Page 1 of 1

Recent blog post by Greg Brooks

Posted: Thu Nov 30, 2017 1:08 pm
by chew8
This blog post by Prof. Greg Brooks has appeared on the website of the Australian Association for Research in Education:

Below are some extracts with my comments in bold:

So who needs to be taught phonics and when?

Some children are enabled to bridge that gulf by being read to copiously, and joining in the reciting of the texts they have heard so often they have them off by heart, until they twig the essential insight that what they are saying is represented by what they can see. For them, phonics is not only unnecessary, but may be a hindrance. Therefore phonics has no place in the teaching of reading to young fluent readers, and testing their ‘phonic knowledge’ is irrelevant and risks causing them to regress in their learning.

There may be two or three categories of ‘young fluent readers’ here: 1. those who enter Reception having already ‘twig[ged] the essential insight’ in the way that Brooks outlines, 2. those whose pre-school experience includes some phonics instruction from parents or others, and 3. those who catch on very quickly once they start school and are reading fluently by the time they take the Phonics Screening Check (PSC) at the end of Year 1.

Some years ago, I did a straw poll of teachers, asking how many children entered Reception already reading well. The consensus was that in most years it was none at all and that it was never more than one or two in a class of 30. That tallies with what I’ve seen in my voluntary work. If the numbers are really so low, it shouldn’t be too difficult for teachers to cater appropriately for these children while still teaching the rest what they need by way of phonics, as my son’s teacher did when he started school in Scotland in 1974 already reading well, having been taught phonics at home but also having been read to a lot – she checked his reading at the start of the year, said it was ‘super’ and asked us if she could let him read in the book corner while she taught the rest, which we thought was just the right thing for him.

Whichever category children fall into, my experience suggests that doing the PSC at the end of Y1 causes no problems for them or their teachers. In my voluntary work, I’ve done past PSCs with children in all three categories – they’ve been very co-operative and have usually whizzed through in 60 seconds or less, so it’s not time-consuming for them or the teacher. Moreover, I’ve been able to follow the later progress of these children, in some cases up to Y5, and they’ve shown no sign of regressing.

What of the children who arrive at school not yet reading? Their most important immediate task is to learn to read, so for them the purpose of phonics is to provide a quick start on the identification of regularly-spelt words, alongside the essential (for English, with its complex orthography) learning of some basic high-frequency but irregularly-spelt words as sight words.

1. The great majority of children enter school not yet reading and needing that quick start on identifying regularly-spelt words in the most child-friendly way possible.

2. Hardly any words consist entirely of irregular grapheme-phoneme correspondences. The synthetic phonics approach is not to teach irregular words purely as ‘sight’ words but to teach them in a way which enables children to see both the regularities and the irregularities. This works well

The experimental evidence shows clearly that phonics in this context works for both normally-developing children and those who are falling behind. But the same body of evidence also shows that (a) the teaching must be systematic and not incidental; (b) it must be embedded in a broad and rich language and literacy curriculum, because there is much more to reading than just word identification, and therefore phonics alone does not constitute teaching children to read.

As far as I know, there’s no evidence that teachers are teaching phonics at the expense of a broad and rich language curriculum.

The flawed case for teaching only synthetic phonics

Jim Rose mostly stuck to saying phonics teaching must be systematic, but in places elided that into saying that systematic phonics is synthetic phonics, which the experimental evidence did not justify, and still doesn’t. There is as yet no evidence that any one form of phonics teaching produces better progress than any other form of phonics.

Brooks is probably referring here to the 2006 review of research by Torgerson, Brooks and Hall. Rhona Johnston and Joyce Watson have re-analysed the data on synthetic phonics presented in that review and have found serious flaws. See Johnston R.S. and Watson, J (2016) The trials and tribulations of changing how reading is taught in schools: Synthetic phonics and the educational backlash. In K Durkin, HR Schaffer (Eds) The Wiley Handbook of Developmental Psychology in Practice: Implementation and Impact, pp 203-221, and Johnston, R. and Watson, J. (2014) Teaching Synthetic Phonics, 2nd edition. Sage (Learning Matters).

I myself have some evidence which I’ve mentioned previously in this RRF forum and which suggests that children’s reading has improved since the PSC was introduced: the results I’ve had in administering the British Ability Scales Word Reading test and the New Salford Reading Test ( ... BAS#p50360 – the whole thread is relevant). I don’t know for sure if the children are typical by national standards, but even if they aren’t, these test results suggest a big improvement at local level since the PSC was introduced – and the improvement is not just in word-reading but also in comprehension.

Following the Rose Report there was a noticeable increase in the number of phonics-based intervention schemes for struggling readers in England, but it was only after the change of government in 2010 that strong official pressure was put behind synthetic phonics, often using a flawed and partial interpretation of the research evidence, and Rose’s conflation of systematic phonics with synthetic phonics. This was expressed in the misleading slogan-like mantra that ‘Synthetic phonics is the best way to teach children to read’, ignoring all the caveats about embedding phonics in the broader curriculum and the dearth of supporting evidence.

I’ve already made some relevant comments above. Brooks himself says in this blog-post that ‘the overriding aim of phonics is the efficient identification of unfamiliar printed words’. For beginners, virtually all words are unfamiliar in their written form, and synthetic phonics is the simplest way for them to work out beginner-type words, requiring only that they should know what sounds to say in response to letters and how to blend those sounds together. Other approaches, such as analytic phonics and onset-rime, add extra complications, as they require that children should already know some words as ‘sight’ words and be able to see how those ‘sight’ words can help them with reading unfamiliar words. Incidentally, there’s nothing to stop children from drawing on ‘sight’ word knowledge or onset-rime strategies in reading the pseudo-words in the PSC if they are able to do so, though sounding out and blending is simpler.

Teachers don’t need a national test to tell them about their own students

What of those children who don’t ‘get it’ the first or second time? There are a few for whom phonics simply doesn’t work, but they are rare and exceptional. An Australian friend who taught in an Infants school (Years 1-2) in England for over 30 years says that she was unable to unlock the door of initial literacy for just one child in all that time. There are others who struggle and fail to progress well. Observant teachers know perfectly well who they are, and need deep professional knowledge to understand them and work round their difficulties. Teachers don’t need a test to identify children who are struggling. When teachers in England were asked about the phonics test a great many said it didn’t tell them anything they didn’t already know.

But the improvement in PSC results over the years shows that many teachers did not know what their pupils could achieve in terms of phonics until the PSC gave them the incentive to teach phonics more systematically. At the end of Y1, when they do the PSC, children will still be encountering many visually unfamiliar words in their text-reading, so competence at working these words out is vital at this stage. The check helps to identify children who lack that competence, and probably also schools which are producing too many children who lack it.

]What teachers of initial literacy do need is better support for helping the strugglers, which was supposed to be part of the follow-up to the phonics test, but is notable by its absence. Money put into that would be well spent, which the money spent on the phonics test is not.

Is it really notable by its absence? What about this? - ... nt-readers.

In the first three years of national operation, the phonics test in England cost £44,000,000 – what a waste! Spend your Australian dollars on good professional development instead!

The £44,000,000 will be a good investment if it means that far fewer children need extra support throughout school and grow up to be illiterate or semi-literate adults.

Re: Recent blog post by Greg Brooks

Posted: Mon Dec 04, 2017 5:24 pm
by john walker
Well done, Jenny! What an excellent rebuttal.
Thank you and best wishes, John

Re: Recent blog post by Greg Brooks

Posted: Mon Dec 04, 2017 7:41 pm
by chew8
Thanks, John. The Brooks blog-post seems to be a potted version of a chapter by him in a new book edited by Margaret Clark (of 'Young Fluent Readers' fame). I've now ordered the book, so should soon be able to see what he and all the other contributors say.

Jenny C.

Re: Recent blog post by Greg Brooks

Posted: Mon Dec 04, 2017 8:07 pm
by john walker
Hi Jenny,
I'll be interested to see what you make of Clarke's new book.
Have you ever come across Alison Clark, who writes the Spelfabet blog? She recently reviewed Clarke's Reading the Evidence: Synthetic Phonics and Literacy Learning, which I feel sure must be the same. Here's the link: ... t-phonics/
Judging from what Alison has to say, you'd better not start reading it later on in the evening, unless, that is, you are desirous of putting yourself into a soporific state ;)
Best, John

Re: Recent blog post by Greg Brooks

Posted: Wed Dec 06, 2017 10:07 pm
by chew8
The Clark book arrived today - I had already read what was available on line, but have so far read only the Brooks chapter and one other (Gardner) in the hard copy. I'll focus here on the Brooks chapter, in keeping with the title of this thread.

1. He writes that the 2006 Torgerson, Brooks and Hall review found evidence in favour of 'systematic' phonics, but ‘not enough evidence to decide whether synthetic or analytic phonics is more effective’. I mentioned in my first post in this thread that Johnston and Watson have found errors in the Torgerson et al. analysis. Brooks goes on to write ‘I was convinced then, and still am, that theory suggests that synthetic phonics is more coherent than analytic phonics as a strategy for young learners’. I agree - more below. He then gives a diagram showing exactly how synthetic phonics (s.p.) works: it starts with the written word - the graphemes in that are identified, those graphemes are translated into phonemes, the phonemes are blended into the whole spoken word, and the meaning of the word is then understood. Again I agree, so no problems there.

2. He then writes ‘I have tried, and failed, to produce a corresponding representation of how analytic phonics is thought or meant to work, and have not found one in the literature’. That made me re-visit what he had said in his 2003 ‘Sound Sense’ paper:

( ... 1&type=pdf)

He talks about s.p. and analytic phonics (a.p.) on pages 11 and 12. He says that in a.p., sounds are not pronounced in isolation – that wasn’t true of the typical Scottish version of a.p. at the time of the Clackmannanshire study, but that’s not important for present purposes. What I think is important is that the Brooks description of a.p. in 2003 and his confessed inability to represent it as he was able to represent s.p. in 2017 shows that it doesn’t teach children directly what they need to do in order to work out unfamiliar words – in fact, he said in 2003 that it involved ‘inferential learning’ and children having to ‘deduce’ things. That, for me, is what makes a.p. less suitable for beginners than s.p. For beginners, virtually every written word is unfamiliar, and it makes sense to teach them exactly how to tackle these words, starting simple, rather than to teach them ‘sight’ words and expect them to deduce from these how to read unfamiliar words.

3. He cites the 2016 Machin et al. paper as showing that s.p. produced an across-the-board improvement at 5 and 7’, but no ‘average effect’ at 11, although there were ‘lasting effects for children who could be considered as having been at risk of under-achievement initially...’. The children in that study, however, had been taught by the Early Reading Development Pilot approach, which fell far short of good s.p.

4. His final short paragraph refers to the Phonics Screening Check (PSC) as ‘an abomination’. We should remember at this point that he approves of s.p. in theory and recognises that it is useful in the identification of unfamiliar words. It is therefore an obvious strategy for children to use for at least half the PSC items, though there's nothing to stop them from using other strategies, too, if they have them – e.g. they might use sight-word knowledge to deduce pronunciations for unfamiliar items if they were able to. Whatever strategies the 2016 PIRLS children used, they did better than the 2011 children - i.e. children who had done the PSC did better in PIRLS, which tests comprehension, than children who had not done the PSC. Moreover, the PIRLS report for England has this on p. 14:

'PIRLS 2016 marks the first opportunity to assess how performance in the phonics check, introduced in 2012 and taken by pupils in England near the end of Year 1, relates to performance in PIRLS. There is a 0.52 correlation between performance on the two assessments, indicating a moderate relationship. Pupils who scored full marks in the Year 1 phonics check are also the highest scoring group in PIRLS 2016, with an average overall PIRLS score of 617. In contrast, pupils who did not reach the ‘expected standard’ in the Year 1 phonics check perform below England’s overall average, with lower phonics check scores being associated with decreasing average PIRLS scores.’

As I see it, the PSC is a quick check on children's ability to decode unfamiliar words at a stage in their schooling where many are still having to do a lot of this in their text-reading and need to be given extra help a.s.a.p if they are not yet competent at an age-appropriate level. I myself don’t think that the PSC is causing an over-emphasis on decoding at the expense of comprehension, but I accept that some people do take this view. What, then, do they make of the 2016 PIRLS results compared with the 2011 PIRLS results?

Jenny C.