The following was published on the RRF website in March 2006.
A copy was also sent to Jim Rose, who expressed his approval of it.
The Reading Reform Foundation warmly welcomes this report, which was published on 20 March 2006. We believe that Jim Rose and his team have dealt very clearly and fairly with the issues. The following headings are not all from the report itself, but reflect areas of particular interest to the RRF.
The searchlights model and the relationship between decoding and comprehension
1. A highlight of the report is the way it deals with the relationship between decoding and comprehension and, in the process, exposes the weaknesses in the ‘searchlights’ model which has always been at the heart of the National Literacy Strategy (NLS) Framework for Teaching. The report states that at the time the searchlights model was introduced in 1998, it ‘encapsulated what was considered to be “best practice” in the teaching of reading. This offered the opportunity, which the NLS has fully exploited, of gradually increasing the emphasis on the importance of phonics for young and for struggling readers’. The implication may be that the original NLS authors knew that what was considered to be ‘best practice’ at the time was not in fact best practice (there was already plenty of research showing this), but went along with it because challenging it too openly might have reduced the chances of having some phonics on the agenda from the start and then gradually increasing the amount.
2. Whatever the reasons for the placing of the searchlights model at the heart of the NLS, the Rose report has now shown convincingly that this model is inappropriate for beginners. There is a section on this in the body of the report (paragraphs 113-130) but the most detailed reasoning is in Appendix 1, which is by Professors Morag Stuart and Rhona Stainthorp. They show how the ‘searchlights’ model has confused ‘word recognition’ and ‘language comprehension’, and they present the case, complete with numerous research references, for explicitly delineating the role played by each. They refer to the ‘simple view of reading’ proposed by Gough and Tunmer in 1986, which is that ‘Reading is the product of decoding and comprehension’. By ‘comprehension’, here, Gough and Tunmer mean comprehension of spoken language. Stuart and Stainthorp point out that ‘when children begin to learn to read, they have already made considerable progress in their language development. They already understand much of what is said to them and can express themselves so that others can understand them. They have not been “taught” language. However, the time limited task that is word reading is generally achieved as a result of direct instruction’ (italics original). In other words, what beginners need above all is not to be taught to understand language (they can already do that) but to be taught to decode. The concept of learning word-reading skills as being a ‘time-limited’ task is a very useful one which surfaces several times in the Rose report. People need to realise that if phonic word-reading skills are taught first and fast, decoding quickly becomes automatic and children can devote their conscious attention more fully to the meaning of what they read.
Prevention rather than cure
3. The Rose report rightly points out that ‘quality first teaching’ can greatly reduce the number of children who fall behind, commenting, ‘Wave 1 teaching needs to be strengthened before any intervention programme is considered’ (para. 142), and ‘It is not the purpose of intervention work to shore up weak teaching at Wave 1’ (Recommendations, p. 71). It also stresses the importance of ensuring that interventions dovetail well with first-time classroom teaching: ‘effective intervention work should focus on the phonic skills children have already met in their mainstream classes but may need more help and time from skilled adults to strengthen and secure those aspects they had not first understood’ (para. 153). This should surely mean that intervention programmes such as Early Literacy Support and Reading Recovery are abandoned, as they encourage strategies (for example guessing at words from pictures and context) which will not fit in with the type of first-time teaching recommended by Rose. The same applies to the Early Reading Development Pilot which the DfES was running during the period in which the Rose review was in progress, although this was intended for mainstream use rather than for intervention.
4. The report states, ‘Good assessment should track performance in all four strands of language: speaking, listening, reading and writing, and identify strengths and weakness in children’s knowledge, skills and understanding, especially those related to mastering word recognition skills’ (para. 59). In para. 61, it praises schools which were observed assessing
· recognition of letters (and groups of letters, such as digraphs)
· the ability to sound out phonemes
· the ability to hear and blend phonemes
· the reading of phonically regular words
· the reading of some irregular words.
Conversely, it criticises schools where gaps were found in assessment: for example, one where it ‘was not possible to find out, from records, whether an individual pupil could sound out and blend specific phonemes, could segment the sounds in words to spell or knew the different ways of writing specific phonemes – the fundamental building blocks for learning to read and write independently’ (para. 234). The implication, we hope, is that while children are still learning the ‘building blocks’, assessment focusing on these will supersede miscue analysis and running records, which do not focus nearly sharply enough on these building blocks.
5. The Rose report comments that ‘nearly half the schools visited did not give enough time to teaching children the crucial skill of blending (synthesising) sounds together. Instead, teachers emphasised hearing and identifying the initial, final and then the middle sounds in words. So, even when children knew sufficient letter-sound correspondences, their lack of skill in blending sounds meant that they could not apply their knowledge to read words they had not seen before’ (para. 232 – see also para. 237). This goes to the heart of the difference between synthetic phonics and the NLS approach which has been in force until now. The NLS approach over-emphasises the hearing side of phonics and does this in a way (initial, then final, then middle sounds in words) which is at odds with the order in which letters need to be looked at and sounds need to be produced in word-reading. By contrast, synthetic phonics teaches children to look at letters from left to right and produce sounds for them, leaving them with the sounds in the right order for blending (synthesising). The blending of letter-prompted sounds is essential for reading words which are unfamiliar in their written form (as most words are to children in the early stages of learning to read). It is essential, too, if a programme is to be called ‘synthetic phonics’.
6. Although the report does not come down firmly in favour of allowing children to practise for a while on decodable texts, it does say that there is ‘some force’ in the view that this is desirable as children can ‘benefit from “quick wins” in practising phonics skills and gaining confidence in reading a whole, albeit short, book. Using such books as part of the phonics programme does not preclude other reading. Indeed it can be shown that such books help children develop confidence and an appetite for reading more widely’ (para. 82). It is good to have the Rose team recognising these things, and also recognising that decodable books of good quality are now available.
7. The report suggests that there is no good reason why children should not be taught the names as well as the sounds of letters at an early stage. Synthetic phonics teachers agree that many children cope well with this, but find that the weaker ones can be confused by it and would therefore argue in favour of delaying the introduction of letter-names at least until digraphs are introduced, which is usually within the first half-term.
Training for teachers
8. The Rose report recognises that current training ‘is not always as effective as it should be’ (para. 176). It quotes comments from an Ofsted survey (para. 179) and from a new Initial Teacher Training tutor (para. 182) to the effect that training in how to teach phonics has been particularly weak. Ensuring that both initial and in-service training are much better in future is clearly essential but will not be easy in view of the large numbers of trainers who still find the multi-cueing ‘searchlights’ type of approach attractive.
Research and common sense
9. Where directly applicable research findings were felt to be inconclusive, the team decided that observation based on common sense was a reasonable guide: ‘While robust research findings must not be ignored, developers of national strategies, much less schools and settings, cannot always wait for the results of long-term research studies. They must take decisions, based on as much firm evidence as is available from a range of sources at the time, especially from replicable and sustainable best practice’ (para. 31); ‘…notwithstanding the uncertainties of research, there is much convincing evidence to show from the practice observed that, as generally understood, “synthetic” phonics is the form of systematic phonic work that offers the vast majority of beginners the best route to becoming skilled readers’ (para. 47). The report comments, ‘Among other strengths, this is because it teaches children directly what they need to know.’ The RRF would endorse this: synthetic phonics explicitly teaches beginners grapheme-phoneme correspondences and the exact way in which they need to use them for reading – they need to produce sounds in response to letters and then blend the sounds together (i.e. synthesise them) into normally-spoken words. Synthetic phonics teachers also teach spelling, at least at first, as the other side of the same coin: they teach children to segment spoken words into phonemes and write down letters for the phonemes. The Rose report, too, recognises that it makes good sense to teach the ‘reversibility’ of reading and spelling (see paragraphs 44, 217, 221).
10. The same common-sense and observation-based approach is used to deal with some objections often raised to teaching phonics systematically to beginners – for example, that such teaching is too ‘formal’, that early systematic phonics leads to negative attitudes to reading, and that ‘one size does not fit all’. The Rose team found that the synthetic phonics teaching which they observed was not ‘formal’ in a bad sense: ‘the best work was formalised in design but taught creatively and with due regard for individual differences in, for example, children’s rates of learning’ (para. 50). They found that the children actually ‘showed positive attitudes to reading and writing’ (para. 36). On the ‘one size does not fit all’ point, they comment that ‘all beginner readers have to come to terms with the same alphabetic principles if they are to learn to read and write’ and that ‘leading edge practice bears no resemblance to a “one size fits all” model of teaching and learning, nor does it promote boringly dull rote learning of phonics’ (para. 34).
The next steps (April to September 2006)
11. It is very encouraging that Jim Rose and his team have seen good synthetic phonics in action: they have realised how well it works with beginners and how groundless the objections are that are commonly raised to it. We must now wait to see whether the draft of the revised framework for teaching, due out soon after Easter, is faithful to the Rose report. If it is not, we must say so during the consultation period. The RRF believes that if the Rose report is properly heeded in both the revised framework and in the training which teachers receive from now onwards, improvements in reading and spelling standards should be noticeable within a year.