We are testing a new beta website for gov.scot go to new site

The Effects of Synthetic Phonics Teaching on Reading and Spelling Attainment





9.1 At the beginning of Primary 1, one group of children learnt to read using the synthetic phonics programme. They were compared with two groups learning to read by analytic phonics programmes; one of these programmes was a standard analytic phonics programme, but the other one contained intensive training to enable children to hear sounds such as phonemes and rhymes in spoken words. At the end of the 16 week training period, the synthetic phonics group were reading words around 7 months ahead of chronological age, and were 7 months ahead of the other two groups. The synthetic phonics group's spelling was also 7 months ahead of chronological age, and was around 8 to 9 months ahead of the two analytic phonics groups. These groups were spelling 2 to 3 months behind chronological age. The synthetic phonics group also showed a significant advantage in ability to identifying phonemes in spoken words, performing even better than the group that had experienced direct training in this skill, despite the fact that these children were from significantly less advantaged homes than the other children. The phonemic awareness programme was found to have no benefits for literacy acquisition.

9.2 The two analytic phonics taught groups then carried out the synthetic phonics programme, completing it by the end of Primary 1. In the meantime the initial synthetic phonics group consolidated their learning rather than moving on to learn new grapheme to phoneme correspondences. During the course of Primary 2 some children in the original analytic phonics taught groups received extra help, but this was not necessary for the initial synthetic phonics taught group. At the end of Primary 2, the initial synthetic phonics taught children were significantly better spellers, and there was a trend towards better word reading skills. When separate analyses of word reading were carried for boys and girls, it was found that early or late synthetic phonics teaching had no impact on the boys reading attainment. However, the analysis for the girls showed that the early synthetic phonics trained group read words significantly better than the group that had received the standard analytic phonics programme first. We conclude that in order to foster good spelling skills, and to assist girls in learning to read, synthetic phonics should start early in Primary 1.

9.3 We have conducted an analysis of the children's performance from Primary 2 to Primary 7, comparing the same children right through in word reading, spelling and reading comprehension. This was to gain an exact measure of whether the gains the children experienced from the Primary 1 programme were maintained, or whether they increased or decreased. It was found for word reading and spelling that the gain in skill compared with chronological age had increased significantly over the years, even though the training programme had ended in Primary 1. In Primary 2, word reading was found to be 11.5 months ahead of chronological age, but in Primary 7 it was 3 years 6 months ahead. For spelling, in Primary 2 it was 1 year ahead, whereas by Primary 7 it was 1 year 9 months ahead. However, for reading comprehension, a different pattern was shown. In Primary 2 the children were comprehending what they read 7 months ahead of chronological age, but by Primary 7 this had dropped to a 3.5 months advantage.


9.4 We also compared the performance of the boys and the girls. In Primary 2, they were found to read words equally well, and there were also no sex differences in spelling ability and reading comprehension. However, in Primary 3 the boys pulled ahead of the girls in word reading and by Primary 7 were reading 11 months ahead of the girls. The boys also spelt better than the girls in Primaries 4, 6 and 7, and by Primary 7 were 8.6 months ahead. The boys were also 3 months ahead of the girls in reading comprehension in Primary 7, but this was not statistically significant. It is very unusual for boys to perform better than girls; in a recent international study of reading comprehension, girls were significantly ahead of boys in all 35 countries (Mullis et al, 2003).


9.5 The girls, despite not having superior literacy skills, had a significantly more positive attitude to reading than the boys on the ATR2 (Ewing and Johnstone, 1981). When answering a direct question about how much they liked reading, girls were found to like reading significantly more than boys. They were also significantly more likely to be a member of a public library. However, when questioned about how much fiction they read, no difference was found between the boys and the girls. This is an atypical finding, as boys are generally found to read less fiction than girls (Mullins et al 2003). It would be desirable to study controls in Scotland in order to determine whether this is unusual. A positive attitude to reading was associated with better word reading and spelling skills, more reading of fiction, and greater use of the public library by the children (and their parents). It was also associated with being able to read and write letters before starting school. However, as these analyses are correlational, one should be cautious about assuming that the findings indicate causation.


9.6 In response to a request from SEED for final feedback at the end of the seven-year research period, a brief questionnaire was sent to the eight Head Teachers of the schools included in the study. All the Head Teachers responded that, in their view, reading, spelling and writing skills had been accelerated by the synthetic phonics programme. One Primary 2 teacher, with thirty years' experience, also responded, observing that not only had the literacy skills been accelerated but also the results were "the best ever achieved" and would not normally have been expected until the Primary 3 stage. It was also stated that the children were very motivated, enjoyed the programme and had improved confidence in their literacy skills. Indeed, one Head Teacher commented that the synthetic phonics programme had empowered both teachers and pupils and had also provided both staff and curricular development opportunities. Another Head Teacher said that it was a professional 'life-changing' experience. All of the respondents agreed that teachers now had higher expectations of their pupils, one Head Teacher remarking that the accelerated pace of teaching and learning had become the norm. Another Head Teacher of a school in an area of deprivation said that they now knew what the children could achieve and that it was possible to help less able pupils to keep pace with the class. In terms of detecting children needing learning support, most Head Teachers commented that they were able to do this much earlier and one said that for some children only a low level of support time was needed for them to catch up.


9.7 An examination was made of the effects of differing socio-economic background, using Clackmannanshire Council's deprivation index. Using this index we divided our sample into advantaged and disadvantaged, according to the Council's categorisation. The expectation was that children from advantaged homes would outperform those from disadvantaged homes. However, for word reading and spelling this was only found to be significant in Primary 7 (only marginally so for reading), where the advantaged children's reading was 6.2 months ahead of that of the disadvantaged children's reading and spelling was 5.8 months ahead. For reading comprehension, the advantaged children were significantly ahead only in Primaries 5 and 7, the superiority at the end of the study being 5.5 months. It is very likely that children learning by the standard analytic phonics approach would show these socio-economic differences much earlier on in their schooling, but further work with a control sample will be needed to examine this issue.

9.8 A questionnaire was sent to the parents in Primary 7, and we achieved a 46.4% response rate. These data were used in correlational analyses, together with Clackmannanshire Council's Deprivation Index. From these analyses we have found that the less deprived the homes the children came from, the better they read and spelt in Primary 7. The parents from less deprived homes reported having more children's and adults' books, and the adults said they made greater use of public libraries. This greater availability of books may explain why socio-economic differences emerge by Primary 7, where home influences may become more important as children spend more time reading independently outside the school curriculum. The children from the less deprived homes were more likely to have attended a mother and toddler group, but virtually all of the children had attended a nursery class. There was no correlation between the deprivation index and the extent to which parents valued learning to read. This equal value placed on education by the less well off may not be found in other parts of the UK, and it would be interesting to establish whether this is so. Having more adults' books in the home was associated with both mothers and fathers having high educational levels, whereas the number of children's books in the home was associated only with the mother's educational level. Interestingly, the more educated the father the more likely the children were to read and write letters of the alphabet before starting school.


9.9 Although the synthetic phonics programme has clearly had a major effect on the literacy skills of these children, it is important to know whether it is just the high achieving pupils who have received a boost, or whether there have also been gains for the lower achieving children. We have no controls for comparison, but we can examine the proportions of low achievers, and the progress of one such child has been studied in detail. We have taken a performance level of more than two years below chronological age as indicating underachievement. This measure cannot be meaningfully made for the children in Primary 2, but is useful from Primary 3 onwards. In Primary 3, only 0.8% of the children were more than 2 years behind in word reading, 0.4% in spelling, and 1.2% in reading comprehension. By Primary 7, 5.6% were more than 2 years behind in word reading, 10.1% behind in spelling, and 14.0% were behind in reading comprehension.

9.10 The question arises as to whether these low achieving children can be helped to attain normal performance levels. Although a revisiting programme was devised and offered to learning support teachers, we only know of a few children who definitely received the programme. One in particular, AF, has had his progress closely monitored over the last 7 years. AF entered school a year late, primarily due to difficulties in language development. He was entered into our analytic phonics and phonemic awareness training programme, having at the start of schooling no measurable level of phonemic awareness, and being unable to give a single rhyme for a spoken word. His only indication of any literacy skills was being able to give the name of one letter of the alphabet. At the end of the analytic phonics and phonemic awareness training programme he was still a non-reader, had no phonemic awareness or rhyme ability, could only give 3.8% of letter names, and knew no letter sounds. He and his class then carried out the synthetic phonics programme, completing it by the end of Primary 1. At the end of Primary 2 we found that he now had a reading age of 5.6 years, but his spelling age was 5.0 years which indicates that he was not able to spell at this stage. In Primary 3, his reading was 6.1 years and he did not sit the spelling test. By January of Primary 4, his reading age was 6.8 years, his spelling was 7.0 years, and his phonemic awareness and nonword reading scores were 100% correct. At this point, he carried out the Phonics Revisited programme. By the end of Primary 5, his word reading was 9.2 years, his spelling age was 8.9 years, and his reading comprehension was 8.0. This performance was quite creditable given that the actual age of his class was 9.7 years, but as he had entered school a year late, his chronological age was 10.6 years. In Primary 5, we tested his receptive vocabulary knowledge, and he gained a score of only 75, where the average is 100. By the end of Primary 6 his reading age was 10.2 years, and his spelling age was 10.1 years, but this still meant he was lagging behind his chronological age of 11.4 years. At this point he carried out a programme which developed advanced blending skills and a more visual approach to spelling. At the end of Primary 7, when he was 12.4 years old, his reading age was 13.1 years and his spelling was 10.5 years. His reading comprehension, however, had fallen back to 7.1 years. Juel (1988) has argued that children who make a slow start always lag behind. However, it is clear that a child whose low achievement that has its basis in severe language development difficulties can achieve a very creditable level of literacy skills with appropriate teaching methods and learning support tailored to his needs.


9.11 It is evident that the children in this study have achieved well above what would be expected for their chronological age according to standardised tests. The actual gains may be much larger than this comparison indicates, as many of the children came from homes experiencing economic deprivation, and receptive vocabulary knowledge scores for the whole sample were somewhat below average. It is hoped that in future work controls matched on socioeconomic background can be studied, so that we can gauge the true gain.

9.12 Overall, we can conclude that a synthetic phonics programme, as a part of the reading curriculum, has a major and long lasting effect on children's reading and spelling attainment. Indeed, these skills were found to be increasing many years after the end of the programme. It is evident that the children have learnt a technique that they can use for themselves, that they have learnt a self teaching technique. Furthermore, although in a recent international study boys were found to have significantly lower levels of reading comprehension than girls in all 35 countries surveyed, the boys in this study comprehended text as well as the girls'. In fact they were slightly ahead, and if this trend continues in the future, it may become statistically significant. Socio-economic differences in literacy skills were non -existent in the early years of the study, only emerging in the upper primary years. Further work will be needed, however, to establish just how great the gains are in comparison with other approaches to teaching reading.