Free places research: why EPPSE takes a different view

Be the first to comment

Professor Kathy Slyva, one of the authors of the EPPSE study, explains why new research into early education places and children's outcomes contradicts their findings.


Oxford University professor Kathy Slyva, author of the EPPSE study

The Government has three stated aims for early education and care: supporting child development, reducing inequalities and enabling parents to enter employment.   The second is particularly important in the United Kingdom where the gaps between educational outcomes for rich and poor are among the largest in the developed world. Several decades of international research has shown that early childhood education reduces the gaps in attainment between rich and poor children – but only if it is of high-quality

Findings from the  Effective Pre-school Primary and Secondary Education project

The Effective Pre-school, Primary and Secondary Education (EPPSE) project followed 3,000 children from early childhood to age 16 and showed that their GCSE results and social development were influenced by what happened to them as young children.  The report on EPPSE outcomes at age 16+ was published in September and showed that early learning at pre-school made a positive difference to the lives of young people at the end of compulsory education.

Two recent papers question the findings of EPPSE, both based on more recent data on children’s development.  There are two different questions: (1) What is the evidence that current policy has improved child outcomes? And (2) What is the evidence that early childhood policy could improve outcomes? The research evidence from the UK and also internationally is quite clear: high-quality early education (not childcare) can improve child outcomes but it may not do so if it is of low-quality.

Key findings of EPPE/EPPSE include:

  • Attending a pre-school had a positive influence on GCSE results.
  • Based on GCSE scores, economists at the Institute of Fiscal Studies were able to predict the monetary benefits to individuals and benefits to the Exchequer. Pre-school attendance was associated with a (predicted) benefit of around £27,000  in lifetime earnings for an individual.
  • Higher quality pre-school improved  total CSE scores, English and maths grades, and social behavioural outcomes.  Children who had attended higher quality pre-school settings also showed more self-regulation and lower levels of hyperactivity at age 16. The quality of pre-school was especially important for children whose parents had low qualifications.
  • Attending pre-school, particularly for a longer period, or attending settings of higher quality, predicted a greater likelihood of following an academic pathway (4 or more A/AS levels) post 16.

Researchers in the EPPSE study compared the outcomes of a large sample recruited before early education was offered free of charge via the (then) Nursery Education Grant. Roughly half of the children in the EPPSE sample attended maintained provision: nursery schools, nursery classes, and nursery schools offering combined care and education. The other half attended provision in the Private, Voluntary and Independent (PVI) sector.

The findings demonstrated the positive benefits of pre-school education for all children, but especially for children whose parents had low levels of education. More importantly, children made more progress in the maintained sector, which Ofsted also reports to be of higher quality. The over-sampling of the maintained sector in the EPPSE design allowed the effect of different levels and types of quality to be studied.

Current statistics show that the majority of children under the age of three are cared for in the PVI sector. Moreover the great expansion in this century of places for 3-5 year old children was in the PVI sector;  places in the maintained sector remained constant.  There is no doubt that the maintained sector has higher quality (see Ofsted 2014 and also Mathers et al., 2007), and higher quality has been shown to lead to better outcomes.

What the EPPSE study has shown is that early childhood education can improve the life chances of poor children, and thereby reduce inequality. However, early education of low quality will not reduce inequalities. The EPPSE study was designed to answer the question ‘Can early education improve children’s life chances?’ This is the reason it over-sampled higher quality settings; a large number of children in high quality settings were needed to show the effects of a wide range of quality.

The recently published studies reported the outcomes of the children across England, many of whom are in the lower quality PVI sector (Blanden et al 2014 and Brewer et al 2014). The results from the National Pupil Database (NPD) sample (children born after the EPPSE children) show only weak effects of pre-school education on children’s development at age seven, diminishing by age 11.  There are three reasons for this-many children in the sample were in settings shown to be of lower quality than the EPPSE study (PVI), EPPSE controlled for social background through detailed measures collected via interviews with families, and the outcome used in the analyses were the Key Stage 1 broad-level assessments. These are graded on a simple ‘level’ scale, which is very coarse and places the vast majority of children at level 2 (‘as expected’).

By way of contrast, EPPSE used psychometrically sound tests at school entry which were individually administered and scored in a highly differentiated manner. For children at ages seven and 11, the EPPSE team had access to more finely differentiated scores from nationally administered tests, rather than the broad-banded levels.  This may explain the fact that EPPSE is ‘optimistic’ about the power of early education to reduce inequalities and the NPD study is more pessimistic. The former studied children from a wide range of settings, many of which were high quality. The NPD sample (Blanden et al, 2014) included many more children in the PVI sector, used coarse ‘levels’ as outcome, and had no measures of quality of provision.                 

There is one other excellent study looking directly at the effects of quality on children’s outcomes. The ‘Two Year Old Pilot’ study (Smith et al., 2009) was funded by the Department for Education and carried out by NatCen and Oxford University. It studied the effects of attendance at two-year-old provision for children in disadvantaged groups. This study found that the two-year-old provision did not improve children’s outcomes if of low quality, but that it improved vocabulary scores on a psychometrically sound test at age three if it was of moderate-to-excellent quality. A follow-up of the same children in primary schools showed that the gains found at age three continued on the Early Years Foundation Stage Profile (Maisey et al. 2013). This more recent study confirmed the earlier findings from EPPSE; quality matters, especially for the poor, and it is necessary if the gap is to be narrowed.

The recent Ofsted report on the early years (Ofsted, 2014) called for improvements in the quality of provision for children living in  disadvantaged neighbourhoods, where quality in the PVI sector was lower  compared to quality of  settings in more advantaged areas. One way to narrow the gap is to improve childcare funding for disadvantaged children, as in the Early Years Pupil Premium.  


Blanden, J., Del Bono, E., Hansen, K., McNally, S., & Rabe, B. (2014). Evaluating a demand-side approach to expanding free preschool education. Retrieved from

Brewer, M., Cattan, S., Crawford, C., Rabe, B., Blanden, J., Del Bono, E., Hansen, K & McNally, S. (2014). Free childcare for 3 year olds: no long term benefits for child development. Institute for Social and Economic Research.

Maisey, R., Speight, S. & Marsh, V. with Philo, D. (2013). The Early Education Pilot for Two Year Old Children: Age Five Follow-up. DfE Research Report RR225.

Mathers, S., Sylva, K. & Joshi, H. (2007). Quality of Childcare Settings in the Millennium Cohort Study. DfES Research Report SSU/2007/FR/025.

Ofsted. (20014). The Annual Report of Her Majesty's Chief Inspector of Education, Children's Services and Skills 2012/13 Early Years. London: Her Majesty’ s Stationary Office.

Smith, Purdon, Mathers, Sylva, Schneider, La Valle, Wollny, Owen, Bryson & Lloyd (2009). Early Education Pilot for Two Year Old Children Evaluation. DCSF Research Report RR134.

Sylva, K. Melhuish, E., Sammons, P., Siraj, I & Taggart, B. (2014). The Effective Pre-school, Primary and Secondary Education Project (EPPSE 3-16) Students’ educational and developmental outcomes at age 16. DfE Research Report RR354.

blog comments powered by Disqus