Evidence Check 1: Early Literacy Interventions - Science and Technology Committee Contents

Supplementary memorandum submitted by the Every Child a Chance Trust (LI 14a)


  1.1 This submission is from the Every Child a Chance Trust, a charity which aims to unlock the educational potential of socially disadvantaged children through the development and promotion of evidence based, early intervention programmes. The charity was initiated by the KPMG Foundation and is funded by a coalition of business partners and charitable trusts. It is a successor body to the KPMG-led coalition that set up the Every Child a Reader 2005-08 pilot initiative to tackle the problem of a static 6% of children (35,000 a year) entering secondary school without even the most basic skills in reading and writing (below National Curriculum Level 3—that is, with literacy skills at or below the level of the average seven year old).

  1.2  In 2008 the Trust provided evidence to Sir Jim Rose's dyslexia review. This submission is based on that evidence.


  2.1  An issue we sought to answer in our evidence to the Rose dyslexia review was whether Every Child a Reader works for dyslexic children.

  2.2  We summarised evidence from the evaluation of Every Child a Reader commissioned by the KPMG Foundation from the Institute of Education. This showed children receiving Reading Recovery making, on average, 20 months progress in word reading age over a year compared to 7 months in the control group receiving a range of other interventions. At follow-up one year later, children who had received Reading Recovery were still doing as well as their chronological age band. They had an average reading age of 7 years 9 months compared to 6 years 9 months in the comparison group. Their superior performance was evident on a wide range of tests—word recognition, phonics, reading comprehension, spelling and grammar. The children who had been involved in Reading Recovery did better than the national average for all children (across the whole ability range) in their end-of-key-stage National Curriculum assessments. 86% of children who had received Reading Recovery achieved the expected level for their age (Level 2+) in Reading, compared to 84% of all children nationally. 77% achieved Level2B+ compared to the national 71%. In Writing, 83% of children who had received Reading Recovery achieved the expected level for their age compared to 80% of all children nationally.

  2.3  The children involved in this evaluation were selected at the start of the study as the very lowest achieving children in Year 1 classes. While there will be many reasons for their literacy difficulties, it is inconceivable that this lowest achieving group would not include large numbers of dyslexic children. Dyslexia affects between 6 and 10% of the population, according to British Dyslexia Association information—and 20% or more are "at risk"' of dyslexia (if a very wide-ranging definition is used) according to a recent No to Failure report. Clearly, any population of very poor readers, such as those who receive Reading Recovery, must by definition include a very much higher proportion of dyslexic learners than 4-20%.

  2.4  Data from three years of Every Child a Reader, and from many more years of Reading Recovery UK annual monitoring reports, consistently shows that eight out of 10 children receiving Reading Recovery are returned to average literacy levels for their age after around 40 hours of one to one teaching, with the remaining two out of 10 ("referred" children) making twice the normal rate of progress (nine months progress in reading age over four to five months teaching). Given that a high proportion of those taught are bound to be dyslexic, the conclusion can only be that Reading Recovery works, and works well, for dyslexic children.

  2.5  The children who receive Reading Recovery, moreover, show a good fit to what we know about dyslexia in the child population. 66% are boys. Many are reported by their teachers to have severe problems with phonological awareness, and many have diagnosed speech and language difficulties. Many are reported to come from families with a history of dyslexia.

  2.6  The following case studies, written by their teachers, describe typical children in Reading Recovery.


    James is a quiet, hard-working little boy who always does his best, but he just couldn't seem to make any progress with his reading and writing in his Reception year and this continued into Year 1.

    His older sister has a diagnosis from an Educational Psychologist to say that she is dyslexic and has struggled all the way through primary school, making minimal progress. James showed all the signs of going the same way.

    James was born with a cleft lip that was corrected when he was a baby, but he is still receiving speech therapy and finds it difficult to make some sounds as well as having immature language.

    James began Reading Recovery in March and at first made slow progress. His mum was very supportive, coming to observe a lesson and doing his homework with him every night. One day while reading, James made an amazing discovery, that "my" is the same as that one on the other page!" We then looked for "my" on every page and in lots of books. He was thrilled to find it was the same everywhere. From that point James started to move forward with confidence and has continued to make good progress. He is now reading simple stories and writing all the time. His class teachers see him as a different child in the classroom, not just in literacy lessons but also in his whole attitude to learning.

    A few days ago James had a speech therapy review and his therapist was surprised at his progress. He had made good progress with sound production but his use of language had risen from the bottom 3% to average levels. James is a much more talkative and confident little boy, bouncing into his lesson each day already knowing what he wants to write about.

    It may well be that James is dyslexic like his sister but we don't think it's going to stop him learning to read and write, thanks to Every Child A Reader and Reading Recovery.


    Vincent scored very inadequately in initial Reading Recovery evaluations. He could only read and write his name and the word "I" after nearly two years at school. Vincent lives with his mother who works (visits his father) and attends an after school club, thereafter is tired when he eventually gets home. His mother did not support his learning during his reception year saying "he should just be playing".

    Vincent has a very poor memory and general knowledge, eg did not know what a calf, foal, lamb, duckling or chick was. He also has huge listening and focusing problems. At the beginning of term his mother told me that his father is dyslexic and had not learnt to read at school. His mother and father are now supporting him consistently. His mother has been to his Reading Recovery lesson in order to be able to support him further.

    Initially Vincent had to learn to focus when reading and writing. He then succeeded in learning all his sounds; establishing left to right directionality, knowing when he had read something incorrectly ("that doesn't make sense"') and attempting to self correct. He still had to say the sounds in even a simple word like "i-t" in order to read the word correctly, which made it very slow for him. He could be reading about for example a "hare" and for an entire book for weeks and then change it to "rabbit" on the final page! These factors all impacted on the speed of his progress.

    After 20 weeks of daily Reading Recovery lessons, however, Vincent has succeeded in catching up with his classmates. His mother is delighted that he has learned to read successfully. For the first time his class teacher says that he has become enthusiastic about his school work.

  2.7  Observation of Reading Recovery lessons by the Every Child a Chance Trust Director (previously a principal educational psychologist, with many years of experience in assessing children for dyslexia, including providing assessments for the then Dyslexia Institute) show that children exhibit familiar dyslexic patterns of persisting b/d confusion, holistic rather than left-right processing (for example, was-saw confusions), confusion of similar sounds such as e/i, omission of word endings such as -ed, and weak short term memory.

  2.8  It is important to note that the children who receive Reading Recovery have a complex profile. Many have very poor home support for learning. Some have poor attendance. Some have social, emotional and behavioural difficulties resulting from trauma or poor parenting. Some have ADHD, Down's syndrome, a history of hearing loss, or general "moderate learning difficulties" (MLD). Many are EAL learners. In some cases the cause of their literacy problems may be one or more of these factors, not dyslexia. It is our view, however, that in many, case these additional factors interact with the phonological awareness problems that characterise dyslexia. The children are both dyslexic but also at the same time challenged by a range of additional factors that impede their learning.

  2.9  Evidence on the impact of other less intensive "layered" interventions used in Every Child a Reader is not as comprehensive as that for Reading Recovery, but still substantial. The original version of the Primary Strategy's Early Literacy Support has been researched by Peter Hatcher at the University of York, as has Reading Intervention and these interventions have been shown to be effective for children who have made a start in reading and are not in the very hardest-to-teach group. Evidence of impact of the Better Reading Partnership scheme, the Fischer Family Trust Wave 3 intervention and Catch Up Literacy has been reviewed by Greg Brooks in his very recently updated research review, "What works for pupils with literacy difficulties?" All were classed as successful for the groups at which they are targeted. Again, it is highly unlikely that the significant numbers of children supported by these schemes did not include many who are dyslexic.


  3.1  Greg Brooks was unable in his exhaustive review to find any evidence for or against the impact of specialist dyslexia teaching (that is, teaching provided one to one and in very small groups by a teacher with a specialist dyslexia qualification). Little formal evaluation appears to have been done in this country. One evaluation of a scheme introduced by the Dyslexia Institute ("Spell It") did not show a significant impact for the school-based intervention, though there was evidence for the impact of a parent-provided teaching programme.

  3.2  A recent report by Chris Singleton, published by the No to Failure charity, presents a different view. We do not, however, consider that this report met its remit in providing Ministers with a review of evidence on interventions for dyslexia. Such a review might be expected to evaluate:

    — evidence on the impact on children's progress of intervention from teachers with specialist dyslexia training—but only four studies in the review involve such teachers; and

    — evidence on the impact of the interventions widely used by specialist dyslexia organisations: Gillingham-Stillman, Alpha to Omega, the Hickey language training course, Units of Sound, the Dyslexia Institute Literacy Programme(DILP). Only one study was presented, the Spell-It study referred to above.

  3.3  Evidence from studies involving specialist dyslexia-trained teachers or methods used under the aegis of dyslexia organisations, or involving children identified as dyslexic or (through screening) at risk of dyslexia are summarised in the table below.[19] The outcomes of these studies can be compared with those for Reading Recovery used with the very lowest attaining six year old children (irrespective of any presumed dyslexia)—Ratio Gain 4.2 (BAS Word Reading).
StudyRatio gain reading accuracy Ratio gain reading comprehension Other or not stated whether accuracy or comprehension
Thomson 1989, Hornsby and Farrer 1990 "Make progress in the order of 18 to 24 months per year" (RG 1.5 to 2)
Hornsby and Miles 1980 RG 1.9 for reading and spelling
Thomson 20031.53 and 2.0 1.33Spelling 1.63
Rack and Walker 1994 Reading 1.0
Rack and Hatcher 2002 Spell It study Reading 1.2
Study of 12 dyslexic children in an independent specialist school—Phonographix 4.5
Lore (2010)—Phonographix 4.1
MTSR—a methodology specifically devised for dyslexic children—reviewed by Greg Brooks Reading RGs ranging from 1.6 to 4.5
Spelling RGs ranging from -3.6 to 13.4
Whiteley (2007) children identified as at risk of dyslexia using DEST screening 40% "benefitted" from a group systematic phonological training (25 hours, 15 weeks) scheme, but 60% made no progress or declined further. After further 15 weeks of one to one teaching the percentage benefitting rose to 66%

  3.4  The argument is made in Singleton's review is that as dyslexic children, without intervention, make very much less than normal progress, ratio gains of the order reported represent good impact. However the research literature overwhelmingly shows that this very much slower rate of progress applies to all poor readers. Brooks reviews a large number of studies with such poor readers that show ratio gains of four and above.

  3.5  It would also be expected that Singleton's review would evidence on the progress made by children identified as being dyslexic, but very few of the studies review meet this criterion as the author reports a dearth of studies of this nature. Instead the review is about what works for children with reading difficulties, and thus repeats that carried out by Greg Brooks for the DCSF in 2007. The only difference is that the Singleton review includes of non-UK studies, but at the same time is very much less comprehensive as it draws from Brooks' review only those studies which used phonologically based interventions, omitting those of a varied and different nature which produced effect sizes or ratio gains just as high (for example, Paired Reading (RG 4.6 comprehension), Catch Up literacy (RG 4.6 comprehension), Better Reading Partnership (RG 7.2 accuracy 9.2 comprehension), Dennis Lawrence's self-esteem based approaches, Inference Training (RG 4.3 accuracy, 17.4 comprehension), Early Literacy Support (effect size equal to a phonologically based comparison programme).

  3.6  The rationale for the omission of studies that were not phonologically based is stated to be that the fundamental problem experienced by dyslexic children is acquisition of phonological decoding skills. This seems as logical as saying that because the fundamental problem with reading experienced by visually impaired children is poor vision, only interventions based on visual methods will be useful. The task of any scientific research in reading is to identify what methods work for children with literacy difficulties, not to pre-suppose what those methods will be.

  3.7  Given the current lack of any robust evidence on specialist dyslexia teaching, the view we expressed to the Rose dyslexia review was that it would be inappropriate and unethical at the present time to offer children such teaching instead of Reading Recovery, or indeed instead of other interventions reviewed by Greg Brooks for which there is already an evidence base.


  4.1  To us, it seems that Reading Recovery embodies many of the features that dyslexia experts would want to see offered to dyslexic learners:

    — use of a highly trained specialist teacher;

    — multisensory learning—as an example, tracing letters in a sand tray, teaching reading through writing and vice versa;

    — techniques to help children analyse the phonemes in a word;

    — rigorous teaching of phonics;

    — cumulative teaching, with teaching points from one lesson or one part of a lesson revisited repeatedly;

    — overlearning—new learning taken to fluency;

    — teaching children syllabification/chunking and

    — an emphasis on metacognition—making the child aware of strategies they are using successfully, so that they can generalise them to new situations and texts.

  4.2  What is different in Reading Recovery, however, is that whilst phonics is taught rigorously and cumulatively it is not taught in a set sequence. The grapheme-phoneme links that are initially taught are chosen to be of high salience for the child (for example, occurring within the child's name, or in book with which they are familiar and enjoy reading). Another difference may be the skill of Reading Recovery teachers in finding ways around a child's profound phonological awareness difficulties, and building from strengths to weaknesses (from the known to the unknown) rather than adopting a deficit model.


  5.1  We emphasised to the Rose dyslexia review that Every Child a Reader is not the only solution to dyslexia. The programme:

    — largely applies to Year 1 and 2 children;

    — only applies to children who cannot read in this age group; this would exclude many dyslexic children who do learn to read reasonably well but whose dyslexia manifests itself in other ways, notably spelling;

    — leaves 15% of those who have received Reading Recovery who have made an average reading age gain of nine months in four to five months' teaching, but not caught up with their peers and so will require further help; and

    — does not address the mathematical difficulties that often characterise dyslexia, or the need for ongoing help with spelling, organisational and study skills.

  5.2  Dyslexia, moreover, does not go away when children become competent readers and writers. It represents a lifelong difference in the way that individuals think and learn, and as such demands appropriate curriculum access arrangements throughout their education.

  5.3  We take the view that dyslexia is a learning difference that only becomes a disability if we fail to adapt the way we teach and the way we organise our classrooms to accommodate the learning styles of children with diverse needs. We have seen at first hand the misery experienced by children whose teachers have not recognised their dyslexia or not responded appropriately to their needs, and agree wholeheartedly with the dyslexia organisations that these issues need to be tackled. Schools need to have effective approaches to dyslexia that include:

    — awareness level training for all class and subject teachers and teaching assistants (such as that universally available to schools through the National Strategies Inclusion Development Programme, which was recently developed in partnership with leading independent dyslexia organisations); and

    — access to advice from specialist knowledge of dyslexia who can help class and subject teachers differentiate their teaching and create dyslexia friendly classrooms.

  5.4  Schools also, however, need to make sure that they tackle the core difficulty for the majority of dyslexic children—difficulty in learning to read—and that they tackle this early on in a child's career before the effects of anxiety and frustration set in and remediation becomes more difficult. Our view is that Every Child a Reader provides schools with a well evidenced means of achieving this goal.

  5.5  Importantly, it offers a route which avoids a policy that provides something at age six to one group of children that is different from that provided to another group. In the past, some schools and local authorities used to provide a two-tier service for children with literacy difficulties—one for sheep (dyslexics) and one for goats (what psychologists have called "garden variety" poor readers). Authorities which ran these services (a dyslexia teaching service and a general learning difficulties service) put vast amounts of time and money into deciding which child merited which service—until research was published which showed there was no benefit in separating sheep and goats as the same teaching approaches worked for garden variety and dyslexic poor readers, all of whom appeared to share the same basic deficit in phonological processing.

  5.5  For this reason, we asked the Rose dyslexia review to consider whether it wanted to recommend:

    — a two tier system that would inevitably be bureaucratic and expensive to administer, or

    — a universal effective provision at age six for all poor readers, that has shown it can get 86% of them to Level 2 or above in Reading at the end of Key Stage 1, compared to the national figure of 84% for all children.

  5.6  Government currently seems to have chosen the first option. We believe this decision was based on good evidence. Government has also made provision for training specialist dyslexia teachers shared between schools, to provide advice on how to make the curriculum accessible for dyslexic children and to work directly with children who fall outside the remit of Every Child a Reader. It will be useful to see research commissioned to evaluate the impact of this work

October 2009

19   A ratio gain represents the ratio of gain in months in reading age to months of intervention. Greg Brooks categorises ratio gains as follows:

RG of 4 or above-remarkable impact

RG of between 3 and 4: substantial impact

RG between 2 and 3: useful impact

RG of between 1.4 and 2: modest impact

RG of 1.0: exactly standard progress-one month's progress for every month the child grows older. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2009
Prepared 18 December 2009