Reading between the lines: early intervention in practice
Jean Gross reflects on a recent UK evaluation of the Reading Recovery programme, and how this illustrates other trends and challenges for early intervention in the current environment.
Momentum is definitely building again for early intervention, stimulated by EIF’s milestone report Realising the potential of early intervention, and by the House of Commons Science and Technology Select Committee’s report on Evidence-based early years intervention. This week, the publication of new research on the long-term impact of Reading Recovery picks up on important themes from these reports, and can help us think through what they mean for early intervention on the ground.
New research on the impact of Reading Recovery
Reading Recovery is a one-to-one tuition programme for children who have made little or no progress in reading after a year in school. It was at the heart of the former national Every Child a Reader programme, of which I was formerly a director.
Evidence for Reading Recovery’s impact is already strong: the programme featured in Graham Allen’s seminal reports on early intervention, and it is highly rated in the EIF Guidebook. Its international evidence comes from multiple successful randomised controlled trials, including a recent evaluation at scale, in a multisite randomised controlled study involving almost 7,000 participating students in over a thousand US schools.
The newly published UK research compares outcomes of a sample of children who received Reading Recovery and a comparison group over a 10-year period. Broadly, it found that children who received Reading Recovery at age six were more than twice as likely to achieve five or more good GCSEs including English and mathematics (49% vs 23% in the comparison group) and less likely to leave school with no qualifications (2% vs 7% in the comparison group). The effect size for GCSE points scores, at 0.49, was greater than that generally seen immediately after the majority of comparable educational interventions. The children who took part in the programme required no intensive special needs support – a Statement of special educational needs, or an education, health and care plan (ECHP) – whereas 10% of the comparison group had a SEN Statement or EHCP at age 14, and 9% at age 16.
The research is not perfect. One strong feature is its use of the National Pupil Database to look at outcomes. This is a fully independent measure, and one which ensured that attrition was low. On the other hand, the children to be followed up were not randomly allocated to Reading Recovery or a control group; this is tricky when the programme’s guidelines call for careful selection to ensure that very lowest-attaining children take part. Instead, a quasi-experimental design was used, with the comparison children matched on a range of variables. In the event, however, the groups proved to be significantly different on one key variable: eligibility for free school meals. The researchers therefore looked at children eligible for free school meals separately, and confirmed still-substantial effects of Reading Recovery on GCSE results, with an effect size of 0.37.
Building on these findings, a study by Pro Bono Economics calculated the potential benefit of Reading Recovery support to be £9,200–£12,100 for each child, compared to around £2,800 in costs – a net benefit of £6,400–£9,300 per child, with a benefit-to-cost ratio of between 3.3 and 4.3. The large part of the benefit is attributable to a child’s higher lifetime earnings as a result of increased employment and higher wages. But £2,900 per child represents savings to local authorities on their high-needs SEN budgets as a result of the reduction in SEN Statements and EHCPs in the Reading Recovery group.
What does this new research teach us about early intervention more widely?
This new research reflects several key themes from EIF’s ‘Realising the potential’ report: the importance of good evidence that is carefully critiqued to understand its strengths and weaknesses, and the potential for long-term economic returns on investment, coupled with shorter-term cashable savings where – as in this case – they can be realised. It also speaks to the well-made argument that early intervention is not just about what happens in the early years – between the ages of 0–5 – and to the need to break intergenerational cycles of disadvantage. Over half of children of the group who received Reading Recovery were eligible for free school meals, and many came from families with low literacy levels. As one Liverpool boy said to me, after completing his programme: ‘I can read now – I’m the only one in my family who can read.’
Finally, some of the challenges that Reading Recovery is facing nationally illustrate all too clearly the wider barriers to early intervention that EIF has described.
Funding: The EIF report notes that providing effective early intervention services requires long-term investment at a level that is sufficient to enable the commissioning and implementation of high-quality interventions by skilled and experienced professionals. Reading Recovery is delivered by qualified teachers, who have a full year of on-the-job training. Fidelity to the programme is ensured by subsequent quality-assurance of their practice from local ‘teacher leaders’.
This is important, but it is costly. Until 2011, schools received half the cost of implementing Reading Recovery from central government, and at that time the programme was reaching 21,000 children a year. Unsurprisingly, after the match-funding ceased, these numbers fell and by 2016/17 they stood at just 4,500.
Short termism: We know that policymakers are required to develop initiatives to tackle the issues of the day, such as rising exclusions from school and street violence, rather than address the underpinning risks that contribute to such problems. Very low literacy levels are one such key risk: there are robust, persistent correlations between poor literacy skills on leaving primary school and school exclusion, antisocial behaviour, crime, and poor adult mental health.
Fragmented responsibility: EIF notes that the good case for investing in early intervention can be undermined by the fact that its benefits often do not accrue to those who make that investment. The returns from Reading Recovery accrue largely to central government (with a smaller proportion to secondary schools and local government), rather than to the initial investor: the primary school. The decline in numbers of schools taking part after 2011 reflects this basic economic dilemma.
Not delivering what works: In the light of its strong evidence base, one would imagine that schools would choose Reading Recovery over programmes with less robust evidence of impact. In fact, they most often tend to choose group interventions delivered by teaching assistants, where there is as yet no long-term follow-up, and where a typical effect size for the best of these is around 0.2–0.3 immediately after the intervention.
So what is the future for this particular early intervention programme, Reading Recovery? That may depend on whether, post-Brexit or non-Brexit, government responds to the call to grow our economy by investing in early intervention within the next comprehensive spending review. Locally, it may be that the new opportunities for primary–secondary collaboration presented by multi-academy trusts and opportunity areas will encourage school collectives to commission Reading Recovery as an early intervention strategy to improve children’s outcomes and reduce costs in the secondary phase. Personally, I will watch, wait and hope.
Jean Gross CBE is an EIF Associate, a former trustee of EIF, and a former director of the Every Child a Reader programme. She writes here in a personal capacity.