House of Commons - Education and Skills

Select Committee on Education and Skills Eighth Report

3 Teaching Methods

16. The focus of this inquiry was on the methods used to teach children to read in the classroom. We therefore took evidence on the approach recommended by the National Literacy Strategy as well as on the alternative programmes currently being used by teachers in schools. The status of the NLS as guidance means that schools and teachers are free to use whatever method they prefer to teach reading, although some of our evidence pointed to pressure from LEAs and others to adopt the NLS approach and to use the approved teaching materials published alongside the strategy.[13]

17. The NLS has been strongly criticised by some, who claim that it is not equally successful with all children, and others who even see its approach as educationally damaging. In this section of our report we examine the teaching methodology set out in the NLS and the critiques put forward by two main groups: those supporting a greater use of phonics and those preferring a broader experience of text and language. We then consider what the available research evidence can tell us about these different approaches.

The National Literacy Strategy

18. The National Literacy Strategy promotes the teaching of reading in its broadest sense from the beginning of a child's education. This includes decoding, comprehension, grammatical understanding and a more general experience of different books and texts. In the NLS scheme, none of these aspects is given priority at any particular time in a child's acquisition of reading. For example, when a child learning to read encounters a word he or she does not know, the child is encouraged to 'work out' the word either by inferring from narrative context or syntax, by sounding out the word or by recognising the shape of the word from a previous encounter. This approach has been termed the 'searchlights' model and is graphically represented in the diagram below.

19. The searchlights model assumes that reading can best be taught using by using a range of strategies simultaneously. The rationale for this approach is that children will learn to read most effectively by exploiting a diverse range of strategies, but it also claims to be more effective for those children who respond better to one particular approach than to others: "The 'searchlights' metaphor attempts to describe a methodology for teaching reading which optimises the range of 'cues' or inputs for the pupil, enabling them to cross refer between them. The more 'searchlights' that are switched on, the less critical it is if one of them fails".[14]

20. In oral evidence, Dr Kevan Collins, National Director of the Primary National Strategy expanded on the rationale behind the NLS approach. He emphasised not only the 'searchlights' model, but also the fact that the NLS established a regular curriculum time devoted exclusively to reading:

"What the child does is they bring the four aspects of the searchlights to bear. They bring their knowledge of phonics to get the first consonant. The dominant consonant is the first thing and they get to bits of the word. They use other information, the context, maybe the picture, the evolving story. They use their syntactic knowledge, the kind of grammar and pattern of English, and they use their graphic knowledge. They bring those things to bear to try and solve that word. There are some words at the beginning of reading which you cannot read and then you have got this great other asset which is an adult to help you. What we encourage children to do is to be active learners and to try new things. I have a problem with texts that are completely bound by what children already know. It is quite helpful to have some words in a text which require you to be active and begin to problem solve because I think that is what a lot of reading is about."[15]

"The principal problem [before the NLS] was that there was no place where literacy (and I think reading is the priority in the early years) where reading and writing was taught. There was no moment in the day when this was our focus. It was lost in an integrated curriculum and literacy teaching. I think we would agree on this requires some very focused and structured teaching in the early years."[16]

21. The NLS has been subject to revision since its inception. Stephen Twigg MP, the then Parliamentary Under Secretary of State at the DfES and now Minister for School Standards, told us that the Department has "sought at every stage of the strategy to keep on top of the research to ensure that we are engaging with academic evidence both from people in this country and from people in other parts of the world, and that is a continuous process".[17] Dr Collins agreed with this claim, saying:

"As well as that historical body of research, which we draw on deeply, we also in an on going way continually reflect on learning as it occurs, and a key area of that has been the developing research around phonics, which has been a piece of literacy learning, a core element, which we have continually updated and developed and, as we move through our support for schools, we keep drawing on it and evolving and developing our resources, our materials and support based on the research. So it draws on a historical base and continues to respond to evolving research that is happening every day."[18]

22. The phonic 'searchlight' has perhaps been subject to the most revision in the lifetime of the NLS. Written evidence submitted by the DfES states:

"Every year Ofsted has reviewed the implementation and practices of the National Literacy Strategy, and now the Primary National Strategy. The programme has therefore had an unmatched regular flow of evidence to inform its continuing development. The DfES has also been proactive in seeking out fresh evidence to inform the way we support the teaching of reading. Last year, in response to a 2002 Ofsted report which identified some weaknesses in the teaching of phonics, the department convened a phonics seminar which drew on a range of expertise, including the Reading Reform Foundation, and was independently chaired by Professor Greg Brooks.

The findings of the seminar were that a major redirection of the phonics element of the NLS was neither necessary nor appropriate. The papers from the seminar, including Professor Greg Brooks' concluding report, can be found on the Primary National Strategy section of the DfES Standards Site. We have also separately published all the background research underpinning the teaching of English at Key Stage 3 in 'Roots and Research', which is a research review compiled by Colin Harrison of Nottingham University.

As a result of the seminar described above, the supplement to Progression in Phonics, Playing with Sounds, was produced for primary schools in Spring 2004, to make it a more detailed and fully resourced programme."[19]

23. In addition, Kevan Collins, Director of the National Primary Strategy, emphasised to us that "the priority for information when you are young is developing the phonics knowledge".[20] Despite these alterations, the NLS approach has been severely criticised, notably by those who support a much greater and more exclusive focus on phonics in the early teaching of reading.

Phonics

24. In recent years the use of 'phonics' has increased in popularity as a method of teaching reading. Put simply, phonics programmes emphasise the importance of establishing a secure correspondence between written letters and the sounds of language in the learner's mind. Phonics programmes often begin by teaching the single letters of the alphabet as sounds (for example, 'kicking k' rather than the letter name 'kay'), later moving on to more complex digraphs[21] and, finally, the irregular spellings of the English language, which do not follow phonic rules.[22] As with other teaching methods, there are a number of variations in the way in which phonics can be taught. The two main variants are 'analytic' and 'synthetic' phonics. Analytic phonics does not necessarily break down words into their smallest units or phonemes.[23] The 'onset-rime' method, for example, divides words into openings (onset) and endings (rime). So 'street' is broken down into 'str-eet'. In contrast, synthetic phonics involves segmenting words into the smallest unit of sound, then teaching children to blend these sounds together to form words. So the word 'street' is broken down into five components: 's-t-r-ee-t'. This is sometimes referred to as 'all-through-the-word' teaching. Our inquiry stimulated a particularly strong response from supporters of synthetic phonics, who argued that this approach leads to much greater improvements in reading standards than the searchlights model put forward by the NLS.

25. The evidence we received in favour of synthetic phonics is based on the belief that an early ability to 'decode' words is the key to later success in reading. Phonics programmes give children the ability to 'sound out' words on a page, even if they do not always understand the meaning of all those words. Supporters of phonics argue that comprehension can be built up subsequently upon this foundation:

"You are not taught contradictory messages and you are not given, in the first instance, words which are awkward, words with complicated phonics, even if they are regular, irregular words, you are given the words that work. By the time you start to be introduced to more difficult words but useful words for reading text, you already understand the principles of the alphabetic code and how to decode words […] within half a term, or a term, you can have a whole cohort of children able to do the most fundamental skill, which is sound out and blend for reading and segment the spoken word for spelling. Now that is very powerful and compared with the mixture that is here [in the NLS], where some phonics will be taught, the results are pretty dramatic.

You can measure how well children can decode words; you can measure how well children can comprehend. What you cannot do is understand their full measure of comprehension if they cannot decode well. They might not do well in a comprehension test, but actually their oral comprehension would be better, they are just not sufficiently competent at reading the words on the page." (Debbie Hepplewhite, Reading Reform Foundation)[24]

If a child can decode a text effortlessly, it means then that all their resources, all their energies go into working out what the book is all about. If you have to work very hard at reading every single word that you come across, asking yourself 'Shall I use a picture cue? Shall I use a context cue? Shall I use a picture cue with a letter cue? Shall I read on a little bit and try to work out what the word is in the middle?', the child cannot make that decision whilst they are reading. The children who are at the lower end find that almost impossible to do, so then they get the image of themselves as not being very good readers." (Ruth Miskin, ReadWrite Inc)[25]

26. An important feature of the phonics approach, which diverges from the methods recommended by the NLS, is that children are only taught to read through texts fully within their current phonological ability. So, although children might encounter words they do not understand, they are not given texts they cannot decode and are therefore not expected to infer words from context or syntax. Many supporters of phonics deplore what they see as 'guessing' strategies and view them as actively damaging.

27. Some submissions further suggest that a phonics approach is particularly effective for children who are at risk of becoming poor readers. This group includes boys, children with English as an additional language and those from socio-economically deprived backgrounds. In her written evidence, Sue Lloyd, co-author of the 'Jolly Phonics' teaching programme, states: "The NLS initiatives were supposed to correct the imbalance between the results of boys and girls, as well as prevent the serious reading failure of the bottom 25%. Fairly soon it was obvious that this initiative was not working for these particular groups […] synthetic phonics is the most effective way to teach reading […] it is possible to have all children, apart from the 2% with clinical disorders, reading and writing fluently before they enter Year 3".[26]

28. Many of those who submitted evidence to our inquiry argued that the introduction of phonics 'first and fast' would raise reading rates significantly from their current level. In response to this critique, supporters of the NLS approach gave two responses. Firstly, the DfES argued that phonics is and always has been an important component of the NLS. It pointed to Progression in Phonics and Playing with Sounds, the two phonics programmes it provides to teachers.[27] It also noted that teachers are able to choose from a number of proprietary phonics programmes now available (including Jolly Phonics) to supplement the NLS.[28] Secondly, Dr Kevan Collins put forward the argument that phonics should not be taught alone, and claimed that it was inappropriate only to teach decoding ability, even at a very early age of reading instruction:

"For us at the Early Years the phonics is the dominant learning but what we are saying is you do not live there; you are there for a while as you put the learning together and then you are moving up, but we run with the grain of what children do as active learners, and they need all this learning."[29]

Asked whether phonics programmes produce better readers, Dr Collins responded:

"No, I would not say that. I would say they work best in being able to demonstrate that children have learned the phonic knowledge. They do not demonstrate that children are learning the other knowledge around the development of context, syntax and the other parts of reading which, in my view, are also important. They do not demonstrate that knowledge at all."[30]

29. Dr Collins also took issue with the idea of limiting children's experience of texts to those within their existing phonological knowledge:

"Controlling the reading environment of a child is a tricky business because there might be the odd book that you have control over but the truth is that children are active readers right across the curriculum and throughout their lives, and what you have to do is give them strategies that allow them to be engaged and positive about that approach and not think, 'I can only read when I read these little books and everything else I cannot read'."[31]

30. We have received conflicting evidence from supporters of phonics regarding the possibility of amending the status of phonics within the NLS. Dr Morag Stuart, Reader in Psychology at the Institute of Education, University of London and an advocate of the phonics methodology, told the Committee that the NLS "is broadly correct to the extent that it recognises that reading should be taught and that there is a role for some kind of phonics teaching in how reading should be taught",[32] but warned that "it has missed an opportunity to get a generation of teachers who understood about reading. I should like to see different models of reading adopted in the National Literacy Strategy guidance to teachers which were in accordance with research evidence and knowledge about reading".[33]

31. In contrast, other witnesses described the NLS approach as fundamentally flawed and called for it to be withdrawn entirely. Debbie Hepplewhite of the Reading Reform Foundation said that the apparent success of the NLS should not disguise its essential methodological problems:

"The National Literacy Strategy brought a huge impetus to the teaching of reading: massive influx of reading material in the form of big books and sets of reading books. A bit of a rod for teachers, saying 'You need to teach literacy. You have to make it very high profile in your schools. You have to plan it very thoroughly. There is such a thing as word level, sentence level, text level and you must account for all these areas'. The difference it made was that the middle to above-average children have absolutely flown on that extra impetus. In my opinion, what it did not do was still train the teachers how to teach reading in the most effective way and this is where the sort of, say, bottom third are still failed, because there is still a lot of grey area about how to do it […] I do not think it [the NLS] is broadly correct. I think its programmes are contradictory."[34]

32. A further complication arises from a difference of opinions concerning the type of phonics that the NLS currently recommends. The DfES claims:

"The approach advocated through the National Strategies is a synthetic phonics approach, as it relies on direct teaching and the recognition and blending of letters to form words. It does not rule out the possibility that children will supplement their knowledge and understanding of a text through inference, and hence could be described as drawing on some elements of an analytical approach. However, it is clear that synthetic phonics is the principal method of instruction."[35]

33. In contrast, other witnesses have suggested that the NLS programmes do not constitute a rigorous form of synthetic phonics. Psychologists Professor Rhona Johnston and Dr Joyce Watson suggest it is more akin to analytic phonics, an approach they view as much less effective:

"In analytic phonics, whole words are presented and pronounced by the teacher, and the children's attention is only subsequently drawn to the information given by letter sound correspondences. The National Literacy Strategy's Progression in Phonics uses an analytic phonics approach, supplemented by a substantial phonemic awareness training programme. The sounding and blending element for pronouncing unfamiliar words is only introduced after children have learnt to read words by sight. Typically in the areas of England in which we have carried out studies of the National Literacy Strategy, it would not be until the third term of the first year at school that the most advanced children would be made aware of the importance of letter sound correspondences in all positions of words, which enables sounding and blending to be taught. An analytic phonics scheme such as this is usually not completed until the end of the third year at school. In synthetic phonics programmes sounding and blending is taught at the start of the year, before books are introduced, and the basic programme can be completed in a period of 2 to 4 months.

The new supplement to Progression in Phonics that was issued last May, Playing with Sounds, still emphasises rhyme and phonemic awareness training as a precursor to learning to read and spell. Early on it also emphasises sound-to-letter training for spelling, rather than letter-to-sound training for reading. In fact 16 letters are taught before children are shown how to blend letter sounds for reading."[36]

34. In accordance with the available evidence, the DfES now seems to have accepted that phonics is an essential methodology in teaching children to read. The present debate revolves around the status of phonics within early teaching of reading and the type of phonics programme that should be used.

Reading for Pleasure

35. Our inquiry also took evidence from other critics of the NLS who advocated alternative teaching methods. We found particular concern from some who considered that the NLS Framework for Teaching is too rigid in structure and leads to a dull and mechanical experience for pupils. These submissions suggested that there is a danger that such uninspiring lessons will demotivate pupils, so that although they may be proficient in reading and writing the English language, they have little enthusiasm to learn and any benefits that this proficiency may have afforded them are lost.

36. Written evidence submitted by the Association of Teachers and Lecturers (ATL) suggests that NLS guidance is self-defeating:

"Within the National Literacy Strategy, the focus on reading and writing, and the consequent marginalisation of speaking and listening has had perverse consequences in terms of exciting children's interest in language and literature. Even more importantly, children need to talk and to experience a rich diet of spoken language, in order to think and learn. Reading, writing and number may be the acknowledged curriculum 'basics', but talk is 'arguably the true foundation of learning'. ATL believes that foundation stage guidance (and that for key stage 1) needs to be strengthened to ensure that practitioners are less constrained by the (perceived or real) prescription of the literacy strategy. Guidance needs to focus on the role of play, of real engagement with stories, of real-life experiences of reading, rather than the small building blocks of literacy. The requirement for the full literacy hour to be in place by the end of reception should be removed.

Teachers were 'unanimous in their view that the Framework's emphasis on language was undervaluing the literature entitlement in the national curriculum. There is widespread regret that covering all the objectives in the time available would exclude the reading of whole novels and teachers would prefer to sacrifice some of the objectives rather than lose this."[37]

37. The Early Childhood Foundation supported this view in their evidence, suggesting that the formal teaching of reading, whether by the NLS, phonics or by some other means, could profitably be delayed until a later stage in a child's cognitive development:

"Phonics teaching is not appropriate for children in pre-school or reception classes. It depends on the accurate pronunciation of letter names (which is open to confusing variation e.g. haitch for H) and their initial sounds. If the practitioner demonstrates the sounds singly, as for example in h-o-p, and creates a gap between the initial and subsequent sounds it is very difficult for the child to 'hear' the complete word. Presented with material which is out of context or uninteresting, children may well repeat sounds or words by rote, but not assimilate these into their knowledge base.

Recent papers published in Scandinavian countries stress the child-centred approach to learning. In many nursery and 'first' schools in Denmark, Sweden and Finland the children are not subjected to restricting formal lessons but are provided with 'systematic support for their growth, development and learning'. (Finnish Family Policy document). In Sweden, 'Educational activities are based on the children's individual capabilities and are linked to what the child has already experienced and learned. Children are encouraged to engage in their own activities and discover things for themselves. The importance of play for a child's development and learning is emphasised both in preschool education and in school age childcare, and is included in the national curriculum for compulsory schools.' (Sweden SE Childcare in Sweden). In the Council of Europe, the Committee of Ministers' Recommendation to member states on child day-care, concerning the care and education of children from birth to eight years [Rec. (2002) 8], stresses play and talking as very important elements in children's learning. All the countries mentioned above start formal schooling later than in the UK and have literacy outcomes far higher than ours, so maybe the approach speaks for itself. [38]

38. In defence of the NLS strategy, the DfES stated that "there are appropriate structured programmes in place for children below school age, which develop their communication and literacy skills within the context of play-based learning. We engage parents and the wider community fully in supporting children to learn to read and to enjoy reading".[39] This view was supported in oral evidence given by Neil McClelland, Director of the National Literacy Trust:

"Whilst we support the use of phonics and synthetic phonics and certainly, in comparison with what things were like in the past, we have moved forward massively and radically well, it is that wider context of that work which we believe is critically important. We obviously agree with the Early Childhood Forum on the issue of early language and books and reading in the home from the earliest age; we have initiatives of our own that build that. As far as the specific question about the strategy is concerned, we certainly believe that more emphasis can be put on encouraging children to read fuller texts, to read more for pleasure, to have more opportunities for extended writing. I actually think the strategy has come to terms with that. We are involved with an initiative funded by the DfES about creating whole-school reading communities. We believe that surrounds and supports the strategy and is a necessary additional component in this complex picture of getting children, particularly children from many disadvantaged communities and homes, to be fluent, motivated, enjoying readers."[40]

39. Current evidence on the best age to begin the formal teaching of reading does not point to a clear answer. It is difficult to account for variables such as the language of instruction and the developmental level of the individual child concerned. Whatever method is used in the early stages of teaching children to read, we are convinced that inspiring an enduring enjoyment of reading should be a key objective. This can be endangered both by an overly formal approach in the early years and by a failure to teach decoding.

Other Teaching Methods

40. We are aware that many different approaches to the teaching of reading exist, apart from those mentioned specifically above. 'Whole word' methods were particularly prominent in the 1970s and 1980s. This approach rests on the hypothesis that many children start to read by learning words, for example, their name, as a whole word, without breaking it down into sounds. Teaching proceeds by familiarising children with the shape of each word as they learn it (this method is also known as 'look and say'). The 'whole language' approach shifts the emphasis not only away from individual sounds, but also from individual words. Children are thought to learn most effectively when immersed in language and literacy through the books read with or by adults, word games and rhymes. There are also those who contend that reading cannot be taught and that it is naturally acquired whatever techniques of intervention teachers employ. Although these views were discussed by witnesses, we have not received any evidence specifically promoting these approaches over and above others in the course of this inquiry and they appear to have been largely overtaken by more recent developments in teaching methodology.

Compromise?

41. The evidence we took during our inquiry reveals a number of competing methodologies concerning the teaching of reading. Those involved in establishing and implementing the NLS themselves accept that its 'searchlights' model represents a pragmatic compromise between these divisions. This is unsurprising, given the context in which it arose, which Ruth Miskin described to the Committee:

"When the NLS was first started I was on one of the advisory bodies right at the beginning with John Stannard. John Stannard was in an almost impossible position when the NLS was being written, because he had all of those pressure groups saying 'We want this' 'We want this' and then I would come in and others would say 'No, but we want this'. What we actually got was a plethora of eclectic messages to teachers, so they had not just one sort of phonics, but they had three sorts of phonics […] You then had the whole word lobby, which was 'Let's all learn 250 words very quickly' as well. So three sorts of phonics and the word level work; then the real books lobby and that is just the beginning. Then you have other lobbies to look at how you organise it, whether you should put children in groups so they could actually apply the knowledge they have at their level, or whether you should mix them all up together. Even when you have decided on a method which actually works, you then have to look at the most effective way to implement it."[41]

42. The question before this Committee was to consider whether the NLS 'compromise' represents a synthesis of best practice drawn from a range of approaches, or a confusing mix of methods that is holding back further improvements in standards of reading. We sought to answer this question by examining the results of research studies conducted in schools, which compared different methods of teaching reading and their consequences for pupils' achievement.

Research evidence

43. Evidence submitted by supporters of phonics as a teaching methodology appears to demonstrate that schools which adopt pure phonics programmes achieve significant improvements in results. In oral evidence, Sue Lloyd cited the achievements of a West Country school using her programme:

"94% are achieving level four at the Key Stage 2 SATS [i.e. the expected level for their age], compared with 77% in England, and this is a large primary school with a low entry assessment. This is a poor social area, a very large school. Look at the level five: 65% achieved level five in this school [i.e. above the level expected for their age] and only 26% in the rest of England. Boys? 33.3% of boys achieved level five in writing compared with 11% in the rest of the schools and it goes on like this as well […] the boys can do just as well as the girls. No significant difference between children with summer birthdays and no children with English as an additional language on the SEN register."[42]

This study also showed that the school achieved just 6% reaching only level three or below, compared to 15% nationally and 0% reaching level two or below, compared to 7% nationally.[43]

44. In some cases where phonics was used 'first and fast', almost 100% of pupils achieved Key Stage test results of the level expected of their age or better and some schools also showed a reduction in the percentage of children identified as having special educational needs. This includes a number of schools serving communities with a high level of deprivation, such as the Kobi Nazrul School in Tower Hamlets, where Ruth Miskin was headteacher. This school achieved results in the top 5% in England, despite taking a significant proportion of children with English as an additional language.[44]

45. When we asked Dr Kevan Collins, Director of the Primary National Strategy, whether he could produce similar results from schools following the NLS programme, he responded:

"In looking at primary schools' results in the Key Stage 1 tests, it is important to bear in mind that published figures take account of all pupils in year 2: including those with special educational needs; those only recently arrived in the country, and who may have little or no English on arrival; and those who are unable to take the tests. Nonetheless, in 1,962 primary schools every pupil achieved at least level 2 for their Key Stage 1 reading in 2004. In another 1,622 primaries, between 95% and 100% of children did so. In these schools, that will typically mean a single child not reaching the expected level. So, in over 20% of all primary schools in England, all or almost all KS1 pupils are reading at at least level 2 by age 7. And these are schools which will be using the support provided by the National Primary Strategy - properly adapted and supplemented for local needs and practices - including the Strategy's central, but not exclusive, role for synthetic phonics."[45]

46. We discussed the best way of comparing the NLS with other programmes, such as phonics, with Dr Morag Stuart, Reader in Psychology at the Institute of Education, University of London. She told us that her research indicated positive benefits for children who received a dedicated synthetic phonics programme, but warned that individual studies had significant limitations. She advised us that a larger comparative study was necessary to confirm her preliminary findings:

"Structured phonics teaching: proof that it works. There is the proof from the [United States] national reading panel's survey of the literature which suggests that structured phonics teaching works better than no phonics teaching or less structured phonics teaching.[46] It is very difficult in the real world to do the kind of research that you would like to be done. It is terribly difficult to match children so that they are comparable on all possible things. We did try to do that in the study that I conducted. We had 50 children taught for a term using Jolly Phonics which is a very nice programme for five-year-olds and it is fun. We had 50-odd children who were not taught [with that programme]. We pre-tested them on a range of measures of language and phonological skills and letter-sound knowledge and various things that we did not expect to change as a result of the teaching and other things that we did expect to change as a result of the differential teaching. We managed to match our groups on almost everything and where we were unable to match groups, we took account of that in the statistical analysis we did. So it is not impossible to do that sort of research, but it is difficult. What our research showed was that the Jolly Phonics teaching was definitely much, much more successful in making children fluent readers of words than the non-phonics teaching. However, that is not the sort of comparison that you are asking for, which is comparing the phonics as taught in the NLS with different phonics teaching programmes. I do not know of any research that has done that."[47]

Dr Stuart later added: "We have some inkling of what works: we do not know the fine details of how best to do things. We have not had proper comparative studies looking carefully at the best way to do things and the best way to do things for different sorts of children, because children differ."[48]

47. We accept Dr Stuart's conclusion that there has not so far been any decisive research evidence determining the value of dedicated synthetic phonics programmes directly compared to the mixture of phonics and other strategies in the NLS. This is not necessarily an argument for the preservation of the NLS: many witnesses argued that the 'searchlights' model itself is not backed up by robust research evidence. Dr Stuart commented, "The model of reading which is presented to teachers [in the NLS] which is this black hole of four things [i.e. four 'searchlights'] operating and disappearing into a text is completely and utterly misleading and bears no relation to any research on reading that I know of."[49]

48. In response to this critique, Dr Collins told us of the research background to the NLS:

"There are two kinds that we draw on. We draw on a body of historical research, and as the literacy strategy and the numeracy strategy were drawn together they were well founded on the core research. I would say that in terms of the literacy strategy, we were very fortunate it was a seminal piece of research done in the late 90s in the United States through Marilyn Jaeger Adams which basically did a full review of all literacy research and informed our work as well as the key research in England drawing on the work in Australia, New Zealand and this country. As well as that historical body of research, which we draw on deeply, we also in an ongoing way continually reflect on learning as it occurs, and a key area of that has been the developing research around phonics, which has been a piece of literacy learning, a core element, which we have continually updated and developed and, as we move through our support for schools, we keep drawing on it and evolving and developing our resources, our materials and support based on the research. So it draws on a historical base and continues to respond to evolving research that is happening every day."[50]

We do not consider that this answer provides conclusive evidence of the National Literacy Strategy's basis in sound research.

49. The Committee was extremely interested to see the recent publication of a seven year longitudinal study, The Effects of Synthetic Phonics Teaching on Reading and Spelling Attainment.[51] This study, carried out in Clackmannanshire, Scotland by Professor Rhona Johnston and Dr Joyce Watson, looked at 300 children in the first year of the Scottish primary school system. It compared three different teaching methods: synthetic phonics; analytic phonics; and an analytic phonics method that included systematic phonemic awareness teaching. At the end of the programme, those children who had been taught by synthetic phonics were found to be 7 months ahead of the other two groups in reading. The other two groups were then given the synthetic phonics programme as well and the progress of all the children was followed for 7 years. At the end of this period, all the children were tested and found to be achieving significantly higher levels in word reading and spelling that would be expected of their chronological age. Unusually, boys were found to be outperforming girls.

50. The Clackmannanshire study is of interest as it appears to show long-lasting benefits from early synthetic phonics training. However, it does raise a number of additional issues. The study showed that although reading comprehension was still significantly above chronological age at the end of the seventh year at school (3.5 months ahead), the advantage was smaller than it had been at the end of the second year at school (7 months ahead).[52] It would be useful to return to these children again in future to see whether the gains in word recognition and spelling continue to persist. The children involved in the study displayed a higher than average level of socio-economic disadvantage, but the study was not able to control for this, to determine which programme was most effective for those with the greatest level of disadvantage. As all children eventually took the synthetic phonics programme, there was no direct comparison of a dedicated phonics programmes with a programme like the NLS 'searchlights' model, which mixes phonics with other approaches. Nevertheless, the Clackmannanshire study is an important addition to the research picture, which increasingly points to synthetic phonics as a vital part of early reading education.

51. We took evidence from one of the authors of the Clackmannanshire study, Professor Rhona Johnston. We asked Professor Johnston whether she thought research should be undertaken comparing a group of children who were being taught by the National Literacy Strategy programme with a similar group, matched very closely, who were being taught by synthetic phonics. She replied:

"Yes, I think that absolutely needs to be done to establish what the facts are. I should stress that my research has been paid for entirely by the Scottish Government. It actually voted £18 million in 1997 to look at early intervention and that money was given to the regions. We were invited by Clackmannanshire to do the study, but that money was only given out if they did pre-tests and post-tests of an experimental and a control group using standardised tests. This is what I think should happen in England."[53]

52. In view of the evidence from the Clackmannanshire study, as well as evidence from other schools where synthetic phonics programmes have been introduced, we recommend that the Government should undertake an immediate review of the National Literacy Strategy. This should determine whether the current prescriptions and recommendations are the best available methodology for the teaching of reading in primary schools. We therefore strongly urge the DfES to commission a large-scale comparative study, comparing the National Literacy Strategy with 'phonics fast and first' approaches. This study should establish:

The relative effectiveness of approaches to teaching reading, such as synthetic phonics, analytic phonics and the methods recommended in the National Literacy Strategy;
The effect of mixing phonics instruction with other methods of teaching, compared to 'phonics fast, first and only';
How long any gains afforded by a particular programme are sustained;
The effect of teaching texts which go beyond a child's existing knowledge of phonics compared to that of limiting instructional texts to those within a child's current decoding abilities;
The effectiveness of different approaches with particular groups of children, including boys/girls, those with special educational needs and those with a high level of socio-economic disadvantage.

The Study should:

Measure and compare attainment by means of standardised testing and not Key Stage test results;
Measure attainment in all the components of literacy (word recognition, reading comprehension, narrative awareness, etc.);
Use control groups to take account of factors which may have a bearing on reading outcomes, for example: teacher knowledge and ability; socio-economic background; gender.

13 Q 12 Back

14 Ev 35 Back

15 Q 248 Back

16 Q 249 Back

17 Q 150 Back

18 Q 151 Back

19 Ev 36-37, paragraphs 27-29 Back

20 Q 196 Back

21 A digraph or bigraph is a pair of letters used to write one sound. This is often, but not necessarily, a sound (or more precisely a phoneme-see footnote 23) which cannot be expressed using a single letter in the alphabet used for writing, e.g. ch, th, sh. Back

22 An exception is made for very common words which would not normally be taught immediately, such as 'the', so that the child is able to access whole texts from an early stage. Back

23 A phoneme is the smallest contrastive unit in the sound system of a language. Back

24 Qq 16 and 91 Back