more regular orthography. Issues in Education, 1,
101–104.
Wimmer, H., & Goswami, U. (1994). The influence of orthographic consistency on reading
development—Word recognition in English and German children. Cognition, 51, 91–103.
Wyse, D., & Styles, M. (2007). Synthetic phonics and the teaching of reading: The debate surrounding
England’s Rose report. Literacy, 41, 35–42.
R. S. Johnston et al.
123
  • ... These are related to the fact that spelling is largely based on word pronunciation and the segmentation of words into sounds. Although phonics is thought to be an effective approach, as mentioned above, the English system of spelling has an opaque orthography and some inconsistent connections between grapheme-phonemes (Johnson, McGeown, and Watson, 2011). There work also mentions a mixed methods approach presupposing an advantage from recognizing words by sight. ...
  • ... As an example of the analytic phonics approach advantage can be that If a child knows the word "look" and needs to read or spell the word "book", a straight-forward link can be made (Callander et al., 2010). Johnston et al. (2012) study made a comparison of 10-year-old-boys and girls who were engaged in learning reading by using analytic and synthetic method as part of their literacy. The study concluded that the group taught by synthetic phonic showed better word reading, spelling, and reading comprehension. ...
    Article
    Full-text available
    Competence in English language literacy can be assumed to be an essential factor for the competitive market in the context of the 21st Century globalization. However, in some cases, pupils learning English as an additional language in same classroom are not homogenous due to their diverse background experiences (Conteh, 2015). For instance, in the UK, there are over 1.5 million EAL learners (The Bell Foundation, 2018); whereas, Cinkara (2017) in Turkey’s context, Turkey welcomed around 2.523.554 Syrian refugees. This study reports the processes of reading English by two Arabic native speaker pupils. Salah (pseudonym) 9-year-old, without prior formal and informal education who was admitted to Year 4 in West of England primary school, UK. Ahmed (pseudonym), 8 years old in Year 3 in a South Turkey school learning English as a subject, possible as his third language, in addition to Arabic and Turkish. The study aims to understand the ways to improve these pupils’ English language reading skills. Action research and structured interviews were used to collect the participants’ data. Thematic analysis was used to identify the themes of the structured interview. It was found that the whole-word approach when reading a word, is the preference for both participants rather than the phonic approach. Further research with larger sample and encouragement of cross-border professional co-operation to improve refugees’ Basic English reading skills is recommended.
  • ... Reading difficulties have also been associated with environmental factors, such as methods in reading instruction (Johnston, McGeown, & Watson, 2011;McGeown, Johnston, & Medford, 2012) and the educational system (e.g., pre-primary education, child's age when they began school, etc.) (Sellès et al., 2015). There is evidence to suggest that the methods used to teach children how to read can influence their skills and predict their reading success and the ways in which their reading development will continue (Connelly, Thompson, Fletcher-Flinn, & McKay, 2009). ...
    Article
    Previous studies have showed that early problems with word decoding can lead to poor performance in text reading and comprehension and suggest that poor readers often struggle with reading deficits throughout their school years. Therefore, early detection of those children who are at risk for slow reading development and/or who belong to the lowest reading profiles is essential in order to organize proper support. The present study explores the heterogeneity and prevalence of latent reading profiles among 769 Finnish- and German-reading students during their first and second school years in three countries (Finland, Germany, and Italy) using latent profile analysis. The results identified three latent profiles among Finnish readers, one of which (sentence-level reading) was identified as developing slowly. Among German-reading students, four latent profiles were discovered, two of which were identified as developing slowly. The results of ordinal logistic regression modeling show that rapid automatic naming (RAN) was significantly related to poorer reading profiles among Finnish- and German-reading students, and that the poorer results in letter-sound connection testing among the German-reading group was also significantly related to poorer reading profiles. Although the educational systems have some differences between Germany and German-speaking areas of Italy, no significant country effect was detected. In addition, a child’s age and spoken language did not significantly affect the student’s reading profile.
  • ... As an example of the analytic phonics approach advantage can be that If a child knows the word "look" and needs to read or spell the word "book", a straight-forward link can be made (Callander et al., 2010). Johnston et al. (2012) study made a comparison of 10-year-old-boys and girls who were engaged in learning reading by using analytic and synthetic method as part of their literacy. The study concluded that the group taught by synthetic phonic showed better word reading, spelling, and reading comprehension. ...
    Article
    Full-text available
    literacy can be assumed to be an essential factor for the competitive market in the context of the 21st Century globalization. However, in some cases, pupils learning English as an additional language in same classroom are not homogenous due to their diverse background experiences (Conteh, 2015). For instance, in the UK, there are over 1.5 million EAL learners (The Bell Foundation, 2018); whereas, Cinkara (2017) in Turkey’s context, Turkey welcomed around 2.523.554 Syrian refugees. This study reports the processes of reading English by two Arabic native speaker pupils. Salah (pseudonym) 9-year-old, without prior formal and informal education who was admitted to Year 4 in West of England primary school, UK. Ahmed (pseudonym), 8 years old in Year 3 in a South Turkey school learning English as a subject, possible as his third language, in addition to Arabic and Turkish. The study aims to understand the ways to improve these pupils’ English language reading skills. Action research and structured interviews were used to collect the participants’ data. Thematic analysis was used to identify the themes of the structured interview. It was found that the whole-word approach when reading a word, is the preference for both participants rather than the phonic approach. Further research with larger sample and encouragement of cross-border professional co-operation to improve refugees’ Basic English reading skills is recommended.
  • ... Findings suggested that students who participated in early intervention made more progress in primary grades and are at lower risk for later reading failure (Coyne et al., 2004;Vellutino et al., 2003), supporting the claim that strategies can enhance children's motivation and engagement on phonics performance. Lemons, Mrachko, Kostewicz and Paterra (2012) and Johnston, McGeown and Watson (2012) Sparks, R. L., Patton, J., & Murdoch, A. (2014). also examined the National Reading Panel (NRP) report (National Institute of Child Health and Human Development, 2000) the magnitude of the effect of systematic phonics instruction. ...
    Article
    Full-text available
    This study examined the effects of strategies on children’s phonic reading performance at Kosober Primary School, Injibara, Amhara National Regional State, Ethiopia. For this purpose, one hundred grade one children of two sections of intact classrooms (n=50) and (n=52) were selected as experimental and control groups respectively and participated in quasi-experimental pre-test and post-test research design aimed to examine improvements on children’s phonic performance. Researchers administered pre-test and post-test to collect quantitative data and observation and teacher self-reflection report to qualitative data. A paired samples t-test was computed to analyze the quantitative data. To analyze qualitative data, researchers employed narratives. Findings indicated that phonic reading strategies could improve children’s phonic performance. Furthermore, findings from observation and teacher’s reflections showed that application of strategies resulted in enhanced children’s phonic performance. Eventually, recommendations and implications for further research were suggested.
  • ... Aujourd'hui, il convient de rester très prudent car aucun consensus n'est établi. En effet, si certaines expérimentations ont annoncé un avantage de l'approche phonique synthétique sur l'approche phonique analytique (e.g., Johnston, McGeown, & Watson, 2012), au moins deux méta-analyses (Ehri, Nunes, Stahl & Willows, 2001;Torgerson et al., 2006) suggèrent au contraire que les deux approches ont une efficacité équivalente (voir aussi Wyse & Goswami, 2008). Une étude récente, menée à très grande échelle (Machin et al., 2018), conclut à un avantage des classes où l'approche phonique synthétique a été utilisée, par rapport à l'ensemble des autres classes de l'étude. ...
    Article
    Full-text available
    La question abordée, et illustrée dans le domaine de l’enseignement du code alphabétique au CP, est celle de l’utilisation des données issues de la recherche pour faire évoluer l’enseignement vers des pratiques plus efficaces. Cet objectif est profondément louable mais demande une extrême rigueur. Une règle particulièrement fondamentale est de ne pas faire dire à la recherche ce qu’elle ne peut pas dire. Ne pas la respecter sera dommageable pour l’école, mais aussi pour l’ensemble de la communauté scientifique concernée. À notre avis, cette règle fondamentale n’a pas été respectée lors de l’écriture de certains passages du « guide fondé sur l’état de la recherche pour enseigner la lecture et l’écriture au CP» diffusé par le Ministère de l’Éducation nationale en 2018, notamment lorsque ce guide aborde de façon très précise comment enseigner le code graphème-phonème. Notre article reprend donc cette question en essayant de lui appliquer la rigueur souhaitée, afin d’éclairer les lecteurs sur ce qui, dans certaines préconisations du fameux guide, relève vraiment de connaissances issues de la recherche et ce qui n’en relève pas. Ainsi, les données scientifiques justifient totalement la préconisation d’enseigner le code graphème-phonème de façon explicite et en respectant une progression logique. Par contre elles engagent à rester beaucoup plus prudent sur la question des méthodes à appliquer pour enseigner le code, car aucun consensus scientifique ne se dégage. Quant à la préconisation d’entrer dans cet apprentissage « par le graphème » plutôt que « par le phonème », c’est une idée qui ne semble justifiée par aucune littérature scientifique, quelle que soit la façon dont on tente de l’interpréter.
  • ... Mcgeown and Medford (2014) found in their study that the participants who were taught using synthetic phonics approach demonstrated greatest gains in literacy skills for both regular and irregular words. Furthermore, Johnston, McGeown and Watson (2012) agreed in their findings which the samples in the treatment group had greater performance in comprehension compared to the control group. Besides, study carried out by Jamaludin et al. (2016) also approved this through their findings which resulted a significant improvement on their experimental group's comprehension skills after receiving synthetic phonics teaching. ...
    Article
    Full-text available
    Reading skill in English Language learning is crucial for the learners' language development. The basic skills in reading involves from letter-sounds to word recognition and language comprehension which some English as Second Language (ESL) readers find it is difficult to acquire. There are overwhelming researches evidence point to the use of a synthetic phonics approach to help young struggling readers to develop their reading ability. This paper discusses the importance of early reading skills development and how synthetic phonics instruction helps ESL struggling learners to read. In addition to that, it also outlines the benefits and the limitations of synthetic phonics instruction and lists some suggestions to help phonics instruction to be more effective.
  • Chapter
    Annually, the results released from the National Assessment Program Literacy and Numeracy testing Australia-wide confirm the huge gaps in literacy performance between students from Indigenous and non-Indigenous Australian backgrounds, particularly for those students living in remote communities. Recent research has shown promising results, indicating that one of the major changes that is likely to show positive effects on Indigenous students’ literacy levels is the provision of scientific, evidence-based, best-practice reading instruction.
  • Article
    From 2006 the British government strongly favoured synthetic phonics as the principal approach for the teaching of initial literacy in state-funded primary schools in England, and since 2010 has made it mandatory. In 2007–2013 just over 100 commercially published phonics schemes were available, and in that same period the government maintained a system of quality assurance, in the form of two (successive and non-overlapping) panels of independent evaluators. Their task was to judge whether commercial publishers’ self-evaluations of their phonics schemes and materials were correct, in the sense of justifying statements that they met the government’s criteria for such schemes, etc. Of the schemes that were judged, just over half (54) were found to contain linguistic errors. In this article the errors are analysed in detail, and classified into three main categories: phonetic inaccuracies, phonic inaccuracies, and misguided pedagogies. The criteria for that classification are stated, and conclusions and recommendations drawn – the main recommendation being that existing schemes need to be scrutinised in detail to ensure that they are fit for purpose. And this would apply to all phonics schemes used anywhere in the English-speaking world, not just in England, even though the criteria for phonetic and phonic accuracy would necessarily differ across accents.
  • Chapter
    Many communities lack adequate after-school programs, especially after-school programs that help children improve academically. To answer this need, our university research team collaborated with two community sponsors to develop an effective after-school program for students who were failing in reading. Some of the children had failed for multiple years. The goal was to bring children up to their appropriate age level in reading and to develop something that could be an ongoing, long-term self-sustaining program for the community. Across eight years, we developed, implemented, and tested an after-school group-centered prevention model, using a program packet to facilitate training with a community volunteer staff. The first-person accounts included in this article describe the development of the program and the development of the volunteer staff needed to continue the program. We present a model that can be used by any community to develop a similar program. The success of our model focused on using a group-centered prevention format, stressing intrinsic motivation, and teaching vowel clustering.
  • Book
    Full-text available
    EXECUTIVE SUMMARY 1. We report here a study of the effectiveness of a synthetic phonics programme in teaching reading and spelling. Around 300 children in Primary 1 were divided into three groups. One group learnt by the synthetic phonics method, one by the standard analytic phonics method, and one by an analytic phonics programme that included systematic phonemic awareness teaching without reference to print. At the end of the programme, the synthetic phonics taught group were reading and spelling 7 months ahead of chronological age. They read words around 7 months ahead of the other two groups, and were 8 to 9 months ahead in spelling. The other two groups then carried out the synthetic phonics programme, completing it by the end of Primary 1. 2. We have followed the progress of all of these children for 7 years, examining their performance in word reading, spelling and reading comprehension. At the end of Primary 2, the boys performed equally well in word reading, regardless of which method they had started with in Primary 1. However, the girls read words significantly less well if they had started with the standard analytic phonics programme. Furthermore, both boys and girls were behind in spelling if they had started with the standard analytic phonics programme, even if it had been supplemented with systematic phonemic awareness training. 3. At the end of Primary 7, word reading was 3 years 6 months ahead of chronological age, spelling was 1 year 8 months ahead, and reading comprehension was 3.5 months ahead. However, as mean receptive vocabulary knowledge (an index of verbal ability where the average is 100) was 93 at the start of the study, this is a group of children for whom normal performance might be expected to be below average for chronological age on standardised tests. Therefore this may be an underestimate of the gains with this method. 4. In all 35 countries surveyed in an international study, including Scotland, it was found that the boys' reading comprehension was significantly behind that of the girls'. In the present study, the boys' word reading was significantly ahead of that of the girls' from Primary 3 onwards; by the end of the study in Primary 7 they were 11 months ahead of the girls. In spelling, the boys were significantly ahead of the girls in Primaries 4, 6 and 7, being 8.6 months ahead by the end of the study. They were also 3 months ahead of girls in reading comprehension, but this difference was not statistically significant. However although the boys read better than the girls, they nevertheless reported a less favourable attitude to reading. 5. It had been expected that children from disadvantaged homes would perform less well than those from advantaged homes. However, this was not statistically significant for word reading and spelling until Primary 7 (and only marginally so for reading) and was only significant for reading comprehension in Primaries 5 and 7. 6. In the early years of the study, the level of underachievement was very low. For example, in Primary 3, only 0.8% of the children were more than two years behind chronological age in word reading, with 0.4% being behind in spelling, and 1.2% being behind in reading comprehension. However, by Primary 7 this had increased to 5.6% behind in word reading, 10.1% behind in spelling, and 14.0% behind in reading comprehension. It is possible that these levels of underachievement are quite moderate for children with a somewhat below average level of receptive vocabulary knowledge. This could be established by carrying out a study of a control sample learning to read by the standard analytic phonics approach. 7. Teachers and Head Teachers have responded very favourably to the programme, having found that the children's reading and spelling skills are very accelerated, that underachievers can be detected earlier and that the children are very motivated. 8. Overall we conclude that the synthetic phonics approach, as part of the reading curriculum, is more effective than the analytic phonics approach, even when it is supplemented with phonemic awareness training. It also led boys to reading words significantly better than girls, and there was a trend towards better spelling and reading comprehension. There is evidence that synthetic phonics is best taught at the beginning of Primary 1, as even by the end of the second year at school the children in the early synthetic phonics programme had better spelling ability, and the girls had significantly better reading ability.
  • Article
    Three studies were conducted to examine individual differences among young children in the extent of use of alternative cognitive processes for word reading. The expectation was that boys, of the same reading attainment level as girls, tend to rely more than girls on access to phonological segments of words. In Study I a predicted pattern of gender differences was tested with 87 seven-year-old children reading pseudohomophones graphemicalty different and graphemically similar to lexically matched words. In Study II predicted differences were tested with the same children reading words of regular and exception grapheme-phoneme relationships. For 84 six-year-old children in Study III predicted differences were examined in the relationship of word reading accuracy to phonological consistency of initial segments of words. The results of the three studies supported the expectation.
  • Article
    To clarify the role of decoding in reading and reading disability, a simple model of reading is proposed, which holds that reading equals the product of decoding and comprehension. It follows that there must be three types of reading disability, resulting from an inability to decode, an inability to comprehend, or both. It is argued that the first is dyslexia, the second hyperlexia, and the third common, or garden variety, reading disability.
  • Article
    A quantitative meta-analysis evaluating the effects of systematic phonics instruction compared to unsystematic or no-phonics instruction on learning to read was conducted using 66 treatment-control comparisons derived from 38 experiments. The overall effect of phonics instruction on reading was moderate, d = 0.41. Effects persisted after instruction ended. Effects were larger when phonics instruction began early (d = 0.55) than after first grade (d = 0.27). Phonics benefited decoding, word reading, text comprehension, and spelling in many readers. Phonics helped low and middle SES readers, younger students at risk for reading disability (RD), and older students with RD, but it did not help low achieving readers that included students with cognitive limitations. Synthetic phonics and larger-unit systematic phonics programs produced a similar advantage in reading. Delivering instruction to small groups and classes was not less effective than tutoring. Systematic phonics instruction helped children learn to read better than all forms of control group instruction, including whole language. In sum, systematic phonics instruction proved effective and should be implemented as part of literacy programs to teach beginning reading as well as to prevent and remediate reading difficulties.
  • Article
    Elementary and middle school children were given a large battery of tests evaluating reading subskills and reading-related cognitive abilities. These measures were used to define latent representing skills and abilities believed to be important components of reading comprehension. Hypothesized relationships among these constructs were specified within the context of a structural model we call the “Convergent Skills Model of Reading Development,” and developmental differences in the relative contribution made by each construct to reading comprehension performance were assessed through confirmatory factor analysis using the LISREL and AMOS programs. Results provide qualified support for the model and were interpreted as consistent with the major premises of both Gough and Tunmer's (1986)15. Gough , P. B. and Tunmer , W. E. 1986. Decoding, reading, and reading disability.. Remedial and Special Education, 7: 6–10. [CrossRef]View all references “Simple View” and Sticht's (1979)35. Sticht , T. G. 1979. “Applications of the Audread model to reading evaluation and instruction”. In Theory and practice of early reading, Edited by: Resnick , L. B. and Weaver , P. A. Vol. 1, 209–226. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. View all references “Audread” models of reading.
  • Article
    3 experiments show differences among children in relative reliance on spelling-sound rules versus word-specific associations in reading words. Ability to read nonsense words (e. g., lut) is correlated more highly with ability to read regular words (cut) than with ability to read exception words (put). Children also differ in their tendencies to make meaning-preserving errors as opposed to errors involving overgeneralization of spelling-sound rules to exception words. Children who rely more on rules are more slowed by reading 2 successive words with the same spelling pattern pronounced differently (maid, said). The tendency to rely on rules as opposed to word-specific associations is correlated with ability to read regular words. Individual differences appear to arise largely from differences in instruction, although there are also consistent sex differences (boys tend to rely more on rules). In experiment 3, performance on nonsense words is improved by instruction to think of analogous words. This finding, plus others, indicates that both analogies and smaller-unit rules are used to apply rules.