Almost every week a Government minister or educational expert criticises Britain's failing schools and the low academic achievement of this country's school children compared with their counterparts in other industrialised countries. International studies have not only highlighted poor attainment and under-achievement in Britain, they have also identified countries where, by contrast, most pupils do well. In Europe, three countries or regions - Hungary, German Switzerland and Flemish Belgium - stand out as being particularly successful.

These countries use markedly different teaching techniques and Classroom Organisation from Britain. However, another important difference is coming to light: the way these countries approach pre-school education. This was the focus of the research behind the Dispatches programme, The Early Years. This booklet looks at why early years education in Hungary, German Switzerland and Flemish Belgium appears to help children do so well when they begin formal schooling, compares it with what happens in this country and suggests ways Britain can change.

Successive British governments have raised concern about educational standards in this country, particularly in comparison with other industrialised countries. These comparative studies not only highlighted low attainment and underachievement in Britain, they also identified countries where by contrast, most pupils do well. In Europe, three countries and regions - Hungary, German Switzerland and Flemish Belgium - have emerged as being particularly successful.

The success of these countries has attracted the attention of academics and others seeking to improve school performance in Britain. They have identified important differences - which have been widely reported -in teaching techniques and classroom organisation.

But another of their findings has been much less well publicised: the increasing emphasis that experts now put on pre-school education in explaining success. This was the focus of the research behind the Dispatches programme. It looks at early years education in Hungary, German Switzerland and Flemish Belgium and compares it with what happens in this country.

Every country expects its nurseries and kindergartens to 'socialise' children and every official devotes a lot of time to this. But much of what is written is in very general terms and has been ignored for the purposes of this report. All countries accept the importance of play in the education of young children. It is assumed throughout our report that effective early years education can only take place through encouraging development of social behaviour and 'play' of one sort or another.

We have set out to uncover exactly what children are being taught, how they are taught and the way their progress is monitored. The research also looks at the objectives of pre school provision and whether its organisation helps achieve these objectives.

The results took us by surprise:

• they confirm the critical importance of pre-school education in Hungary, German Switzerland and Flemish Belgium;

• they show that after stripping away marked cultural differences, it is possible to identify an almost identical approach in all three countries;

• they show that there are remarkable similarities between this approach and practice in the Pacific Rim, particularly in Japan but also in Korea and probably Taiwan.

The results have profound implications for Britain, which follows an exactly opposite approach to these countries If correct, they mean Britain will not be able to solve its educational problems until it reforms its early years provision. This applies as much to mathematics as any other subject.

Hungary sprang into the limelight with the 1990 International Assessment of Educational Progress (IAEP) carried out by the United States government. This compared educational attainment in 20 countries and showed Hungary and Switzerland to be the highest performing European nations, not far behind Korea and Taiwan. Hungary did less well in the broader 199oThird international Mathematics and Science Study (TIMSS) but ranked highly in the narrower 1996 Kassel study of mathematics, coming second to Singapore in the eight countries analysed.

Hungary also performed well in the 1992 International Association for Evaluation of Achievement (IEA) Study of Reading Literacy where it was praised for having particularly high achievement relative to expectation from socio-economic factors. After correction for the age of the sample. Hungary's 14-year-olds were graded second out of 32 countries.

This performance attracted the interest of Professor David Burghes of Exeter University who is using Hungarian teaching methods in his secondary and primary mathematics projects now running in 150 schools. His understanding of the Hungarian system has convinced him that Hungary's early years provision is crucial in explaining the country's high rankings in these surveys. He believes the full potential of Hungarian teaching methods will only be realised in Britain when they are preceded by a 'Hungarian style' pre-school cycle.

If anything, Switzerland's success is even more dramatic. In the 1990 IAEP study, while its pupils achieved about the highest average scores in mathematics, the lowest tenth of Swiss pupils performed far better than corresponding pupils in any other country. The tests in science showed Swiss pupils at a similar advantage. The 1996 TIMSS study confirmed Switzerland's position, showing it-with Flemish Belgium -as the top performing country in western Europe.

Switzerland did less well in the 1992 IEA Study of Reading Literacy, but given the high number of Swiss pupils being taught in a foreign language, the eleventh position out of 32 countries at age nine, and seventh position at age 14, remained impressive.

The Swiss performance has attracted the attention of researchers at the National Institute of Economic and Social Research and of the Barking and Dagenham Education Authority. Together they have introduced Swiss-style whole-class interactive teaching into the borough's schools. These were originally introduced into Year 4 classes (eight to nine-year olds) but it is now accepted that for many pupils this was too late to have ready impact.

An interim report on the reforms accepted that the minimum standards set for the Year 4 pupils were low 'because the foundations simply had not been established for the full breadth of the class. It is now generally agreed amongst our teachers that standards can be substantially raised if the initiative is introduced earlier.' Swiss methods are now being introduced in Year 2 and Year 1 classes.

But even this is viewed as too late. Now, in common with Burghes and his Colleagues in Hungary, those involved in the Barking and Dagenham reforms have become convinced that Swiss success is dependent on its pre-school cycle of education and that until a similar pre-school cycle is operating in Barking and Dagenham it will be unlikely to reach Swiss standards of attainment.

In the same interim report quoted above, it is noted that in Swiss kindergartens Children become used to making extensive oral contributions and have been given all the 'necessary poise and skill' to do so. It comments: 'teaching children to speak correctly and effectively to a larger audience is a key to the reduction of under-achievement at subsequent stages of schooling across the whole curriculum.'

In a paper on School Readiness and Pupil Attainment, Professor Sig Prais has taken the analysis further. He points out the importance the Swiss (like the Hungarians) place on children successfully completing kindergarten before moving on to school proper. He argues this is an essential part of Swiss educational success and concludes: 'Greater flexibility in age of school entry than currently practised in England may be a precondition for the extension of whole-class teaching and for more efficient teaching and learning.'

Those involved in the barking and Dagenham reforms are now actively investigating Swiss kindergarten practice and are anxious to move their own early years practice towards this.

Recognition of Belgium's educational success is also relatively recent. Although it did well in the first IEA mathematics study in 1967 (when it was second only to Japan) it was not until the TIMSS results in 1996 that it began attracting attention in Britain. At both age 13 and 14, Flemish-speaking pupils came ahead of all other European pupils in mathematics and very nearly so in science. Their performance was close to pupils in the Pacific Rim.

German-speaking pupils in Switzerland, and Flemish-speaking pupils in Belgium, have consistently outperformed their French-speaking counterparts. French practice-which is closer to that in Britain-is dealt with at the end of this chapter.

According to Hungarian educationalist Jozsef Nagy: Children with a calendar age of six can demonstrate a biological difference of plus or minus one year, a difference in mental development of plus or minus two and a half years and a difference of plus or minus three years in social development. And this is without including the least developed and most advanced, representing 5% at the end of each scale.'

In the early 1980s Nagy surveyed all the main school-based attempts this century to overcome such variation. He concluded that school was incapable of blurring these differences. 'The result,' he said,' is that the school career of those entering is predetermined by their stage of development on entry.'

Education systems in Hungary, Switzerland and Flemish Belgium accept this and believe that the problem can only be properly tackled in the pre-school cycle. All make it explicit (as do the Japanese and Koreans, what children should achieve by the end of kindergarten and emphasise the importance of this being achieved. In setting out such objectives for kindergartens, the 1996 Flemish Department of Education Core Curriculum states: 'It is important that as many children as possible achieve these objectives . . . it is crucial that problems are pointed out in due time and they are properly remedied,'

The objectives set out and the methods used to pursue them are consistent across successful pre-school systems.

In Hungary, Switzerland and Flemish Belgium the goal of pre-school provision is to prepare children for effective formal learning, which begins on entry to school proper at the age of six or seven. The implicit and sometimes explicit aim is that the pre-school cycle should reduce the socio-economic and genetic variation found in young Children and pass on to schools homogeneous groups of children who can be taught together and who are all ready for the formal learning and rapidly escalating whole-class interactive teaching they will experience at primary school.

Remarkably similar methods are used to achieve this in all three countries. In almost identical ways, children are taught:

• attention, listening and memory skills;

• appropriate group behaviour;

• conceptual understanding (seen as essential for subsequent mathematical success);

• phonological and motor skills (seen as essential for subsequent success in reading and writing).

Teaching is highly structured: it aims at slowly consolidating knowledge and confidence with tangible concepts before moving on to representational material. It avoids the abstract.

Teaching is dominated by an oral linguistic approach which places primary importance on developing mastery of spoken language. This skill is often given as the primary purpose of pre-school education. Emphasis is placed on young children being protected from failure or even the perception of failure, both of which are seen as immensely damaging to their subsequent school careers.

It is considered essential that children complete this specialist 'cycle' of education successfully before moving on to formal schooling. In all three countries arrangements are in place to facilitate this.

This is the first priority when children enter kindergarten and is viewed as the foundation of all that will happen subsequently. The teaching is done predominantly in whole-class groups which are often referred to as 'circle time' The teaching is highly structured, continuous and progressive.

ATTENTION: For the youngest age groups, a number of simple but effective attention-developing devices were used and observed in all the kindergartens visited in all three countries. The most common were eye contact games. For example, children sit in a circle around the teacher and have to catch her eye before being allowed to leave the group.

LISTENING: Similar devices were used for developing listening skills among younger children. The most common were 'stop/go' games in which children pursuing different activities must listen for a specific signal to stop, such as a drum beat or even a specific number or rhythm of beats.

For older children there are much more sophisticated listening and auditory memory games which are used across the curriculum, although they featured particularly in music sessions. The most common was discriminating between, and matching, musical instruments. For example: six instruments would be displayed and played in turn, a child would then be blindfolded while one instrument was picked up, played and then returned to its original place; the child would then have to identify that instrument.

MEMORY: As their attention and listening skills develop, children are introduced to games and activities intended to improve auditory memory. Such games build on attention and listening skills and make increasing demands on children as they grow older.

For instance: in a typical game (observed in Hungary) a bean bag was passed round a group of five year olds; when the teacher signalled stop, the child with the bag was asked to give a number between zero and five, the bag then continued to circle the group until the teacher again signalled stop and a second child gave a number. The bag circled again and when it stopped a third time, the child holding it was asked to total the two numbers.

The Hungarian Ministry of Education's Kindergarten Handbook puts particular emphasis on the value of music lessons in developing memory. For instance, it says children aged three to four should be taught to recite six short nursery rhymes. At four to five they should be taught to sing, with help from the teacher, three to four songs. By five to six, they should be taught to sing 10 songs on their own.

All pre-school systems aim at developing appropriate group behaviour. What stands out in these successful systems is just how thoroughly it is pursued. The Flemish Belgium kindergarten Core Curriculum says: 'being able to participate in co-operative forms of behaviour is a specific skill that can be learned and practised.'

The Swiss kindergarten Rahmenplan (meaning Framework) published by the Federation of Kindergarten Teachers, says kindergarten classes of 18 are ideal and should not be broken up too much. It argues: 'The child should find an identity and be able to feel at home in the group. An over-frequent and variable division of the group gives the child too little opportunity to form relationships.'

All children in all the kindergartens observed were taught to cope with formal whole-class teaching sessions at least once and usually twice a day. All these groups teach the linguistic skills and confidence required to sustain intensive interactive involvement with the rest of the class.

Conceptual understanding

Kindergarten teachers in all three systems see the development of conceptual understanding, that is the conceptual grasp of space. size, quantity and time, as the development of intellect and therefore an essential part of kindergarten practice.

Giving special help to slower children or those from deprived backgrounds is seen as an absolute priority. The aim is to move children forward from a simple, concrete level of understanding to the ability to manipulate these concepts at an entirely oral/linguistic level.

These skills are regarded as an essential preparation for later success in mathematics. In all three systems the conceptual understanding expected of every child by the end of kindergarten is set out in some detail. In all three countries this includes conceptual understanding of space (for example in, on, under); size (for example: longer; shorter); quantity (for example: as much as, fewer than) and time (for example: yesterday, tonight).

Once more, teaching is highly structured, continuous and progressive. Sophisticated examples of such teaching were seen in every kindergarten visited in all three countries.

In a Brugge kindergarten class of five and six-year-olds, for instance, the teacher targeted the concepts, 'as many as, biggest, smallest, bigger than, smaller than, in, on, between' as well as one-to-one correspondence using the class topic of 'rabbits'. The whole activity was carefully structured in that there was a progression from the hands-on comparison and manipulation of three dimensional material (lots of

different-sized toy rabbits) to comparison and spatial orientation of two dimensional representative material (drawings of different-sized rabbits).

In a Zurich kindergarten, two big drawings of ladybirds were placed in the middle of the class. Children took turns in following the teacher's instructions, such as, 'Give two leaves to the red ladybird and three leaves to the yellow ladybird.' The children were then told to make the ladybirds have the same number of leaves. After each turn, the teacher asked all the children how many more had been necessary to make it the same. The teacher then varied the game and told children to close their eyes while she fed the ladybirds. When the children opened their eyes they were asked which ladybird had more or less leaves and what they would have to do to ensure that each had the same number. A child would then add or subtract the appropriate number of leaves. Finally the children took away a follow-up activity for individual work: they were asked to draw in the correct number of leaves for the ladybirds shown on their cards.

In all cases, teaching involves ensuring children have absolute confidence with manipulating concrete objects before moving them on to representations of these objects. This is made explicit in the Flemish Core Curriculum which states that such activities should not be hurried and should not make use of abstract forms such as written numbers.

It continues: 'Too large and particularly too early an emphasis on the abstract may lead to a method of hearing and blandly repeating: of blindly applying learned procedures and reasoning at the cost of real understanding. When we give children time to gain understanding ... they will automatically have fun doing maths later on.'

Neither the Hungarian Kindergarten Handbook, the Swiss Rahmenplan, nor the Flemish Core Curriculum refer to the now extensive body of British and US evidence that identifies phonological skills-such as a child's awareness of rhyme, the rhythm of speech and the structure of a word or syllable-as the primary mechanism that promotes subsequent success in reading. Nevertheless the handbooks and kindergarten teachers in all three countries place enormous emphasis on teaching precisely these skills and clearly appreciate their relevance to the subsequent development of literacy.

Music lessons are seen as a key place for teaching phonological skills. The Swiss Rahmenplan is typical in listing the teaching of rhythmic expression, rhythmical games, differentiation of musical instruments by sound and pitch, musical dialogue in suggested tone sequences, rhyming presentations, grading and so on.

Extensive teaching of these skills was seen in every kindergarten visited and not just in music lessons. For example, the syllabic structure of words and sentences was taught in a Hungarian kindergarten by clapping or tapping out the rhythm of a well-known song. Children were also taught to clap out the rhythm of a song for others to identify. An awareness of sounds within syllables was taught in a Belgium kindergarten PE lesson by the teacher making the initial sound of children's names. As they heard their own sound, children left the large group and joined smaller groups of children whose names started with the same sound.

Similarly, the teaching of the gross and fine motor skills necessary for the subsequent development of proper handwriting was once again observed in every kindergarten visited. Exercises to achieve this were often highly structured. For example, in a Belgium kindergarten children were taught to make flowing letter-like shapes in a large sand table before being taught to make them in much smaller sand trays and then on paper.

Spoken language

In all three pre-school systems, the teaching of spoken language is emphasised above everything else. All three kindergarten curriculum publications make this explicit. The Swiss Rahmenplan is typical. It says: 'The development of social contact and thought is most dependent on perfecting [a child's] capacity for oral expression. The cultivation of oral language is therefore the most essential aspect of the promotion of mental facility in the kindergarten.'

In all three systems, whole-class groups or 'circle time' are used day in, day out to teach children the confident use of precise, accurate spoken language.

While spoken language is given such emphasis, written language is rigorously excluded from kindergartens in all three systems. Words, letters and numbers are neither used in lessons nor displayed in the classroom. In some kindergartens they even exclude children's names, using small pictures instead.

Avoiding failure

Kindergarten staff in all three systems were unanimous about the reason for rejecting written language. Their job, they said, was to hand over to primary school teachers, children who were confident and secure and ready for the rapidly escalating formal learning they would encounter at primary school. While some children at kindergarten can cope with reading and writing, many cannot: they are not ready to move from the concrete and representational to the purely abstract that reading and writing entail. Seeing others succeed while falling behind themselves would engender a sense of failure in children who were struggling, and thus undo all the work of the kindergarten.

For this reason, although kindergartens could not stop parents teaching their children to read, they did not encourage it. They pointed out that primary school teachers were skilled at teaching reading and writing and that it was better left to them. Similarly, although all kindergartens have books children can use, they are neither encouraged nor discouraged from doing so.

Although Hungarian and Flemish-Belgian children enter primary school at the beginning of our Year 2, having received no teaching at all in reading and writing, within one term almost every ordinary Hungarian and Flemish-Belgian child can read and write. Teachers talk of an 'explosion' in literacy. This is evident from examining exercise books: to British eyes the progress is staggering.

In German-Swiss primary schools, reading and writing move more slowly but the reasons for this are clear. Firstly, German-Swiss children must learn to read and write a different language from the one they speak. German-Swiss is a collection of local oral dialects, some far removed from standard written German which all Swiss children must master. Secondly, 17% of German-Swiss primary school children are foreigners being taught in a second language.

Nevertheless, Swiss 14-year-olds came seventh in the 1992 IEA international study of literacy. This impressive score included both German and lower performing French-speaking pupils. Adjusting for this raises the German-speaking Swiss into fourth position.

Interestingly, German-Swiss nine-year-olds did not do well in the IEA study. In an analysis of the figures, Philipp Notter of Zurich University concludes: 'A large part of this difference between the third and eighth class [nine and 14-year-olds] can be traced to the fact that Swiss Year 3 children read more slowly and work more thoroughly.

'Many ask the question: should children in the third class be required to read so quickly? Many are against it. As Diem (1990) established, one of the best ways of preventing dyslexia is to leave children the time to learn to read and write.

'The results of the Swiss Year 8 children, who found the test easy, shows that children who are taught more slowly [early on] later learn to read quickly and without problems.'

In handwriting too, despite their slow start, Swiss children seem to have no difficulty in overtaking British children. From direct observation Graham Last, a senior schools inspector for Barking and Dagenham, concluded in 1996: 'In handwriting, pupils catch up in a matter of weeks with their English counterparts in terms of the accurate formation of letters and consistency of size, and overtake them in the second grade [seven and eight-year-olds] where they learn to join up.

'Waiting until fine motor control and hand-eye co-ordination are well established through appropriate activities in the kindergarten allows children to make rapid progress.'

By contrast, the teaching of mathematics in the first year of primary school everywhere moves very slowly. Children only cover addition and subtraction of numbers up to 20. But observation of first year mathematics lessons and video evidence of Swiss children show them operating across these numbers at a more sophisticated level than British pupils and with total understanding.

There is no evidence that more able children suffer from their later start in reading, writing and recorded numbers. TIMSS found that the top 5% of pupils in both Flemish Belgium and German Switzerland comfortably outperformed their British counterparts in mathematics, with scores of

684, 685 and 665 respectively. It was the same for the top 25%: this time the scores were 609, 607 and 570 respectively.

French-speaking Swiss

The French-Swiss pre-school system, with its more academic, individualistic approach, has traditionally introduced reading and writing earlier than the German-Swiss system. In a major 1992 reform, the French-Swiss curriculum moved closer to the German system but still retained its emphasis on formal reading, writing and number work

The new curriculum, for instance, calls for children: 'To be put into contact with all kinds and forms of the written word ...' It suggests children 'should observe, compare and make sense of various written material ... contribute to the preparation of a kindergarten newspaper book, invitation or exchange of letters . . . write posters . . . communicate invitations in writing . . .' In mathematics it suggests that children should 'recognise numbers' and 'sort and classify according to number'.

Traditionally the French-Swiss education system has had a higher failure rate than the German-Swiss system. In 1990/91, for instance, French-Swiss cantons had between two and three times more pupils repeating years than German-Swiss Cantons.

In the 1992 IEA Literacy Survey, French-speaking pupils did significant worse than German-speaking pupils (a mean score of 521 compared with 543). Notter analysed the data to identify the role of socio-economic status, age, class size and make up (ratio of foreign language speaking pupils per class). He concluded that the 'weaker reading performances of French-Swiss school children cannot be explained by these factors.'

French-speaking Belgians

The French-Belgian pre-school system has a more differentiated approach than the Flemish system, with children moving into mixed age groups and pursuing formal reading and writing in kindergartens

according to their ability. In the French-Belgian kindergarten classes observed, children were surrounded by written material and being taught reading and writing.

French-speaking Belgian children experience greater problems with literacy than their Flemish peers. A senior French Ministry official said that at age 11, one in four French-speaking children have significant difficulty with reading and writing. It may be significant-given the evidence about the cumulative impact of early failure-that in the 1992 IEA study of literacy, the performance of French-Belgian children fell from thirteenth position at age nine, to twenty-third at age 14. Flemish children did not take part in the 1992 IEA study, but do not appear to have any general literacy problems.

In the 1996 TIMSS study, French-speaking Belgian pupils did significantly worse than the Flemish in mathematics and much worse in science. The relative position of French-speaking pupils in both subjects fell between the ages of 13 and 14, the two age groups tested: in mathematics from eighth to thirteenth, and in science from twenty-second to twenty-eighth.

The contrast between what happens in successful pre-school systems abroad and what happens in Britain could hardly be more complete. Britain does not do - or does badly - that which elsewhere is viewed as essential. And it does with considerable vigour that which elsewhere is seen as dangerous.

While British early years education is dedicated to prodding the wellbeing of children and developing their confidence, for many it actually does the reverse. The available evidence suggests that for many children British early years education is a disaster that diminishes the effectiveness of the entire education system.

It is to be hoped that fresh evidence will emerge that will reverse or at least soften this judgement. But at the moment, on the evidence available, it remains the only judgement possible.

Unlike the cohesive, structured pre-school systems of Hungary, German-speaking Switzerland and Flemish Belgium, Britain's provision is a chaotic hotchpotch without shape or direction. In 1995, Kate Buchanan, a researcher at the Institute of Education, surveyed the three main groups involved in nursery education: Montessori, the Pre-school Playgroups Association (PPA) and Traditional. She found that while there was agreement about aims (for example, to develop self-confidence and encourage a positive attitude towards learning) there was no agreement about what this meant in practice, beliefs or training of staff. The confusion is intensified by the range of provision and the increasing numbers of four-year-olds moving into school reception classes

Indeed confusion is endemic in British early years provision. Nothing better illustrates this than the Effective Early Learning Research Project which a third of local education authorities are using to try to evaluate and improve early yews provision. Yet it neither attempts to set out the purpose of early years education nor the outcomes that might reasonably be expected from it. Nor does it even define what is meant by quality. Indeed it specifically refuses to do so, arguing that quality is 'a value laden, subjective and dynamic concept which varies with time and place'. Nor is this an isolated view: it has the support of leading figures in the British pre-school world and even some abroad.

Such views have created a vacuum in British early years education and a confusion of purpose that has stunted the development of technique. Many nursery and reception teachers have - on their own - intuitively grasped the need for change. But inadequate training, lack of support and conflicting demands have made it difficult for them to achieve much. There are, in British terms, islands of excellence but they are few and far between.

The wide variation in practice makes it difficult to identify exactly what is happening within British early years education. Yet when its methods are compared to those of successful pre-school systems elsewhere, what little evidence there is points to dramatic failings.

A fairly extensive search of academic literature has found no British discussion of how attention, listening and memory skills should be taught or what skill level should be considered acceptable for children of different ages. There are complex references to 'child involvement', notably in the Effective Early Learning Research Project carried out by Worcester College of Higher Education, which probably refer to a child's attention span, but there is no hint here of the sophisticated attention, listening and memory teaching observed in Hungary, Switzerland or Belgium.

Many education authorities have introduced pupil profiles which are passed on as children move from state nurseries to primary school. These may refer to attention and listening problems but often only tell primary school teachers what will quickly become all too apparent.

Observation of early years classes does occasionally reveal good practice and sometimes good remedial teaching, but it does not reveal anything like the sustained and structured approach to teaching attention, listening and memory skills seen elsewhere.

Appropriate group behaviour

The development of social skills and the ability to interact with others would be universally accepted as among -he goals of British early years provision. But once again such aims have never been properly defined, acceptable skill levels for different age groups never established and methods for monitoring the acquisition of such skills never put in place.

And once again observation of early years classes does not reveal anything like the sustained, formal whole-class and small-class interaction which figures so prominently in the pre-school systems of Hungary, Switzerland and Belgium. It actually reveals the opposite: even when children are involved in group work they are usually pursuing individual activities which mitigate against the development of appropriate group behaviour by, as one experienced British observer put it to the authors, encouraging the egocentric behaviour so characteristic of young children.

Conceptual skills

At this point the contrast between methods used in British early years provision and successful systems elsewhere begins to become dramatic. As early years educationalist Chris Athey showed in 1990, few involved in early years teaching have been taught the importance of cognitive and conceptual skills or how to teach them. This, she says, is the area where improvement in early years education is most needed.

Observation shows plenty of so called 'experiential learning' but none of the carefully graded and monitored teaching of the essential concepts of space, size, time and quantity found in Hungary, Switzerland and Belgium.

Going by published reports and observation, there is no evidence that the importance of such teaching is understood in British early years education. In 1994 a study by speech and language therapists found 12 out of 28 children in a reception class had considerable problems with either language or phonological awareness which had not been detected but which would delay reading progress.

Spoken language

Recent evidence suggests that there is a stark failure of British early years education to teach effective spoken language, compared with successful systems elsewhere. A 1995 survey of 10 classes of four-year olds showed that out of 300 two-minute observations collected over three months, only 10 showed spoken interaction between children or between children and adults.

Another 1995 study shows that in one urban area, 25% of young children were showing signs of serious problems with spoken language. An early years literature search has revealed little or no awareness of the problem and little or nothing on how spoken language should be taught or its acquisition monitored.

Nor does it seem likely that this weakness will be rectified. The Curriculum for the training of Primary English teachers, which comes into force in September 1998, has six pages on the teaching of reading and writing but only two short paragraphs on the teaching of spoken language. These come after the sections on reading and writing.

If the problem was simply that British early years education was failing to do that which is considered essential elsewhere, the position would be serious. But it is far worse than this. British early years education is actively pursuing exactly those policies that have been identified abroad as being dangerous and damaging.

The ever earlier introduction of reading and writing

Professor Tricia David, who is responsible for early childhood education at Canterbury Christchurch College, in 1993 identified what she calls a Gaderene rush' to formal reading and writing in the British early years curriculum - actively encouraged by Government policy.

The desirable outcomes for nursery education drawn up in 1996 by the School Curriculum and Assessment Authority state that by the time children reach five years old, they should 'associate sounds with . . . syllables . . . words and letters'. They should also 'recognise their own names and some familiar words' as well as 'recognise letters of the alphabet by shape and sound'. In their handwriting it says, 'they should use familiar words and letters . . . and write their names with appropriate use of upper- and lower-case letters.'

The Baseline Assessment Proposals published by SCAA later in 1996 make even more frightening demands on young children. Within their first half-term at school (which today means when they are four and, for many, when they are only just four) children are now given tests to see whether they can: 'Recognise letters by shape and sound . . . read familiar words in a range of contexts . . . read simple texts . . . use identifiable letters to communicate meaning . . . write their names with appropriate upper - and lower-case letters . . . attempt to write sentences . . . attempt to spell unfamiliar words.'

It is difficult to see how any nursery can now avoid introducing the alphabet and formal reading and writing skills to three-year-olds. This is three to four years earlier than in the successful school systems of Hungary, Switzerland and Belgium. And it is being done despite overwhelming evidence that it will inevitably damage many young children.

Peter Blatchford, of the special needs department at the Institute of Education, in 1990 found a high correlation between children's pre-school skills and their performance at ages seven and 11. Dorothy Bishop, senior research scientist at the Medical Research Council's applied psychology unit at the University of Cambridge, in the same year found a clear association between expressive language problems in young children and subsequent difficulty in reading and writing among 83 British children she studied. She found that if these children's problems had been resolved by the age of five, literacy development was normal, but if not, they suffered persisting speech and reading difficulties.

Bishop believes a likely reason for this is that at five, the children in her 1982-84 sample were introduced to reading and writing and those who still had expressive problems were damaged in the process. This is in line with other evidence that children with poor spoken language skills have great difficulty with reading and writing.

The 1992 IEA study of reading literacy in 32 countries (Britain did not take part) showed that the age at which children began reading was associated with a gender gap in literacy. It concluded: 'It is clearly a plausible hypothesis that boys are too immature to begin reading formally at age five, and that their difficulties are represented in low achievement, relative to girls, at both age nine and 14.'

An as yet unpublished reworking of the IEA data for 27 countries found that in only four of these did children begin reading at age five. But that in all of these countries-and in only these countries - was there a gender gap in reading attainment at age nine. In Britain children begin reading at or before age five and a later replication by the National Foundation for Education Research (NFER) of the IEA test to include England and Wales suggests that Britain has the biggest gender gap of all at age nine.

In 1995, Marion Jongmans and her colleagues at the Institute of Education showed how accurately 6-year-old children perceive their strengths and weaknesses with motor and/or reading skills and raised the possibility that the perception of specific weaknesses could quickly turn into a 'global' sense of failure.

In his classic 1986 paper, Keith Stanovich, of the Ontario Institute for Studies in Education, set out the evidence - even then overwhelming - for why it is so important that spoken language and phonological awareness skills should be in place before children are exposed to formal reading and writing. 'Their absence can initiate a causal chain of escalating negative side effects . . . poorer readers read less than their peers . . . and deficient decoding skills result in unrewarding early reading experiences that lead to still less reading,' he said.

'Lack of practice . . .delays the development of automatic reading . . slow, capacity draining word recognition processes require cognitive resources that should be allocated to comprehension. Thus reading for meaning is hindered, unrewarding reading experiences multiply and practice is avoided or merely tolerated without real cognitive involvement . . . and because reading itself is an important contributor to the development of many language and cognitive skills . . . the downward spiral continues,'

Even by the mid 1980s research had identified every line in this downward spiral. A spiral so vicious, that American researchers have labelled it 'The Matthew Effect' after Matthew 13: 'For whosoever hath, to him shall be given, and he shall have abundance but whosoever hath not, from him shall be taken. even that which he hath.' It is a trap the pre-school systems of Hungary, German speaking Switzerland and Flemish speaking Belgium do everything they can to avoid. It is a trap into which Britain is now throwing its young children.

It might be thought that the early introduction of reading, writing and numbers would benefit the more privileged and more able. and give them an advantage over their peers in other countries who are deliberately held back. While this may have been true in the past (as recently as the 1990 IAEP survey, the top 5% of British pupils did very slightly better in mathematics than the top 5% of Swiss pupils) it no longer seems the case. Neither the 1992 IEA literacy survey, nor the 1995 TIMSS mathematics study show any such advantage. In fact the latter, as Professor Prais has shown, put the top 5% of British pupils below those in Belgium and Switzerland, as well as below those in Austria, France and the Netherlands.

Until there are proper controlled trials it will not he possible to determine the exact impact British early years practice has on the less privileged and less able. But many kindergarten teachers abroad have no doubt about what would happen if they had to follow the British approach. They say it would mean a lot of their Children, particularly boys would have difficulty coping and would fall ever further behind. This is exactly what the evidence suggests is happening in Britain

In 1992 a report from the NFER found that reading standards had fallen between 1985 and 1990 in 19 out of 26 local education authorities investigated. In the majority of cases the decline was due to an increase in the percentage of pupils in the lowest scoring groups rather than an all round decline. The report quoted a second study carried out by the NFER which confirmed that there had been a national decline in standards between 1987 and 1991 More recently, the NFER found a recovery to near 1987 standards between 1991 -1 99S but stressed there had been little or no improvement in literacy since 1948.

In 1995 the SCAA assessment of the Key Stage 2 Standard Assessment Test reported that almost half of all teachers said that children found the 'amount of reading required of them in the mathematics test somewhat daunting.'

In 1996 a further NFER study assessing British reading standards and comparing them with those established in the 1992 IEA study found that British nine-year-olds would have come sixteenth out of the 28 countries involved. It found that Britain's average score was lowered by a 'long tail' of pupils who achieved scores well below the average. It found, too, that pupils tested in 1995 and 1996 appeared to have made slower progress in the intervening 12 months than children had in 1987.

The 1996 TIMSS study showed that the relative position of England in mathematics had deteriorated since previous comparative studies. Average British 13-year-olds were at least a year behind other western European pupils and the gap was growing.

In 1996 Professor Alan Smithers and Pamela Robinson, of Brunel University, found the gender gap in British educational attainment had widened dramatically since 1979. In that year both sexes performed equally well at 0-level GCSE. By 1996, however, boys had fallen far behind with a score of 39.8 compared with 49.3 for girls. The gap is biggest in English.

In schools for children with learning disabilities in the UK, boys outnumber girls by two to one; in schools for those with behavioural difficulties, the ratio is nearer six to one.

Academics and others involved in early years provision in Britain are now issuing warnings about the damage being done to young children. But sadly, these warnings are having little or no effect. Few mainstream early years teachers have been trained in the specialist areas of cognitive and linguistic development which feature so prominently in successful provision elsewhere.

The available evidence suggests a significant proportion of children are being damaged educationally as a result. Britain's numeracy and literacy problems and the long tail of under-achievement indicate that the proportion of children involved could be as high as 30-40%.The evidence suggests, too, that because of their slower early development, boys are more at risk than girls. It is not unreasonable to speculate that this explains Britain's large and growing gender gap in school attainment and the worsening plight of boys in our education system .

Fortunately Britain does possess - in abundant measure - the expertise necessary to begin the process of rebuilding early years provision so that it can better serve, not only the needs of individual children, but the education system itself.

.

In comparing British early years provision with that in countries with successful pre-school systems, there is an unexpected twist. There are in Britain around 300 speech and language units attached to mainstream primary schools. Those associated with such units - specialist teachers, speech and language therapists, psychologists and academics - make up a professional group with an intimate knowledge of the cognitive and linguistic development processes that shape successful pre-school provision abroad. With colleagues in other areas of special education, they have all the skills necessary to tackle the crisis Britain now faces.

Only children who have utterly failed in mainstream education are placed in speech and language units. Dealing with the profound speech and language disorders of such children is only part of what these units do. They also deal with the educational and behavioural problems caused by these disorders. These often include extensive attention and listening problems, as well as a lack of basic conceptual understanding and literacy problems. The units are of variable quality but the best have developed an approach which, in all important respects, is identical to that in successful pre-school systems abroad.

The methods used in the best speech and language units are research-based and often superior to those in successful pre-school provision abroad.

The teaching of attention, listening and memory skills and the monitoring of their acquisition is part of the core training of speech and language therapists and specialist teachers working in this field.

Schemes to help teach and monitor these skills, such as the Wolfson Programme, have been in use for 20 years.

The teaching of appropriate group behaviour is another core part of the training of those involved with such units. Again schemes to help teach and monitor these skills, such as the hello or Social Use of Language Programme, have been developed. In dealing with nursery children, speech and language therapists have been progressively moving toward the whole-class techniques that feature so much in pre-school systems elsewhere.

The teaching of cognitive and conceptual skills has also been a major feature of speech and language units. Assessments such as the Boehm test have been used for years to monitor children's grasp of basic concepts and to facilitate intervention. Some speech and language units, such as those in Essex, have even developed their own programmes for this.

The teaching of spoken language and phonological awareness is, of course, central to the work of speech and language units. Once more, influential schemes, such as the Derbyshire Language Scheme and Living Language, have been developed to help teach and monitor these skills. More recently, programmes to specifically promote phonological awareness have been developed such as Metaphon, or the Essex scheme described by Clare North and Michelle Parker of Powers Hall Infant Speech and Language Unit in Witham.

One area where speech and language units lag behind good early years provision elsewhere is in teaching motor skills, particularly those necessary for handwriting. The need for this is widely agreed but resource limitations, curriculum demands and the need for specialist help have held it back.

Speech and language units - like speech therapy in general - are now accepted as achieving significant success in dealing with language delay and disorder in children. Much more important for the purpose of this paper however, is whether they are successful in dealing with educational problems. Here the evidence, while not extensive, is encouraging.

Given the severity of the handicap suffered by most children attending speech and language units, and their earlier catastrophic performance, it would be astonishing if the units returned them to normality. But while evidence shows that children's educational and behavioural problems often increase after leaving the units, it also shows that white in them they made considerable progress. This is borne out by the fact that around 60% of children in speech and language units make enough progress to allow them to reintegrate with their peers in mainstream classes. There is also a great deal of anecdotal evidence suggesting that at the point of reintegration, the attention and listening skills and even conceptual grasp of unit children is superior to many of their peers in mainstream classrooms. No research has been done on this but a study of mathematical skills does show children in a speech and language unit outperforming mainstream children on judgements of size and number.

This sort of evidence has, for some time, convinced many involved with language units that the methods they use would benefit younger, mainstream children. They have come to this view despite being quite unaware of the comparative evidence showing that precisely these methods are being used in successful pre-school systems elsewhere.

There is a good theoretical base for the view that the methods used in speech and language units would benefit younger, mainstream children.

For many years, cognitive psychologists and linguists have believed that all children move through the same basic stages of language development. It is now believed that most children with speech and language problems move through the same basic stages as other children, but either do so very slowly, or have got stuck at some point. So many of the methods used to help such children would also help mainstream children move through the same stages, albeit at a younger and more appropriate age.

Recent evidence about the impact of such methods on mainstream children, while limited, is positive. For example, in I 990 Chris Athey showed that teaching cognitive skills to under-privileged pre-school children for two years increased their IQ scores from 90 to 1 I O.The iQ scores of a more privileged control group of children remained unchanged at 125.

In 1993, Wendy Best and her colleagues at Riverside Health Authority showed that teaching attention, listening and language games to three-year-olds in a day nursery led to significant improvements in their ability to interact with others and focus on tasks.They also improved their grasp of basic concepts, despite the fact the extra teaching amounted to only two sessions a week (of about 35 minutes) for three months. A control group made no such improvement.

In 1997, educational psychologist Bob Daines and his colleagues at Wallands Primary School in Sussex showed a remarkable improvement in conceptual understanding among children in a mainstream nursery class after only a seven-week intervention trial. At the end of the trial there was virtually no overlap between the scores of the experimental and control groups.

Such trials depend upon co-operation from mainstream teachers. But the ease and enthusiasm with which some teachers pick up the new ideas is impressive. In one trial, a Year 1 teacher, who had been taught about phonological awareness, realised that some children with reading difficulties found the division of spoken words into syllables difficult to grasp. She formed the children into a band with musical instruments and assigned to each child the name of a grocery item which had the same number of syllables as those in his or her name. The children were then required to beat out the number of syllables, chanting either name or grocery item, and marching to reinforce the syllabic rhythm.

There is also good anecdotal evidence that individual teachers and even schools that have seen speech and language therapy and other special education techniques in use are trying to incorporate some of these into their own mainstream teaching.

But while this limited evidence does show that even brief intervention can achieve significant change in mainstream children, it also suggests any educational advantage is soon washed away as children return to conventional teaching styles.This is in line with findings about remedial reading programmes for mainstream children: impressive gains made during the intervention period diminish quickly for many of them after the intervention is withdrawn.

The evidence suggests that 8ritain's early years education, far from helping young children, actually damages many of them. Unlike successful pre-school systems abroad - which move slowly from the concrete to the representational and avoid the abstract - British early years provision rushes children into abstract letters, words and numbers. While elsewhere primacy is given to developing confidence and precision in spoken language, here teaching is dominated by reading, writing and recorded arithmetic.

While brighter children and those from more privileged backgrounds can cope with the demands this makes, less fortunate children suffer, lose confidence and probably never recover. It seems likely that this helps explain Britain's long tail of under-achievement.

The need to avoid damaging young children in this way is understood in the successful education systems of Hungary, German-speaking Switzerland and Flemish-speaking Belgium. This explains the emphasis in these systems on a highly structured, oral approach which always consolidates knowledge and confidence with the tangible before generalising to the more abstract.

Underpinning this is a desire in these systems to compress socio-economic and genetic variation in children before they encounter formal teaching. It is accepted that this cannot be done in the school system itself and must therefore be done in a pre-school cycle of education. Each of these systems has extended the role of pre-school provision to help achieve this goal: kindergarten teachers see their role as passing on to primary school homogeneous groups of children ready for the accelerating formal teaching they will encounter. This also allows primary schools to gain the full benefit of whole-class interactive teaching.

It is significant that while British early years provision rejects this approach, it has been developed quite independently by those involved in school-based speech and language units in Britain (and to some extent by those involved with other areas of special needs education). it is significant, too, that on the few occasions that their techniques have been used with mainstream children, the results have been impressive. There are good theoretical reasons why this should be the case.

The evidence suggests that until Britain adopts a similar approach for all its children, attempts to reduce under-achievement will achieve little. This is just as true for numeracy as for literacy.The evidence suggests early years practice that leads to competence in one will lead to competence in the other.

It also suggests that the introduction of such an approach must be accompanied by matching changes in later primary school years if the likely gains are not to be lost.

The evidence suggests that the best way forward is to take the techniques developed in speech and language units, adapt them for mainstream children, and transfer their use into mainstream classes. However, previous experimental use of such techniques on mainstream children suggests any educational advantage is soon 'washed out' when children return to conventional teaching. It is therefore important that the use of such techniques in early years education is accompanied by the introduction in primary years of Hungarian, German-Swiss and Flemish-Belgian teaching styles.

In this way the 'push' of good pre-school education and the 'pull' of matched primary education found elsewhere would be replicated in Britain. Fortunately the existence and success of the Gatsby school mathematics projects, with their emphasis on promoting precisely this form of teaching in primary schools, make this possible.

These changes would mean the introduction of a pre-school cycle lasting until the end of Year 1, with school proper delayed until the beginning of Year 2.The pre-school cycle would provide intensive preparation for reading, writing and recorded arithmetic but not actually introduce them. This would be done at the start of Year 2.

Literacy would then be taught far more quickly than at present, although it would continue the pre-school emphasis on oral, wholeclass methods. By contrast, numeracy would be taught much more slowly than at present, although this too would continue the pre-school emphasis on oral, whole-class methods and carefully consolidated progress. Differentiation would be used to help slower pupils keep up with the rest of the class, rather than enable quicker ones to move ahead.

The evidence also suggests that it would be important to put in place the same flexible school starting policies seen elsewhere to allow a few Children to move to formal schooling a year early while-more importantly-giving younger, slower or less privileged children an extra year in the pre-school cycle.

However strong the case for reform, the extent of the change that appears necessary is daunting. It cannot be done quickly, nor should it be. To do so would be to repeat a fundamental failing in British education: the refusal to test innovation properly before imposing it.

This leaves, though, an overwhelming case for an extensive trial of the pre-school methods outlined in this paper.

Subject to funding being found, the planning stage should begin at Easter 1998 and initially involve nursery school children who will move into primary reception classes in September 1998.These children should be taught in much the same way as three and four-year-olds in Hungary, Switzerland and Belgium. But the expertise of those involved with speech and language units should be used to draw up the actual programme. It will be essential that the teaching k done by mainstream nursery teachers who have been taught the special techniques. It would be helpful if these nursery teachers could see these techniques in use abroad before using them. The progress of the children should be monitored closely and the resulting lessons fed into a wider trial programme for nurseries starting in September 1998.

By the beginning of the school year 1998-1999, a full programme should have been prepared for the reception classes that the trial group will enter. The programme should again reflect teaching in Hungary, Switzerland and Belgium but once again its preparation should draw on the expertise of those involved in speech and language units. Once more, it wilt be essential that teaching is done by ordinary reception class teachers who have been taught the special techniques. Again it would be useful if they could see these techniques in action abroad before using them. This reception year, too, should be carefully monitored and the resulting lessons fed into a wider trial programme for reception classes starting in September 1999.

By the beginning of the school year 1999-2000, a full programme should have been drawn up for the Year 1 classes that the trial group will enter. As before, this should be based on practice in the final pre-school year in Hungarian, Swiss and Belgian kindergartens. It should be planned using the expertise of those involved in both speech and language units and the existing Gatsby projects. Again it is essential the teaching is done by retrained primary school teachers. As before, their efforts should be monitored and the resulting lessons fed back into the wider trial programme.

By the beginning of the school year 2000-2001, those responsible for the existing Gatsby projects should have drawn up a full curriculum for the Year 2 classes that the trial group will enter. Once more this should reflect Hungarian, Swiss and Belgian practice. The hope would be that by the end of Key Stage 1 tests, the trial group of children would have been brought to much higher overall standards of attainment than their peers in the conventionally taught control groups. If the approach has proved successful, experiences should be fed into the cascading trial programme following these children through their primary school careers.

It will be important that the children involved should continue to be taught in the same whole-class way during their subsequent years at primary school and that they should continue to be monitored.

Any such trial would require relaxing the national curriculum so that the children can be taken to the end of Key Stage 1 in a way markedly different from that currently required. It will be important that the schools and local education authorities involved are supported at every level in seeking such a relaxation.

Although the children involved will be taken through an intensive educational programme, the trial will involve postponing the introduction of reading, writing and written numbers by two to three years. Parents will need to know that if, at any stage, there is a possibility the children are not making the progress expected, the trial would be called off and the children given remedial help. It is therefore important to continually monitor their progress against that of a control group and that the results of such monitoring are offered to parents.

The trial must develop an approach which-if successful-can be easily and effectively replicated. Yet it clearly involves some re-training for the nursery, reception and primary teachers involved. Therefore we would need to bring in people with experience in training teachers in the necessary techniques.

If the trial succeeds, it will be essential that those responsible develop proposals for in-service training for existing nursery, reception and primary teachers, as well as proposals for specialist training schemes for future pre-school teachers.

The trial should include schools where there has been concern over standards. This is not simply to provide a better test of the methods being used, but also to give a measure of protection to the children involved. Given that the new method would have most potential benefit for children who are unlikely to do well under the existing teaching system, they would have little to lose in taking part in the trial but (along with their parents and teachers) much to gain.

The above document is available in full and with further information and contacts from Channel 4 Television:

Dispatches - The Early Years

ISBN 1 85144 212 x

£3.00