A document with the above heading was circulated to LEAs, apparently in March 2005. This document contains some disturbingly questionable statements.
It claims that the phonics in the NLS is as effective as the phonics in the Clackmannanshire study ‘if it is done properly’, as in ‘schools that follow a systematic approach as suggested in Progression in Phonics’. But it bases this claim on the assumption that achieving Level 4 or above in the Key Stage 2 test is exactly equivalent to achieving at or above chronological age level in the particular standardised test of comprehension used in the Clackmannanshire study – the Macmillan Group Reading Test. In fact, however, this equivalence has not been demonstrated. In a letter which I wrote to Ruth Kelly on 14 March 2005, I mentioned the performance of the Clackmannanshire children on the Macmillan test and asked, ‘Is it known that NLS teaching produces better comprehension scores among comparable children on this test?’. I also mentioned the Clackmannanshire children’s spelling score on the Wide Range Achievement Test. The reply which I had from the DfES, dated 19 April, stated, ‘We do not administer these tests so we don’t have any national information about the performance of children taught through NLS approaches on these particular tests’.
Phonics PhAQs recognises the same problem, stating that there are ‘issues to be resolved over the comparability of National Curriculum test scores and tests which produce reading ages’. How, then, can it state in the very same sentence that ‘the current 83% figure for the number of children achieving at or above their chronological age in reading seems to compare well with the results in this study’? If there are ‘issues of comparability’ (and there are), the comparison simply should not be made. The fact is that there is no foundation for official claims that children reaching Level 4 or above in the Key Stage 2 tests are comprehending as well as, or better than, the Clackmannanshire children. The reverse is more likely, as one would expect the Macmillan test to have been restandardised if it were true that 83% of eleven-year-olds were ‘achieving at or above their chronological age in reading’.
Another problem with the Phonics PhAQs document is that its definition of ‘analytic phonics’ does not tally with the definition in the Clackmannanshire study. There is of course always some latitude in the way that things are defined, but if a judgement is to be made about whether the phonics in the NLS is or is not like the ‘analytic’ phonics in the Clackmannanshire study, then it is the Clackmannanshire definition which should be used. With analytic phonics as defined in this study, children are taught about grapheme-phoneme correspondences systematically (rather than learning about them incidentally ‘by deduction ... from texts’, as suggested in Phonics PhAQs), but
(a) more slowly than in synthetic phonics programmes,
(b) usually only after text-reading has begun, and
(c) with the emphasis at first only on letters and sounds at the beginnings of words.
(a) and (c) seem true of the NLS, and (b) seems at least partially true – text-reading in the NLS may not begin before the teaching of letter-sound correspondences begins, but the two certainly seem to proceed in tandem. Press reports which ‘equate the analytic phonics used in the Clackmannanshire study with the approach to phonics in the Strategy’, far from being ‘entirely incorrect’ as PhAQs states, may be quite close to the mark.
Phonics PhAQs states, ‘Phonics in the Strategy is fundamentally synthetic – it teaches children to hear and segment the sounds in a spoken word and shows them the letters for those sounds so that they can spell the word. It also teaches blending so that children can read words using the letters they have learned’. The order here reflects the way that the original NLS publications and subsequently Progression in Phonics and Playing with Sounds have placed more emphasis on segmenting for spelling than on blending (synthesising) for reading.
PhAQs continues, ‘However, in the course of reading and spelling, children who have understood that words consist of letters which represent phonemes, will deduce information about words independently. Once a child has understood that the same letters can be used in different words and that these correspond to the sounds they [sic] hear in words, they can attempt to read and spell more advanced words than those in the phonics programme. The Strategy would want to encourage this problem solving behaviour. In that sense it also endorses analytic phonics’. It needs to be noted, though, that analytic phonics, as defined in Phonics PhAQs, does not have a monopoly on this type of grapheme-phoneme-based problem-solving – synthetic phonics encourages it, too.
The type of problem-solving behaviour which synthetic phonics does not encourage is the use of grammar, context and pictures for word-identification purposes. This type of problem-solving is encouraged by the NLS, according to Dr Kevan Collins’s reply to Q248 in the Education and Skills Committee session on 8 December 2004: ‘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 ... 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’. This sanctioning of word-identification which is not grapheme-phoneme based seriously weakens Ruth Kelly’s claims on 2 March and 21 March that ‘synthetic phonics is at the heart of our approach’ and that the literacy hour approach ‘is now almost entirely based on synthetic phonics’ (see Editorial of this Newsletter).
All in all, then, this Phonics PhAQs document does not seem to give a very accurate account of the NLS’s position in relation to the Clackmannanshire study.