british educational research journal synthetic phonics and the

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [University of Winchester]On: 16 July 2010Access details: Access Details: [subscription number 773572795]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

British Educational Research JournalPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713406264

Synthetic phonics and the teaching of readingDominic Wysea; Usha Goswami 1a

a University of Cambridge, UK

To cite this Article Wyse, Dominic and Goswami 1, Usha(2008) 'Synthetic phonics and the teaching of reading', BritishEducational Research Journal, 34: 6, 691 — 710To link to this Article: DOI: 10.1080/01411920802268912URL: http://dx.doi.org/10.1080/01411920802268912

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t713406264

http://dx.doi.org/10.1080/01411920802268912

http://www.informaworld.com/terms-and-conditions-of-access.pdf

Synthetic phonics and the teaching of

reading

Dominic Wyse* and Usha Goswami1

University of Cambridge, UK

A review of the teaching of early reading in England commissioned by the UK Government

recommended that synthetic phonics should be the preferred approach for young English learners.

In response, all English schools have been told to put in place a discrete synthetic phonics

programme as the key means for teaching high-quality phonic work. In this paper we analyse the

evidence presented by the review to support the change to synthetic phonics. We show that the

review provided no reliable empirical evidence that synthetic phonics offers the vast majority of

beginners the best route to becoming skilled readers. We analyse the available empirical evidence in

English, and show instead that the data support approaches based on systematic tuition in phonics.

There is also evidence that contextualised systematic phonics instruction is effective. However, more

research is needed, particularly with typically developing readers, in order to determine whether

contextualised systematic phonics is more effective than discrete systematic phonics.

The importance of systematic phonics instruction in relation to the teaching of

reading has been increasingly recognised by English-speaking countries. In 2000 the

USA’s National Reading Panel advocated systematic phonics instruction as part of a

balanced programme of reading teaching (National Institute of Child Health and

Human Development [NICHD], 2000). In 2005, the Australian government

enquiry concluded that systematic, direct and explicit phonics instruction was

important as part of an integrated approach to reading teaching (Australian

Government, Department of Education Science and Training, 2005). In England,

the government enquiry into the teaching of early reading (hereafter the Rose

Report) reached a narrower conclusion (Rose, 2006). Rose’s recommendation was

that the systematic phonics approach selected by schools should be synthetic phonics.

The Rose Report also put forward a ‘simple view of reading’ (Rose, 2006, Appendix

by Stuart & Stainthorp). This adaptation of the classic ‘simple view’ put forward by

Gough and Tunmer (1986) recommended clear differentiation between word

recognition processes and language comprehension processes, so that teachers could

assess performance separately in each and plan different kinds of teaching for each

dimension. One author of this Appendix explains, ‘confounding the two dimensions

*Corresponding author: Faculty of Education, University of Cambridge, 184 Hills Road,

Cambridge CB2 8PQ, UK. Email: [email protected]

British Educational Research Journal

Vol. 34, No. 6, December 2008, pp. 691–710

ISSN 0141-1926 (print)/ISSN 1469-3518 (online)/08/060691-20

# 2008 British Educational Research Association

DOI: 10.1080/01411920802268912

Downloaded By: [University of Winchester] At: 14:34 16 July 2010

is not conducive to effective teaching in either’ (Stuart, 2006, p. 9). The simple

model provided the basis for the recommendation in the Rose Report that systematic

phonics should be taught ‘discretely’ (p. 70).

Synthetic phonics and the simple view of reading is being introduced

immediately to all English schools. Hence, a universal intervention is being

introduced on the basis of the evidence reviewed by Rose (2006). As noted by

Rutter (2006), it is always desirable to determine the efficacy of an intervention

under optimal research conditions such as a randomised controlled trial before

launching an intervention universally. For programmes intended to make a real

difference in the long term, such as synthetic phonics, the research evaluation

should be long term, and it must be recognised that subgroups (for example,

children with learning difficulties) may require something different. Our first

question in this article is whether any of these desirable conditions were met by the

evidence base considered by Rose (2006). Rose outlined his remit as ‘to examine

… what best practice should be expected in the teaching of early reading and

synthetic phonics’ (2006, p. 7). We therefore focus on the evidence that he

gathered to meet that remit concerning best practice.

As will become evident, one difficulty in analysing the evidence base underpinning

the report is that Rose did not consider empirical research evidence as the primary

form of data relevant to recommending changes in national policy on the teaching of

reading. His approach is formalised in paragraph 47 of the review:

The review’s remit requires a consideration of ‘synthetic’ phonics in particular …

through examination of the available evidence and engagement with the teaching

profession and education experts. Having followed those directions, and notwithstanding

the uncertainty of research, there is much convincing evidence to show from the practice

observed that … synthetic phonics … offers the vast majority of beginners the best route

to becoming skilled readers … (p. 19, emphasis added)

Stuart (2006), a major adviser to Rose, also argued that the inconclusive nature of

available research evidence required Rose to seek recourse to observation of selected

teaching practices in order to make national recommendations: ‘the available

research evidence is insufficient to allow reliable judgements of the relative

effectiveness of implementing different approaches to systematic structured phonics

teaching’ (Stuart, 2006, p. 11). Rose argued that empirical research evidence cannot

provide the answers that he required, as ‘In recent years, there has been a

convergence of opinion among psychologists investigating reading that little progress

towards understanding how reading happens in the human mind is likely to be made’

(p. 75, emphasis added). He stated, ‘focusing on the practice observed in the

classroom and its supportive context, rather than debating the research, is

[therefore] not without significance for this review’ (p. 62). Our second question

in this article is therefore whether the empirical research base on the teaching of early

reading is really so uncertain.

An alternative view would be that a great deal of progress has been made in

understanding the acquisition and development of reading, and its neural under-

pinning in the human mind. To paraphrase Frith (1998), ‘reading literally changes

692 D. Wyse and U. Goswami


the brain’ (p. 1051). Importantly, it does so in somewhat different ways in different

languages (Goswami, 2006). Understanding the particular demands made by a

language like English is crucial for the successful early teaching of reading in English.

This is because written English is different from many other alphabetic languages.

The phonological complexity of syllable structures in English, coupled with the

inconsistent spelling system, mean that direct instruction at levels other than the

phoneme may be required in order to become an effective reader. In countries like

Greece, Finland, Italy and Spain, syllable structure is simple (mainly consonant–

vowel syllables) and there are 1:1 mappings between letters and sounds.

Accordingly, most children are fluent readers long before the end of their first year

of instruction (Seymour et al., 2003). For such languages, teaching reading by

synthetic phonics can be extremely effective (Landerl, 2000). In English-speaking

countries, such early fluency has proved elusive. Our third question in this article is

what empirical studies can tell us about the effective teaching of reading in English.

As will be illustrated, the complexity of reading acquisition in English makes it

unlikely that the universal adoption of one method, synthetic phonics, without any

evidence of proof of concept (for example, via randomised controlled trials), will

deliver the results expected by Rose and by the UK Government.

The case for systematic phonics

One of the most significant contributions to questions about research evidence and

the teaching of reading was the American National Reading Panel (NRP) report on

teaching children to read in English (NICHD, 2000). This was an extensive meta-

analysis of research evidence which addressed a number of questions about early

literacy, including: ‘Does systematic phonics instruction help children learn to read

more effectively than non-systematic phonics instruction or instruction teaching no

phonics?’ (ch. 2, p. 92), and ‘Are some types of phonics instruction more effective

than others? Are some specific phonics programs more effective than others?’ (ch. 2,

p. 93). As far as differences between analytic and synthetic phonics programmes

were concerned, the NRP concluded that ‘specific systematic phonics programs are

all significantly more effective than non-phonics programs; however, they do not

appear to differ significantly from each other in their effectiveness although more

evidence is needed to verify the reliability of effect sizes for each program’ (NICHD,

2000, ch. 2, p. 93). In the NRP, the effect sizes for the phonics approaches classified

as ‘synthetic’ versus ‘analytic’ were statistically equivalent.

More recently, England’s Department for Education and Skills (DfES) commis-

sioned a large-scale study of approaches to the teaching of reading. The

methodology of the NRP was refined to produce a meta-analysis focused on studies

that were randomised controlled trials (RCTs). On the basis of their study

Torgerson et al. (2006, p. 49) concluded that ‘There is currently no strong RCT

evidence that any one form of systematic phonics is more effective than any other’.

As RCTs are one form of the ‘optimal research conditions’ noted by Rutter (2006),

this is an important finding.

Synthetic phonics 693


Two rigorous forms of reviewing large numbers of research studies, the meta-

analyses conducted by the NRP and the meta-analysis restricted to RCTs by

Torgerson et al. (2006), converge in concluding that the systematic teaching of

phonics is crucial for literacy acquisition. This conclusion was also recognised by

Rose (2006), who stated, ‘Having considered a wide range of evidence, the review

has concluded that the case for systematic phonic work is overwhelming …’ (p. 20).

However, he continued, ‘… is overwhelming, and much strengthened by a synthetic

approach’. He then listed the core components of any systematic phonics

programme, namely learning grapheme-phoneme correspondences, and learning

to blend and segment. Despite this common ground between systematic phonics and

synthetic phonics, it is stated that ‘synthetic phonics … offers the vast majority of

beginners the best route to becoming skilled readers’ (p. 19). Rose wrote, ‘analytic

phonics is good, but synthetic phonics is better’ (p. 19, quote attributed to Rhona

Johnston). In the section of the Rose Report outlining what best practice should

comprise (pp. 15–28), the footnotes refer to various forms of empirical evidence such

as the NRP and Torgerson et al. reviews, and one published empirical study on

synthetic phonics, Johnston and Watson (2005). The section of the Rose Report

presenting evidence from practice (pp. 61–69) focuses on visits made by Rose and

his team to the Scottish educational authority of Clackmannanshire.

The case for synthetic phonics

The studies known in the UK as the ‘Clackmannanshire studies’ have been written

up in a number of forms by Johnston and Watson (Johnston & Watson, 2003, 2004,

2005; Watson & Johnston, 1998). Subsequent to the adoption of a reading tuition

programme called by these authors ‘synthetic phonics’, children were reported to

make considerable gains in reading. The Rose Report summarised these gains as

follows:

At the end of Primary 7 [Year 6 in England, i.e. age 10 to 11], 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 … was 93 at the start of this study … this may be an underestimate of the

gains with this method. (p. 61)

Johnston and Watson’s (2005) work is not a research paper, rather it is a summary of

one of their studies written for a government website. However, one of their

publications concerning their Clackmannanshire data sets has been peer-reviewed

(Johnston & Watson, 2004), and Experiment 2 in that paper are the same data

discussed in Johnston and Watson (2005), the main evidence used in the Rose

Report. We therefore focus on analysing the data presented in Experiment 2 of

Johnston and Watson’s (2004) paper.

Johnston and Watson (2004) reported two studies, although chronologically

Experiment 2 was conducted before Experiment 1. Experiment 1 will not be

analysed in depth here (see Goswami, 2008, for a thorough analysis), as it was not

mentioned as evidence in the Rose Report. Experiment 2 was an intervention study,



with both types of intervention administered by the authors themselves. Intervention

was extra to normal classroom tuition, and began six weeks after school entry. This

extra training was given twice weekly for 10 weeks, and was based on a set of 114

printed words. A control group (‘no-letter training group’) was ‘exposed to the print

vocabulary shown to the other two groups, but a ‘‘look-and-say whole word’’

approach was adopted’ (p. 346). One intervention group (‘accelerated letter learning

group’, also referred to as the ‘analytic phonics’ group) was taught two letters per

week in the initial position in the 114 words. ‘Various games were played where the

pupils matched pictures and words, however their attention was drawn only to

the initial sounds of the words and the letters representing those sounds’ (p. 347).

The second intervention group (‘synthetic phonics group’) was also taught two

letters per week, but learned these letters in all positions of the words (‘accelerated

learning and blending of the letter sounds in initial, middle and final positions of

words’, p. 347). The experimenter demonstrated how the words could be read

by saying each letter sound in a word distinctly from left to right, joining them together

smoothly without pausing between each sound. The children practised reading each

word in this way, first of all with the experimenter, and then independently … using

lower case letters and a magnetic board. (p. 347)

Therefore, only the synthetic phonics group was taught to read the 114 target words

independently.

Johnston and Watson (2004) provide post-test data by group for the participants

three months after the end of the intervention, and nine months after the end of the

intervention. Three months after the intervention ended, the synthetic phonics

group had a mean reading level of 5 years 7 months (mean age in all three groups at

this point was 5 years 5 months). The accelerated letter learning group and the no-

letter control group each had a mean reading age of 5 years 3 months, a significantly

lower level of attainment. The superior performance of the synthetic phonics group

was maintained nine months later, by which time the other two groups had also been

taught (by their classroom teachers) to analyse letters in other positions in words—

yet had failed to ‘catch up’ (p. 351). The authors concluded:

Experiment 2 … [controlled] for speed of letter learning; the synthetic phonics group

still read and spelt better than the analytic phonics group. It is concluded that synthetic

phonics was a more effective approach to teaching reading, spelling and phonemic

awareness than analytic phonics. (p. 351)

Reading comprehension was reported to be equivalent for all three groups (p. 251).

However, the study design is not rigorous enough to enable a conclusion about the

superiority of one teaching method against the other. This is because the absolute

amount taught to the children in each intervention group is confounded with the

labels given to the teaching methods. Johnston and Watson (2004) define ‘analytic

phonics’ as ‘children [learning] letter sounds in the context of words that they have

been taught to recognize by sight’ (p. 329). A stringent comparison of analytic versus

synthetic teaching methods would require each group to be taught the same total

number of letter-sounds in the context of all positions in words, and to be taught the

same component skills of sounding out, blending and decoding the 114 target



words. To compare teaching methods, the required design should contrast whether

this teaching occurred in the context of sight words (i.e. analytically) or began from

discrete sounds (i.e. synthetically). In Johnston and Watson’s study, only the

synthetic phonics group were taught sounding out, blending and reading of the 114

words in the target list. This is not an optimal research design for exploring the

research question of whether synthetic phonics methods are superior to analytic

methods (Experiment 1, which is not cited by Rose, did not use a rigorous research

design either; see Goswami, 2008). It is currently unknown whether any differences

would be found between children’s progress if the correct research design were to be

used. Regrettably, the Rose Report did not recognise that such a study was required

before recommendations about national policy could be made, nor indeed that an

RCT of the method was required before synthetic phonics was rolled out nationally.

This is particularly important in the light of the UK experience with Sure Start (part

of the UK equivalent of No Child Left Behind; see Rutter, 2006). The efficacy of

Sure Start was not determined under optimal research conditions before its launch,

and early evaluations did not find evidence of clear benefit.

It is important to note that the Rose Report did accept that the methodology of the

Clackmannanshire study ‘received some criticism by researchers’ (p. 61). Rather

than presenting these criticisms, the report discussed a visit by Rose and his team to

Clackmannanshire:

The visit provided the review with first-hand evidence of very effective teaching and

learning of phonic knowledge … focusing on the practice observed in the classroom and

its supportive context, rather than debating the research, is therefore not without

significance for this review. (p. 62, emphasis added)

Rose tells us, ‘One teacher said ‘‘I have never seen results like this in 30 years of

teaching’’’ (p. 63). Such anecdotes share similarities with those supplied by teachers

who have received training in various commercial phonics packages ‘THRASS2 is

like a new religion! I have seen the light—the answer to how to teach English

spelling’ (THRASS, 1999); ‘Jolly Phonics has given our children the chance to

succeed in reading and writing’ (Jolly Phonics Case Study, 1999). Regarding such

packages, the Rose Review stated, ‘all these programmes were highly systematic and

the perceived, sharp differences that divided their advocates appeared to make little

difference to the claimed success rates’ (p. 20). Curiously, it is not recognised that

exactly the same logic applies to the Clackmannanshire studies.

The case against synthetic phonics

Recent cross-language research suggests that there are two critical factors that affect

the acquisition of efficient phonics skills across languages. One is the phonological

complexity of the language. Most of the world’s languages have words made up of

syllables with a simple or consonant–vowel (CV) structure. English words made up

of simple CV syllables include ‘yoyo’ and ‘baby’. Languages with a simple CV

syllable structure include Italian, Spanish and Finnish. Words in these languages

tend to be longer than words in English, but easier to segment into phonemes. This



is because the onset-rime level of analysis of the spoken syllable and the phoneme

level of analysis of the spoken syllable are identical. This is shown in Figure 1.

Children seem to acquire phonic recoding skills much faster when the phonology of

their language has a simple syllabic structure. In English, only 5% of monosyllables

are CV (De Cara & Goswami, 2002). The primary syllable type in English is CVC

(43% of monosyllables), followed by CVCC (21%) and CCVC (15%). Note,

however, that languages like German do not have a simple syllabic structure either.

In fact, words in German are frequently exactly the same as words in English. Words

like sand (Sand), ball (Ball), wine (Wein) and mouse (Maus) are phonologically

identical. For words like these, onset-rime segmentation is not equivalent to

phonemic segmentation.

Nevertheless, German children learn phonics much faster than English children

(Frith et al., 1998; Landerl, 2000). This is because of the second critical factor

Figure 1. Schematic example of the phonological complexity of a familiar, early-acquired word in

English versus Italian



affecting the efficient acquisition of phonics skills, the consistency of the symbol-to-

sound mapping (Ziegler et al., 1997). In English, one letter or letter cluster can have

multiple pronunciations (e.g. ‘cough’, ‘rough’, ‘though’; this is also true of languages

like Danish). In other languages, letters are always pronounced in the same way (for

example, German, Greek, Italian, Spanish). In English, a single speech sound

(phoneme) can have multiple spellings (e.g. the phoneme /e/ in hurt, dirt, Bert). In

other languages, a phoneme may be almost always spelled in the same way (e.g.

Italian, Serbo-Croatian). English is exceptionally inconsistent, because it suffers

from a large amount of inconsistency in both reading and spelling. This inconsistency

inhibits the rapid acquisition of grapheme–phoneme recoding skills (Wimmer &

Goswami, 1994; Goswami et al., 1998, 2001; Seymour et al., 2003). The acquisition

of phonic recoding skills during the first year of reading tuition across the European

Community nations is shown in Table 1, adapted from Seymour et al. (2003). It

should be noted that the English-speaking children tested were schooled in Scotland

using predominantly phonics methods. The table shows clearly that children

learning to read highly transparent orthographies where the spoken language has a

simple CV syllable structure are extremely efficient in grapheme–phoneme decoding

within months of entering school.

Direct cross-language comparisons of teaching beginning readers by synthetic

phonics support the empirical evidence shown in Table 1. Landerl (2000) compared

English children taught via synthetic phonics to German children taught using

similar methods. She commented that in German initial reading was typically taught

via ‘a straightforward phonics teaching regime … through heavy emphasis on

teaching of letter–sound correspondences and blending and sounding out words’

(p. 240). She therefore sought out schools in England following a similar regime, in

order to compare the progress made by English children receiving ‘straightforward

Table 1. Data (% correct) from the large-scale study of reading skills at the end of grade 1 in 14

European Community countries (adapted from Seymour et al., 2003)

Language Familiar real words Pseudo-words

Greek 98 92

Finnish 98 95

German 98 94

Austrian German 97 92

Italian 95 89

Spanish 95 89

Swedish 95 88

Dutch 95 82

Icelandic 94 86

Norwegian 92 91

French 79 85

Portuguese 73 77

Danish 71 54

Scottish English 34 29



phonics’ with that of German children. The phonics method chosen was a

commercial programme in use in the UK, Jolly Phonics (Lloyd, 1998). An

additional contrast was with the ‘standard’ method of reading tuition in England at

that time (preceding the National Literacy Strategy of 1998), which was classified as

‘a mixed approach providing a combination of whole word and phonics methods’

(p. 241)—‘children in this school received a phonics lesson once a week’ (p. 247).

Thus two English schools adopting contrasting methods to the teaching of initial

reading were compared to a German school utilising a synthetic phonics approach.

The outcome measure was non-word reading at grades 1, 2, 3 and 4 in each case.

The important study was Experiment 2, for which Landerl herself carried out the

testing of all participating children (in Experiment 1 the synthetic phonics group was

tested by advocates for the Jolly Phonics method, and the data from these children

were compared to children from a different education authority tested years

previously by Goswami’s research group). In Experiment 2 there were 111 children

in the ‘English standard’ group, 87 children in the ‘English phonics’ group, and 102

children in the German group. Non-word reading accuracy at Grades 1 (age 6) and

2 (age 7) was found to be equivalent for the two sets of English children. Both

groups were significantly poorer than the German children. At Grade 3, non-word

reading accuracy was much better, with the English children now making relatively

few errors (28% for the standard group, and 7% for the phonics group). This was the

first time that the two groups differed in non-word reading. By Grade 4, the

difference had disappeared. Landerl concluded:

the main difference between the two English groups was in Grade 3. In Grade 1, both

groups of English children committed a high number of errors, and in Grades 2 and 4

the difference was comparably small, because the English standard children performed

almost as well as the English phonics children. (p. 250)

The equivalence of the English standard and English phonics children is remarkable

given that the English standard teachers reported only administering one phonics

lesson a week.

One objection to cross-language comparisons is that cultural differences may

explain differences that are attributed to language per se. One empirical solution is to

look for cultures where bilingualism is common. In Wales, for example, children

grow up with either Welsh or English as their native tongue. Although they live in the

same geographical areas, their parents can choose whether to have them schooled in

Welsh or English. The schools themselves follow a similar (phonics-based, e.g. Jolly

Phonics) reading curriculum, are in the same catchment areas (in some towns, they

are side by side), and are administered by the same education authorities. Welsh is

an extremely consistent orthography. The mappings from graphemes to phonemes

are unambiguous. However, the syllable structure is complex. Welsh syllables can

have complex onsets, as in streic and cnau, or complex codas (consonant phonemes

after the vowel), as in sinc and peint. Hence, in terms of orthographic transparency,

Welsh and English differ markedly, whereas in terms of phonological complexity,

they are extremely similar. Spencer and Hanley (2003) utilised these unique features

of primary schools in Wales in order to carry out cross-language comparisons of



reading acquisition when cultural factors were held constant. The children were in

their second year of formal reading instruction, and were on average six years old.

Spencer and Hanley reported that the Welsh-speaking children were significantly

more accurate in both word reading and non-word reading compared to the English-

speaking children. For real words, the Welsh children read on average 81% of words

correctly, compared to 59% correct for the English children. For non-words, the

Welsh children read on average 78% of words correctly, compared to 44% correct

for the English children. This was not due to a disparity in the difficulty of the items

used in English and Welsh. For the non-words, for example, the Welsh-speaking

children decoded 68% of the English items correctly. The English-speaking children

only decoded 33% of the Welsh items correctly. The Welsh children had also

developed superior phonemic awareness (measured by a phoneme counting task).

Recall that the phonological complexity of words in English and Welsh is similar.

The finding that phonemic awareness was better in Welsh hence reflects the

importance of learning 1:1 grapheme–phoneme correspondences for the rapid

development of phonemic awareness. Spencer and Hanley concluded that their

findings ‘provided strong evidence that reading acquisition is heavily influenced by

the transparency of the alphabetic writing system’ (p. 14). When followed up a year

later, the Welsh-speaking children were still significantly better at word reading,

non-word reading and phoneme counting than the English-speaking children.

Children learning to read a transparent orthography will hence acquire reading skills

more quickly, even when teaching methods are based on synthetic phonics. When

tested again three years later, the differences between the Welsh-speaking and

English-speaking children had disappeared (Hanley et al., 2004). This was the sixth

year of formal reading instruction, and the children were now 10 years old. Hanley

et al. noted, however, that at this test point the English children had significantly

better reading comprehension skills than the Welsh children.

Clearly, cross-language comparisons suggest that linguistic factors rather than

teaching method per se explain the slower acquisition rates shown by English-

speaking children. However, one point made by Johnston and Watson (2005) was

that synthetic phonics should be taught first, before any other kind of reading tuition

was received (‘there is evidence that synthetic phonics is best taught at the beginning

of Primary 1’ [the first year of formal schooling], p. 9). A comparable training study

in this respect focusing on English only was reported by Walton et al. (2001). They

contrasted two types of phonics training, one based on phonemes (called letter

recoding) and the second based on larger units (called rime analogy). These two

training methods are characteristic of the wider literature comparing ‘synthetic’ with

‘analytic’ methods. This literature usually regards analytic phonics as going from

‘whole to part’, focusing initially on larger grain sizes (Moustafa & Maldonado-

Colon, 1998). Synthetic phonics is regarded as going from ‘part to whole’, focusing

initially on the smallest grain size of the phoneme. For the children studied by

Walton et al., who were pre-readers, this was the first direct tuition in reading

received by the participating children. The two groups were taught the same pre-

reading skills (phonological awareness of initial, medial and final phonemes,



rhyming and the letter–sound correspondences), and the treatments were controlled

so that the only differences between the training concerned whether the first steps of

reading were taught using a letter–sound recoding strategy or a rime analogy

strategy. Two training sessions of 25 minutes were given twice a week for 11 weeks.

Reading outcomes were then assessed. For word reading, a significant advantage

was found for the rime analogy group. For non-word reading, the two groups

performed at an equivalent level. In each case, the trained children outperformed a

control group who had received normal classroom teaching. A set of longitudinal

post-tests given four months later showed equivalent patterns. Walton et al.

concluded that ‘the study demonstrated that Grade 1 pre-readers with weak letter–

sound and phonological skills will develop reading ability relatively quickly if given

experience with the rime analogy or letter recoding strategies and the related pre-

reading skills’ (p. 178, emphasis added). The essential finding was that as long as

tuition was systematic, either large-unit (‘analytic’) or small unit (‘synthetic’)

phonics lead to similar gains.

Although carried out with very different samples, the experimental studies by

Walton et al. (2001), Landerl (2000) and Spencer and Hanley (2003) all reach the

same conclusion. The conclusion is that no one method of teaching phonics to

children learning to read in English appears to be superior to any other method. This

is exactly the same conclusion as was reached by the meta-analyses reported by the

NRP and by Torgerson et al. (2006). There is no empirical research base to justify

the Rose Report’s recommendation that the teaching of reading in England must

rely on synthetic phonics.

How do we teach early reading in English?

What, then, should teachers of early reading do? An empirical approach to this

question is to analyse the teaching methods featured in the phonics instruction

studies that were included in the NRP and Torgerson et al. (2006) reviews. Table 2

categorises the 43 studies included in the two reviews by the pedagogical emphasis of

the most effective instruction. Table 2 shows that the largest number of studies

featured instruction where phonics was contextualised at sentence and text level as

part of the teaching programmes. Interventions with an emphasis on discrete

phonics, where comprehension teaching and work with whole texts were covered

separately from the phonics instruction, were smaller in number but were also shown

to be effective. Whole language, whole-word teaching and broader approaches were

also shown to be effective in some studies.

The studies categorised in Table 2 show that a range of different teaching

approaches, including contextualised systematic phonics instruction and discrete

systematic phonics instruction, can be effective. However, in view of the

recommendation by the Rose Report that phonics teaching should be discrete, a

relevant question is, which teaching approach is more effective, contextualised

phonics instruction or discrete phonics instruction? In order to answer this question

there is a need to identify studies which compare directly the two kinds of



approaches. The selection criteria we used for studies which would address this

question were as follows:

1. explicit comparison of contextualised systematic phonics instruction versus

discrete systematic phonics instruction;

2. rigorous experimental design, preferably randomised controlled trial;

3. sample of typically developing readers;

Table 2. Methods of teaching reading implemented in research studies

Pedagogical emphasis

of effective interventions

Studies Total

Discrete phonics work Bond et al. (1995–96) (included spelling work, songs and

books); Foorman et al. (1991); Haskell et al. (1992)

(onset-rime group outperformed others); Johnston

and Watson (2004); Lovett et al. (2000) (sub-syllabic,

particularly focused on rime); Mantzicopoulos et al.

(1992); Oakland et al. (1998); Snider (1990); Stuart

(1999)

9

Whole language teaching,

including literature

teaching

Eldredge (1991) (included some systematic phonics);

Freppon (1991); Griffith et al. (1992); Klesius et al.

(1991); Wilson and Norman (1998)

5

Contextualised phonics

instruction: Phonics

instruction contextualised

with sentence-level and/or

text-level work such as

reading of connected text.

Reading recovery and

modified reading

recovery type programmes

included.

Berninger et al. (2003); Blachman et al. (1999); Brown

and Felton (1990); Evans and Carr (1985); Foorman

et al.(1997); Foorman et al. (1998); Greaney et al.

(1997); Martinussen and Kirby (1998); Santa and

Hoien (1999); Tunmer and Hoover (1993); Umbach

et al. (1989); Vickery et al. (1987)

12

Whole word teaching Gittelman and Feingold (1983) (included intersensory

phonics method); Lovett et al. (1990); Lovett and

Steinbach (1997) (included metacognitive training for

strategy monitoring); O’Connor and Padeliadu (2000)

4

Broad approach and/or

curriculum focus

Gersten et al. (1988); Lum and Morton (1984) (focus on

spelling more than reading); Marston et al.(1995);

Vandervelden and Siegel (1997) (argued against

sounds pronounced in isolation)

4

No statistically significant

gains overall for any

approach

Lovett et al. (1989) (showed different gains for decoding

skills and OWLS approaches); Silberberg et al. (1973)

(included synthetic phonics); Skailand (1971);

Torgesen et al. (1999) (advocates combined word-

level and comprehension teaching); Torgesen et al.

(2001) (Embedded Phonics as effective as Auditory

Discrimination in Depth); Traweek and Berninger

(1997)

6

Limited information about

teaching approaches

Fulwiler and Groff (1980); Leach and Siddall (1990)

(work with parents); Leinhardt and Engel (1981)

3



4. standardised measures of comprehension in addition to standardised

measures of word reading included.

None of the studies met all four criteria. This is because no studies of this kind have

been carried out with typically developing readers. Only two studies met three of the

criteria: the studies of children showing delayed development by Berninger et al.

(2003) and Torgesen et al. (2001).

With regard to the methodological criteria for selection of rigorous experimental

trials, we adopted Torgerson et al.’s (2006) quality assessment criteria, used to

identify the RCTs included in their main analysis (and in two cluster trials), which

were as follows: (a) reporting of method of random allocation; (b) sample size

justification; (c) intention to teach analysis, where all participants are analysed in

their original randomised groups as opposed to excluding participants after

randomisation; (d) blinded assessment of outcome.

Of the two studies that met our criteria for comparing discrete phonics instruction

with contextualised phonics instruction Torgesen et al. (2001) was the only study

that also met one of the experimental trial criteria, the intention to teach analysis.

Sixty children between the ages of eight and ten who had been previously identified

as learning disabled were recruited for participation in the study by Torgesen et al.

(2001). The children were randomly assigned to one of two groups: (a) auditory

discrimination in depth (ADD); (b) embedded phonics (EP). The teaching was

provided by educational therapists on a one-to-one basis in two 50-minute sessions

each day of the week. Teaching was provided over a period of 8–9 weeks until 67.5

hours of instruction were accomplished.

The ADD programme had three major goals. The first goal was to provide a basis

for accurate discriminations between phonemes. The second goal was to teach

children to use their knowledge of phonemes to monitor and represent sequences of

sounds in spoken syllables. The third goal was to teach self-monitoring skills that

allowed children to discover methods to correct themselves during their phonics

decoding attempts. All 44 English consonant and vowel phonemes were taught and

the programme provided extensive practice in reading and spelling individual words.

The EP programme was designed to provide direct, explicit instruction in word-level

reading skills while providing extensive opportunities to read and write meaningful text.

The programme included practice in reading sight words; practice in spelling sight

words; word games for fluency with sight words; phonics ‘mini lessons’; oral reading of

trade books or basal readers; and writing activities using sight words.

Outcomes were very similar for both instructional methods. The effects of

intervention were both substantial and stable over the course of the two-year follow-

up period. Torgesen et al. (2001) argued that the goal of any intensive intervention is

to produce large changes in reading ability that were maintained over time and that

‘the clear conclusion from this research is that the EP and ADD conditions were

equally successful’ (p. 51). In view of the fact that two approaches that varied

substantially were equally effective, Torgerson et al. argued that within structured

language approaches the components of instruction could be arranged according to

teacher and student preferences.



Berninger et al. (2003) did not meet any of the Torgerson et al. (2006) quality

assessment criteria but will be assessed here because it was a cluster RCT. The study

included the hypothesis that ‘instruction that is aimed at all the necessary

components of a functional reading system is more effective than instruction that

is aimed at an isolated component of the reading system’ (Berninger et al., 2003,

p. 103). In the study, second-grade teachers in eight schools serving diverse student

populations were asked to refer their poorest readers. A test battery covering

vocabulary, word identification and word attack was then administered by project

personnel in order to determine the final sample. Forty-eight pairs of children were

randomly assigned to four conditions: (a) explicit and reflective word recognition;

(b) explicit and reflective reading comprehension; (c) combined explicit word

recognition and explicit reading comprehension; or (d) treated control that only

practised reading skills without any instruction. The combined condition featured

explicit instruction in the alphabetic principle and explicit language cueing at the

word, sentence and text level to stimulate the text-based component of reading

comprehension. The teaching consisted of twice-weekly sessions of 20 minutes each

for a total of 24 lessons. The word recognition training began with 10 minutes of

explicit instruction in the alphabetic principle, both in and out of word context. In

any one session, 12 to 24 correspondences were introduced or reviewed, and

altogether, 107 correspondences were practised four or five times during the course

of the tutorial. During the next six minutes of training, children practised applying

these correspondences to decoding, which involved pronouncing single words.

During the last 10 minutes teachers provided explicit language cueing at the word,

sentence and text levels (Berninger, 1998), in order to stimulate the text-based

component of reading comprehension. Levels of language cueing to develop text-

based comprehension always contained word-level prompts, sentence-level prompts

and text-level prompts. An example of a text-level prompt included: ‘Tell the plot or

main events in the story so far’ (p. 107).

Berninger et al. reported that (c), the combined word recognition and reading

comprehension treatment, was the most effective in increasing phonological

decoding as shown by gains in non-word and real word reading. Given that most

at-risk readers have difficulty with phonological decoding, this is an important

finding. However, all three explicit treatments and the practice control treatment led

to significant improvements in reading comprehension. Berninger et al. (2003)

theorised that explicit instruction in comprehension combined with explicit word

recognition training facilitated learning to decode written words because it helped to

develop broad-based metalinguistic awareness. This was thought to generalise across

levels of language in the functional reading system.

The main conclusion to be drawn from this analysis of phonics instruction studies

is that a range of ways of teaching systematic phonics is effective. Considering

contextualised systematic phonics instruction versus discrete systematic phonics

instruction, there is a need for more research to compare the effectiveness of the two

approaches. In particular, no studies of typically-developing readers currently exist.

The two studies that have most rigorously investigated this question differ in their



findings. Torgesen et al. (2001) found that contextualised instruction and discrete

instruction were equally effective; Berninger et al. (2003) found that contextualised

phonics instruction was more effective for word decoding but that there was no

difference for reading comprehension.

Our analysis of empirical evidence does not support the advice based on the Rose

Report’s ‘simple view of reading’ that schools should put in place a ‘discrete

programme as the key means for teaching high-quality phonic work’ (DfES & PNS,

2006). Rather, the studies reviewed here suggest that many forms of systematic

phonics instruction are successful. Contextualised methods in which systematic

phonics are carefully integrated with whole text work and comprehension activities

are effective (Berninger, 2003, method c). According to our analysis, the key features

of phonics (learning grapheme–phoneme correspondences, learning to segment and

blend) as identified by Rose can be taught by a range of systematic methods.

Conclusion

As a result of the recommendations of the Rose Report (Rose, 2006), the UK

Government’s Primary National Strategy (PNS) Literacy Framework and its

guidance, which affects all early years and primary teachers in England, was

changed. The guidance which accompanies the PNS literacy framework makes it

clear that schools and early years settings should ‘put in place a systematic, discrete

programme as the key means for teaching high-quality phonic work’ and that

‘Shared and guided reading sessions should not be used to replace discrete phonics

teaching but they can provide opportunities to reinforce children’s developing

phonic knowledge and skills, in the context of achieving the ultimate goal of the

sessions, which is the development of comprehension’ (DfES & PNS, 2006, p. 7).

The UK Government in England, under the auspices of the Primary National

Strategy, has also published a detailed programme called ‘Letters and Sounds:

principles and practice of high quality phonics’ (Primary National Strategy, 2007)

which teachers are expected to follow. Following the teaching of general orientation

to sound discrimination in the nursery years, daily lessons for a six-week period

feature ‘discrete phonics’ teaching. Teachers must ‘teach at least 19 letters, and

move children on from oral blending and segmentation to blending and segmenting

with letters’ (p. 48). Application of this knowledge during the Letters and Sounds

lessons is limited to ‘Read or write a caption (with the teacher) using one or more

high-frequency words and words containing the new letter (week 3 onwards)’

(p. 49). This is followed by further discrete teaching, lasting for up to 12 weeks. The

purpose of this phase ‘is to teach another 25 graphemes, most of them comprising

two letters (e.g. oa), so the children can represent each of about 42 phonemes by a

grapheme’ (p. 74). Application at this stage is to ‘Read or write a caption or sentence

using one or more tricky words and words containing the graphemes’ (p. 75). This

pattern of a limited context for application of grapheme–phoneme correspondences

continues through year one (age 5 to 6) until year two (age 6 to 7), at which point

phonics instruction moves to an emphasis on spelling.



As demonstrated here, the Rose Report’s conclusion that synthetic phonics should

be adopted nationally in England as the preferred method for the teaching of reading

is not supported by empirical research evidence. Rather, as reviewed here, the

available research evidence supports the importance of systematic tuition in phonics

at a variety of grain sizes (e.g. phoneme, onset-rime). There is also evidence that

systematic phonics instruction that is contextualised can be effective although there

is a need for more research. As Torgerson et al. (2006) argued, the current empirical

database suggests that there is no evidence that ‘any one form of systematic phonics

is more effective than any other’ (p. 49).

The main research study which seems to have influenced the conclusions drawn

by Rose (2006), namely Johnston and Watson (2004), does not enable any

conclusions to be made concerning the optimal method of phonics instruction. In

particular, it does not support the claim that analytic methods (from word to sound)

are inferior to synthetic methods (from sound to word). This is because the children

were not taught the same material or the same component skills by the two methods.

Importantly, the findings reported by Johnston and Watson (2004) do not bear at all

on whether phonics should be taught discretely or in isolation.

The rationale for the change to synthetic phonics in England was also, in part,

built on a critique of the previous National Literacy Strategy ‘searchlights’ model.

This model advocated the use of four ‘strategies’ as a means to understanding text:

‘phonic (sound and spelling); knowledge of context; grammatical knowledge; word

recognition and graphic knowledge’ (Department for Education and Employment

[DfEE], 1998, p. 4). In the Appendix to the Rose Report, Stuart and Stainthorpe

(Rose, 2006, Appendix) argue, on the basis of their analysis of cognitive

psychological research, in favour of a ‘simple view of reading’. One of the most

significant differences between their simple view and the ‘searchlights’ model is that

word recognition processes and language comprehension processes are seen as two

separable dimensions of reading for the purposes of teaching: ‘The case for change

that we discuss … rests on the value of explicitly distinguishing between word

recognition processes and language comprehension processes’ (Rose, 2006, p. 74).

However, reading is not simple. Reading is one of the most complex achievements of

the human brain. Human brains that learn to read English may in fact develop extra

neural architecture that is not developed by brains learning to read more consistent

alphabetic orthographies (Goswami & Ziegler, 2006). In written language, as in

spoken language, the ultimate aim is communication and comprehension. We argue

that teachers are more likely to help children to achieve this aim if government

recommendations for practice are built on a rigorous synthesis of the full range of

evidence, including research about different languages and effective reading

teaching.

Notes

1. Significant parts of the research evidence referred to in this article, including its synthesis in an

abbreviated version of the current section entitled ‘The case for synthetic phonics’, were

submitted to the Rose Review as written evidence.



2. THRASS is the acronym for Teaching Handwriting, Reading and Spelling Skills (THRASS

UK Ltd).

References

Australian Government. Department of Education, Science and Training (2005) Teaching reading.

Report and recommendations. National enquiry into the teaching of literacy (Barton, Australia,

Department of Education, Science and Training).

Blachman, B., Tangel, D., Ball, E., Black, R. & McGraw, D. (1999) Developing phonological

awareness and word recognition skills: a two-year intervention with low-income, inner-city

children, Reading and Writing, 11, 293–273.

Bond, C., Ross, S., Smith, L. & Nunnery, J. (1995–96) The effects of the Sing, Spell, Read, and

Write program on reading achievement of beginning readers, Reading Research and

Instruction, 35, 121–141.

Brown, L. & Felton, R. (1990) Effects of instruction on beginning reading skills in children at risk

for reading disability, Reading and Writing, 2, 223–241.

De Cara, B. & Goswami, U. (2002) Similarity relations among spoken words: the special status of

rimes in English, Behavior Research Methods, Instruments & Computers, 34(3), 416–423.

Department for Education and Employment (DfEE) (1998) The National Literacy Strategy

framework for teaching (Sudbury, DfEE Publications).

Department for Education and Skills (DfES) & Primary National Strategy (PNS) (2006) Phonics

and early reading: an overview for headteachers, literacy leaders and teachers in schools, and

managers and practitioners in early years settings, Guidance Papers. Available online at: http://

www.standards.dfes.gov.uk/primaryframeworks/literacy/Papers/ (accessed 2 February, 2006).

Eldredge, L. (1991) An experiment with a modified whole language approach in first-grade

classrooms, Reading Research and Instruction, 30, 21–38.

Evans, M. & Carr, T. (1985) Cognitive abilities, conditions of learning, and the early development

of reading skill, Reading Research Quarterly, 20, 327–350.

Foorman, B. R., Francis, D. J., Fletcher, J. M., Schatschneider, C. & Mehta, P. (1998) The role of

instruction in learning to read: preventing reading failure in at-risk children, Journal of

Educational Psychology, 90, 37–55.

Foorman, B. R., Francis, D. J., Winikates, D., Mehta, P., Schatschneider, C. & Fletcher, J. M.

(1997) Early interventions for children with reading disabilities, Scientific Studies of Reading,

1, 255–276.

Foorman, B. R., Francis, D., Novy, D. & Liberman, D. (1991) How letter–sound instruction

mediates progress in first-grade reading and spelling, Journal of Educational Psychology, 83,

456–469.

Freppon, P. (1991) Children’s concepts of the nature and purpose of reading in different

instructional settings, Journal of Reading Behavior, 23, 139–163.

Frith, U. (1998) Editorial: Literally changing the brain, Brain, 121, 1051–1052.

Frith, U., Wimmer, H. & Landerl, K. (1998) Differences in phonological recoding in German-

and English-speaking children, Scientific Studies of Reading, 2, 31–54.

Fulwiler, G. & Groff, P. (1980) The effectiveness of intensive phonics, Reading Horizons, 21, 50–54.

Gersten, R., Darch, C. & Gleason, M. (1988) Effectiveness of a direct instruction academic

kindergarten for low-income students, The Elementary School Journal, 89, 227–240.

Gittelman, R. & Feingold, I. (1983) Children with reading disorders. Efficacy of reading

remediation, Journal of Child Psychology and Psychiatry and Allied Disciplines, 24, 167–191.

Goswami, U. (2006) Neuroscience and education: from research to practice?, Nature Reviews

Neuroscience, 7, 406–413.

Goswami, U. (2008) Learning to read across languages: The role of phonics and synthetic

phonics, in: K. Goouch & A. Lambirth (Eds) Understanding phonics and the teaching of

reading: critical perspectives (UK, Open University Press), 124–143.



Goswami, U. & Ziegler, J. C. (2006) Fluency, phonology and morphology: a response to the

commentaries on becoming literate in different languages, Developmental Science, 9(5),

451–453.

Goswami, U., Gombert, J. E. & de Barrera, L. F. (1998) Children’s orthographic representations

and linguistic transparency: nonsense word reading in English, French, and Spanish, Applied

Psycholinguistics, 19(1), 19–52.

Goswami, U., Ziegler, J., Dalton, L. & Schneider, W. (2001) Pseudohomophone effects and

phonological recoding procedures in reading development in English and German, Journal

of Memory & Language, 45, 648–664.

Gough, P. B. & Tunmer, W. E. (1986) Decoding, reading and reading disability, Remedial and

Special Education, 7, 6–10.

Greaney, K., Tunmer, W. & Chapman, J. (1997) Effects of rime-based orthographic analogy

training on the word recognition skills of children with reading disability, Journal of


Griffith, P., Klesius, J. & Kromey, J. (1992) The effect of phonemic awareness on the literacy

development of first grade children in a traditional or a whole language classroom, Journal of

Research in Childhood Education, 6, 85–92.

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 10 years of age, Quarterly Journal of Experimental Psychology,

57(8), 1393–1410.

Haskell, D., Foorman, B. R. & Swank, P. (1992) Effects of three orthographic/phonological units

on first-grade reading, Remedial and Special Education, 13, 40–49.

Johnston, R. & Watson, J. (2003) Accelerating reading and spelling with synthetic phonics: a five year

follow up (Edinburgh, The Scottish Executive Education Department; Insight 4).

Johnston, R. & Watson, J. (2004) Accelerating the development of reading, spelling and phonemic

awareness skills in initial readers, Reading and Writing, 17, 327–357.

Johnston, R. & Watson, J. (2005) The effects of synthetic phonics teaching of reading and spelling

attainment: a seven year longitudinal study. Available online at: http://www.scotland.gov.uk/

Resource/Doc/36496/0023582.pdf (accessed 10 December 2006).

Klesius, J., Griffith, P. & Zielonka, P. (1991) A whole language and traditional instruction

comparison: overall effectiveness and development of the alphabetic principle, Reading

Research and Instruction, 30, 47–61.

Landerl, K. (2000) Influences of orthographic consistency and reading instruction on the develop-

ment of nonword reading skills, European Journal of Psychology of Education, 15, 239–257.

Leach, D. & Siddall, S. (1990) Parental involvement in the teaching of reading: a comparison of

hearing reading, paired reading, pause, prompt, praise, and direct instruction methods,

British Journal of Educational Psychology, 60, 349–355.

Leinhardt, G. & Engel, M. (1981) An iterative evaluation of NRS: ripples in a pond, Evaluation

Review, 5, 579–601.

Lloyd, S. (1998) The phonics handbook (3rd edn) (Chigwell, Jolly Learning).

Lovett, M. & Steinbach, K. (1997) The effectiveness of remedial programs for reading disabled

children of different ages: does the benefit decrease for older children? Learning Disability

Quarterly, 20, 189–210.

Lovett, M. W., Ransby, M. J., Hardwick, N., Johns, M. S. & Donaldson, S. A. (1989) Can

dyslexia be treated? Treatment-specific and generalized treatment effects in dyslexic

children’s response to remediation, Brain and Language, 37, 90–121.

Lovett, M. W., Warren-Chaplin, P. M., Ransby, M. J. & Borden, S. L. (1990) Training the word

recognition skills of reading disabled children: treatment and transfer effects, Journal of

Educational Psychology, 82(4), 769–780.

Lum, T. & Morton, L. (1984) Direct instruction in spelling increases gain in spelling and reading

skills, Special Education in Canada, 58, 41–45.



Mantzicopoulos, P., Morrison, D., Stone, E. & Setrakian, W. (1992) Use of the search/teach

tutoring approach with middle-class students at risk for reading failure, Elementary School

Journal, 97, 573–586.

Marston, D., Deno, S., Kim, D., Diment, K. & Rogers, D. (1995) Comparison of reading

intervention approaches for students with mild disabilities, Exceptional Children, 62, 20–

37.

Martinussen, R. & Kirby, J. (1998) Instruction in successive and phonological processing to

improve the reading acquisition of at-risk kindergarten children, Developmental Disabilities

Bulletin, 26, 19–39.

Moustafa, M. & Maldonado-Colon, E. (1998) Whole-to-parts phonics instruction: building on

what children know to help them know more, The Reading Teacher, 52, 448–458.

National Institute of Child Health and Human Development (NICHD) (2000) Report of the

National Reading Panel. Teaching children to read: an evidence-based assessment of the

scientific research literature on reading and its implications for reading instruction: Reports of the

subgroups. NIH publication no. 00-4754 (Washington, DC, US Government Printing

Office).

Oakland, T., Black, J., Stanford, G., Nussbaum, N. & Balise, R. (1998) An evaluation of the

dyslexia training program: a multisensory method for promoting reading in students with

reading disabilities, Journal of Learning Disabilities, 31, 140–147.

O’Connor, R. E. & Padeliadu, S. (2000) Blending versus whole word approaches in first grade

remedial reading: short-term and delayed effects on reading and spelling words, Reading and

Writing, 13, 159–182.

Primary National Strategy (2007) Letters and sounds, Available online at: http://www.standards.

dfes.gov.uk/clld/las.html (accessed 21 December 2007).

Rose, J. (2006) Independent review of the teaching of early reading (Nottingham, DfES Publications).

Rutter, M. (2006) Is sure start an effective preventive intervention? Child and Adolescent Mental

Health, 11(3), 135–141.

Santa, C. & Hoien, T. (1999) An assessment of early steps: a program for early intervention of

reading problems, Reading Research Quarterly, 34, 54–79.

Seymour, P. H. K., Aro, M. & Erskine, J. M. (2003) Foundation literacy acquisition in European

orthographies, British Journal of Psychology, 94, 143–147.

Silberberg, N., Iversen, I. & Goins, J. (1973) Which remedial reading method works best? Journal

of Learning Disabilities, 6, 18–27.

Skailand, D. B. (1971) A comparison of four language units in teaching beginning reading,

paper presented at the Annual Meeting of the American Educational Research Association, New

York.

Snider, V. (1990) Direct instruction reading with average first-graders, Reading Improvement, 27,

143–148.

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), 1–28.

Stuart, M. (1999) Getting ready for reading: early phoneme awareness and phonics teaching

improves reading and spelling in inner-city second language learners, British Journal of


Stuart, M. (2006) Teaching reading: why start with systematic phonics teaching?, Psychology of

Education Review, 30(2), 6–17.

Torgerson, C. J., Brooks, G. & Hall, J. (2006) A systematic review of the research literature on the use

of phonics in the teaching of reading and spelling (London, Department for Education and

Skills).

Torgesen, J. K., Alexander, A. W. & Wagner, R. K. et al. (2001) Intensive remedial instruction for

children with severe reading disabilities: immediate and long-term outcomes from two

instructional approaches, Journal of Learning Disabilities, 34(1), 33–58.



Torgesen, J. K., Wagner, R. K. & Rashotte, C. A. et al. (1999) Preventing reading failure in young

children with phonological processing disabilities: group and individual responses to

instruction, Journal of Educational Psychology, 91, 579–593.

Traweek, K. & Berninger, V. (1997) Comparisons of beginning literacy programs: alternative

paths to the same learning outcome, Learning Disability Quarterly, 20, 160–168.

Tunmer, W. & Hoover, W. (1993) Phonological recoding skill and beginning reading, Reading and

Writing, 5, 161–179.

Umbach, B., Darch, C. & Halpin, G. (1989) Teaching reading to low performing first graders in

rural schools: a comparison of two instructional approaches, Journal of Instructional

Psychology, 16, 22–30.

Vandervelden, M. & Siegel, L. (1997) Teaching phonological processing skills in early literacy: a

developmental approach, Learning Disability Quarterly, 20, 63–81.

Vickery, K., Reynolds, V. & Cochran, S. (1987) Multisensory teaching approach for reading,

spelling, and handwriting, Orton-Gillingham based curriculum, in a public school setting,

Annals of Dyslexia, 37, 189–200.

Walton, P. D., Walton, L. M. & Felton, K. (2001) Teaching rime analogy or letter recoding

reading strategies to pre-readers: effects on prereading skills and word reading, Journal of


Watson, J. & Johnston, R. (1998) Accelerating reading attainment: the effectiveness of synthetic phonics

(The Scottish Office: Education and Industry Department). Available online at: www.

standards.dfes.gov.uk/pdf/literacy/rjohnston-phonics.pdf

Wilson, K. & Norman, C. (1998) Differences in word recognition based on approach to reading

instruction, Alberta Journal of Educational Research, 44, 221–230.

Wimmer, H. & Goswami, U. (1994) The influence of orthographic consistency on reading

development—word recognition in English and German children, Cognition, 51(1), 91–103.

Ziegler, J. C., Stone, G. O. & Jacobs, A. M. (1997) What is the pronunciation for -ough and the

spelling for u/? A database for computing feedforward and feedback consistency in English,

Behavior Research Methods, Instruments & Computers, 29(4), 600–618.



british educational research journal synthetic phonics and the

Documents