Select Committee on Education and Employment Appendices to the Minutes of Evidence


Memorandum from Sula Wolff FRCP, FRCPsych, DCH (EY 14)


1.  The National Commission on Education (1993), many of whose recommendations have been enacted by successive governments, reported that the percentage of British school leavers who go on to higher education and obtain a degree or higher level diploma compares well with that in other countries. But a large majority of children fail to benefit from their education despite adequate abilities. The percentages of children who at 16 and 18 achieve comparable school qualifications were at that time much lower in the UK than in Germany, France and Japan.

  2.  An academic self-image is formed between 5 and 8 years, and is both an outcome and a mediator of schooling. A positive academic self-concept enhances learning, and the correlation between perceived competence and school achievement increases steadily from 9-12 years (Entwisle et al, 1987). Sylva and Wiltshire (1993) have convincingly argued that the most important impact of early education is on children's aspirations, motivations and commitment to school, rather than on the effects of what they are specifically taught. They argue further that a "mastery" orientation to learning, based on confidence that they are able to succeed, leads children to learn for the sake of learning rather than for adult approval, and enables children to persist in the face of difficulty. In contrast, children with a "helpless" orientation to learning (which is not, or not necessarily, related to IQ), saw difficulties not as challenges but as indicative of low ability, viewed educational tasks as tests, were convinced they would fail and tended to give up in the face of difficulty.

  3.  The enduring positive outcomes of good pre-school education, such as the High Scope Perry pre-school programme, have been well and repeatedly documented. Sylva (1994) makes the point that its main effect was to enable children to develop a commitment to schooling, which begins as a response to school success and is both motivating and promoting of cognitive progress. Participation in this pre-school programme was cost effective and strongly related to later school achievements and also to reduced rates of truancy, school drop out, delinquency, dependency on welfare and even, in girls, teenage pregnancy (Barnett and Escobar, 1990). Such pre-school programmes have the greatest effect on children from disadvantaged backgrounds. For such programmes to be effective, their curriculum needs to be based on play and active learning rather than on formal instruction, but on play that is guided and extended by adults rather than free play alone (Sylva and Wiltshire, 1993). We need to be clear that these effective pre-school programmes took place in the USA where the statutory age of school entry is six years. Although not based on strictly controlled comparisons, there is evidence that pre-school education in nursery schools in the UK is more effective than education in nursery classes within primary schools (Osborne and Milbank, 1987).

  4.  It is possible, but not yet proven, that if high quality pre-school programmes could be guaranteed for all socially disadvantaged children, it might not be necessary to consider changing the age of entry to formal schooling. Yet it is in my view quite unrealistic to rely on such an eventuality.

  5.  The development of language related skills, including reading and spelling, in boys lags behind that of girls, at least in the early years. Boys on average arrive at school less mature than girls and are more likely to start their school careers with a sense of failure.

  6.  I have recently argued that the start of formal schooling should be delayed, as in most other countries, until children are six years old (Wolff, 1999).

  7.  Britain is out of step with most European countries in its statutory age of school entry. Six years is the usual age of entry, except for the Scandinavian countries where it is seven. The age of five for school entry was set in Britain in the last century for reasons that no longer apply: to remove children from the unhealthy conditions and exploitation within their neighbourhoods and homes, and to enable them to complete their education in time for an early entry into the labour market (Woodhead, 1986).

  8.  Research into the effects of age of school entry is of two kinds. Within the UK and the USA, school attainments have been measured for children born in different seasons of the year and who thus start school at the beginning, middle or end of their statutory age of school entry (at the age of six in the USA). The problems with these studies is that the age span of school entry investigated is quite short; season of birth may have unknown, independent effects; the sex ratios of the different intakes have not always been controlled for (Tymms, 1998); and that, if testing is done only one or two years after school entry, the results are confounded by length of schooling.

  9.  Sharp et al (1994), however, in their review of these studies (not all of which found early school entry to be disadvantageous) and a report of their own findings, conclude that overall children who start school close to their fourth birthday do not do as well as children born at the same time of the year who start school later. They stress that teachers may underestimate the abilities of the youngest children in their class because of these children's poorer co-ordination and attention span, and that their expectations and style of teaching may not be appropriate for the children's development level.

  10.  More compelling are international comparisons of children's educational progress in relation to age of school entry. The International Educational Achievement Study of reading literacy and instruction in 32 school systems makes fascinating reading (Elley, 1992). Sadly the countries surveyed did not include the UK, because the National Foundation for Educational Research did not approve of the tests used and was, one deducts, not able to influence these (Brooks et al, 1996).

  11.  In this international study numerous background variables were considered in relation to the children's tested attainments. These variables included measures of general health and welfare, adult literacy rates, gross per capita national product, public expenditure on education, language of tuition compared with language spoken, complexity of language, teacher training, childrens' access to reading materials and a number of others including the child's gender and the statutory age of school entry in the country concerned. Testing was done at nine and 14 years.

  12.  Of the 32 countries surveyed, only three began formal schooling, as in the UK, at five (New Zealand, Ireland and Trinidad and Tobago); one began at five and a half; eight at seven; and 20 at six years of age. In most developed and developing countries children start school at 6. Only in New Zealand, Ireland, Trinidad and Tobago were children formally instructed in reading at five.

  13.  The authors found that a late start to formal instruction was no handicap. Those who started at seven were often the best at nine, despite having had a shorter time at school. Across all countries, other things being equal, starting school at six was best of all.

  14.  In all countries girls outstripped boys in overall levels of achievement, especially in narrative reading at nine. When length of schooling before testing and all other relevant factors were controlled for, the gender differences in overall achievement were greatest in children who had started school at five, both in tests administered at nine and even more so in those given at 14. The authors write: "A plausible interpretation of these trends is that a policy of an early start in formal reading instruction is too early for many boys, and implies persisting problems for many of them" (pp 107-108). They also say " countries where formal instruction begins at age five, the boys fell further behind the girls in both comprehension and word recognition. An early start to formal instruction may not be beneficial for many late maturing boys" (pp 226).

  15.  Among many other important findings of this study was that reading was best promoted when the stress was put on reading for enjoyment rather than on reading skills.

  16.  All child mental health professionals are well aware of the often devastating effects on children of school failure. Not only are they deprived of the opportunities of a rewarding education and working life; school failure contributes to the development of anti-social conduct and delinquency. Both curtailed education and delinquency are commoner in boys than girls, and both have far reaching consequences for the wider community. (It was different even 20-30 years ago when boys with no educational qualifications could still find manual jobs they were good at, which gave them an entry into the adult world, raised their self-esteem and often helped them to give up previous delinquent behaviour and delinquent associates).

  17.  Mortimer and Whitty (1997) have recently argued that in order to overcome the effects of disadvantage on children, progress on two fronts is needed: school improvement as well as tackling the structural inequalities in society. Yet it is unrealistic to expect that high quality pre-school education will, at least in the short term, be available for all children from disadvantaged backgrounds. Nor is it realistic to rely on any rapid decline in the numbers of children growing up amidst poverty and other socially disadvantageous circumstances despite active political measures to counteract social exclusion and child poverty.

  18.  And there is another important point to be made, one that it is currently unfashionable to mention because it is out of step with "politically correct" ideas. Not all children are equally able. There are real differences in intelligence between children, often of course compounded by their sociocultural environment. Much emphasis is, rightly, placed on raising teacher expectations of disadvantaged children. It has been shown, for example, that even children of good abilities and with adequate verbal skills, as demonstrated in the home environment, often arrive at school lacking in social confidence and social skills, so that their verbal and other abilities tend to be underestimated by their teachers (Tizard and Hughes, 1984). And when teacher expectations are too low, children's school progress is impeded. Of course this needs to be corrected.

  19.  Yet, when teacher expectations are too high and children are for reasons of intellectual immaturity unable to fulfil these, they start their school lives with a sense of failure, likely to affect the whole of their subsequent school careers. The choice is between setting the start of formal learning at an age level at which almost all children can cope or, as at present, at a level at which a large minority of children are, for reasons of intellectual immaturity, often compounded by co-existing social disadvantage, unable to follow the school curriculum. Moreover, the International Study of Reading Literacy (Elley, 1992) suggests that children of above average abilities are not handicapped by a later start of formal learning.

  20.  For all these reasons, and despite the many logistic difficulties that will ensue, I urge the Committee to consider very seriously raising the statutory age of school entry to the age of six years.


  Barnett, W S and Escobar, C M (1990) Economic costs and benefits of early Intervention. In: S J Meisels and J P Shonkoff (eds) Handbook of Early Childhood Intervention. Cambridge: Cambridge University Press, pp 560-583.

  Brooks, G, Pugh, A K, Schagen, I (1996) Reading Performance at Nine. Slough, Berkshire: National Foundation for Educational Research and the Open University.

  Elley, W B (ed) (1992) The IEA Study of Reading Literacy: Achievement and Instruction in Thirty-Two School Systems. International Studies in Educational Achievement. Oxford: Pergamon.

  Entwisle, D R, Alexander, K L, Pallas, A M, and Cadigan, D (1989) The Emergent Academic self-image of first graders: its response to social structure. Child Development, 58, 1190-1206.

  Mortimer, P and Whitty, G. (1997) Can School Improvement Overcome the Effects of Disadvantage? Institute of Education, University of London.

  National Commission on Education (1993) Learning to Succeed, Report of the Paul Hamlyn Foundation. London: Heinemann.

  Osborn, A F and Milbank, J E (1987) The Effects of Early Education: A Report from the Child Health and Education Study. Oxford: Clarendon Press.

  Sharp, C, Hutchinson, D and Whetton, C (1994) How do season of birth and length of Schooling affect children's attainment at key stage 1? Education Research, 36, 107-121.

  Sylva, K (1994) School influences on children's development. Journal of Child Psychology and Psychiatry, 35, 135-170.

  Sylva, K and Wiltshire, T. (1993) The impact of early learning on children's later Development: A review prepared for the RSA inquiry "Start Right". European Early Childhood Education Research Journal, 1, 17-40.

  Tizard, B and Hughes, M. (1984) Young Children Learning: Talking and Thinking at Home and at School. London: Fontana Press.

  Tymms, P (1998) Starting school: A response to Chris Whetton, Caroline Sharp and Douglas Hutchinson. Educational Research, 40, 69-71.

  Wolff, S (1999) Starting school: Do our children start too young? Young Minds, Issue No. 42, 10-11.

  Woodhead, M (1986) When should children go to school? Primary Education Review, 25, 10-13.

Sula Wolff, FRCP, FRCPsych, DCH
formerly Consultant Child Psychiatrist,

Royal Hospital for Sick Children, and

Senior Lecturer,

University Department of Psychiatry,


January 2000

previous page contents next page

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

© Parliamentary copyright 2001
Prepared 11 January 2001