Clackmannanshire revisited

The Clackmannanshire study is often cited as demonstrating the positive impact of synthetic phonics (SP) on children’s reading ability. The study tracked the reading, spelling and comprehension progress, over seven years, of three groups of children initially taught to read using one of three different methods;

  • analytic phonics programme
  • analytic phonics programme supplemented by a phonemic awareness programme
  • synthetic phonics programme.

The programmes were followed for 16 weeks in Primary 1 (P1, 5-6 yrs). Reading ability was assessed before and after the programme and for each year thereafter, spelling ability each year from P1, and comprehension each year from P2. After the first post-test, the two analytic phonics groups followed the SP programme, completing it by the end of P1.

I’ve blogged briefly about this study previously, based on a summary of the research. It’s quite clear that the children in the SP group made significantly more progress in reading and spelling than those in the other two groups.  One of my concerns about the results is that in the summary they are presented at group level, ie as the mean scores of the children in each different condition. There’s no indication of the range of scores within each group.

The range is important because we need to know whether the programme improved reading and spelling for all the children in the group, or for just some of them. Say for example, that the mean reading age of children in the SP group was 12 months ahead of the children in the other groups at the end of P1. We wouldn’t know, without more detail, whether all the children’s scores clustered around the 12 month mark, or whether the group mean had been raised by a few children having very high scores, or had been lowered by a few having very low scores.

At the end of the summary is a graph showing the progress made by ‘underachievers’ ie any children who were more than 2 years behind in their test scores. There were some children in that category at the end of P2; by the end of P7 the proportion had risen to 14%. So clearly there were children who were still struggling despite following an SP programme.

During a recent Twitter conversation, Kathy Rastle, Professor of Psychology at Royal Holloway College London (@Kathy_Rastle), sent me a link to a more detailed report by the Clackmannanshire researchers, Rhona Johnston and Joyce Watson.

more detail

I hoped that the more detailed report would provide more… well, detail. It did, but the ranges of scores within the groups were presented as standard deviations, so the impact of the programmes on individual children still wasn’t clear. That’s important. Obviously, if a reading programme enables a group of children to make significant gains in their reading ability, it’s worth implementing. But we also need to know the impact it has on individual children, because the point of teaching children to read is that each child learns to read.

The detail I was looking for is in Chapter 8 “Underachieving Children”, ie those with scores more than 2 years below the mean for their age. Obviously, in P1 no children could be allocated to that category because they hadn’t been at school long enough. But from P2 onwards, the authors tabulated the numbers of ‘underachievers’. (They note that some children were absent for some of the tests.) I’ve summarised the proportions (for boys and girls together) below:

more than 1 year behind (%)

P2 P3 P4 P5 P6 P7
reading 2.2 2.0 6.0 8.6 15.1 11.9
spelling 1.1 4.0 8.8 12.6 15.7 24.0
comprehension 5.0 18.0 15.5 19.2 29.4 27.6

more than 2 years behind (%)

P2 P3 P4 P5 P6 P7
reading 0 0.8 0 1.6 8.4 5.6
spelling 0.4 0.4 0.4 1.7 3.0 10.1
comprehension 0 1.2 1.6 5.0 16.2 14.0

The researchers point out that the proportion of children with serious problems with reading and spelling is quite low, but that it would be “necessary to collect control data to establish what would be typical levels of underachievement in a non-synthetic phonics programme.” Well, yes.

The SP programme clearly had a significantly positive impact on reading and spelling for most children. However that clearly wasn’t true for all of them. The authors provide a detailed case study for one child (AF) who had a hearing difficulty and poor receptive and expressive language.  They compare his progress with that of the other 15 children in P4 who were one year or more behind their chronological age with reading.

Case study – AF

AF started school a year later than his peers and his class was in the analytic phonics and phonemic awareness group.  They then followed the SP programme at the end of P1.  Early in P2, AF started motor movement and language therapy programmes.

By the middle of P4, AF’s reading and spelling scores were almost the average for the group whose reading was a year or more behind, but his knowledge of letter sounds, phoneme segmentation and nonword reading was better than theirs. A detailed analysis  suggests his reading errors are the result of his lack of familiarity with some words, and that he’s spelling words as they sound to him. Like the other 15 children experiencing difficulties, he needed to revisit more complex phonics rules, so a supplementary phonics programme was provided in P5. When tested afterwards, the mean scores for the group showed spelling and reading above chronological age, and AF’s reading and spelling improved considerably as a result.

During P6 and P7 a peripatetic Support for Learning (SfL) teacher worked with AF on phonics for three 45 minute sessions each week and taught him strategies to improve his comprehension. An cccupational therapist and physiotherapist worked with him on his handwriting, and he was taught to touch type.  By the end of P7, AF’s reading age was 9 months above his chronological age and his spelling was more than 2 years ahead of the mean for the underachieving group.


The ‘Clacks’ study is often cited as conclusive proof of the efficacy of SP programmes. It’s often implied that SP will make a significant difference for the troublesome 17% of school leavers who lack functional literacy.   What intrigued me about the study was the proportion of children in P7 who still had difficulty with functional literacy despite having had SP training. It’s 14%, suspiciously close to the proportion of ‘functionally illiterate’ school leavers.

Some teachers have argued that if all the children had had systematic synthetic phonics teaching from the outset, the ‘Clacks’ figures might be different, but AF’s experience suggests otherwise.  He obviously had substantial initial difficulties with reading, but by the end of primary school had effectively caught up with his peers. But his success wasn’t due only to the initial SP programme. Or even to the supplementary SP programme provided in P5. It was achieved only after intensive, tailored 1-1 interventions on the part of a team of professionals from outside school.

My children’s school, in England, at the time when AF was in P7, was not offering these services to children with AF’s level of difficulty. Most of the children had followed an initial SP programme, but there was no supplementary SP course on offer. The equivalent to the SfL teacher carried out annual assessments and made recommendations. Speech and Language and Occupational therapists didn’t routinely offer treatment to individual children except via schools, and weren’t invited into the one my children attended. And I’ve yet to hear of a physiotherapist working in a mainstream primary in our area.

As a rule of thumb, local authorities will not carry out a statutory assessment of a child until their school can demonstrate that they don’t have the resources to meet the child’s needs.  As a rule of thumb, schools are reluctant to spend money on specialist professionals if there’s a chance that the LA will bear the cost of that in a statutory assessment.  As a consequence, children are often several years ‘behind’ before they even get assessed, and the support they get is often in the form of a number of hours working with a teaching assistant who’s unlikely to be a qualified teacher, let alone a speech and language therapist, occupational therapist or physio.

If governments want to tackle the challenge of functional illiteracy, they need to invest in services that can address the root causes.


Johnston, R & Watson, J (2005), “The Effects of Synthetic Phonics teaching on reading and spelling attainment: A seven year longitudinal study”, The Scottish Executive website


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s