How do schools raise the quality of their pupils’ achievements? Can local education authorities still have an impact? What do headteachers do to raise standards? What is happening inside the classrooms of teachers who succeed in increasing their pupils’ performance? Which individual children in a class improve, and what do they actually do?

These are a few of the questions we have addressed in the Leverhulme Primary Improvement Project, a two-year research programme I am directing at Exeter University. We have been investigating, at four different levels, what people do in primary schools to improve literacy in general and reading in particular. This is a preliminary account of analyses which I shall also be presenting tomorrow at the annual conference of the British Educational Research Association in Lancaster.

Even defining “improvement” is not as straightforward as it looks. Although we tested several hundred pupils at the beginning and end of the school year, we used reading test scores as just one criterion.

In our case studies, there were individual children whose test scores moved up only a little, but who devoured every Roald Dahl book they could find and whose enthusiasm for, and understanding of, what they were reading deepened considerably.

Using a “zoom” technique to move from the general to the individual, we studied: * local authorities in general and Birmingham in particular; * what happens inside a national sample of nearly 1,400 schools; * the processes, over a full school year, in the classrooms of 35 teachers in different parts of the country, ranging from Reception up to Year 6; * eight individual children in each of these 35 case study classes, 280 pupils altogether, to see which ones seemed to make good progress and which did not.

Responses to our national questionnaire from a random sample of 1,395 schools, came from every local authority in England. Most primary schools said they gave a high priority to literacy, as one would expect. But their hopes are pinned on people more than systems.

More than 90 per cent said that they had a language co-ordinator who took special responsibility for literacy - though 10 per cent of them were the head or deputy. However, many postholders are not paid any special allowance for their responsibilities, nor do they have any extra time to do the job away from the classroom (non-contact time).

The role of the language co-ordinator is generally regarded as crucial. Almost all schools said that a principal responsibility was to make sure that all teachers knew the school’s language policy, though in practice not all the individual teachers we interviewed were fully clued up.

Beyond this assignment, the 10 most common functions of the language co-ordinator, in descending importance, can be seen in the table. The need to monitor language teaching and give regular advice to other members of staff, mentioned by 84 per cent and 81 per cent of all schools, were regarded as being right at the top of schools’ priorities.

Almost all schools either have a written language policy or are currently preparing one. Most of these policies involve a wide range of people. About 90 per cent of schools said it was compiled mainly by the language co-ordinator and the class teachers, working with the head. Some three-quarters mentioned the special needs co-ordinator, about half cited the governors, and a quarter involved the local education authority and classroom assistants.

Most schools believe in the “pick and mix” approach when it comes to books and teaching methods. Some 99 per cent use reading schemes, with 54 per cent using several. About half the schools (key stage 1 - 48 per cent; key stage 2 - 47 per cent) employ colour coding to identify the difficulty of books.

More than 90 per cent of schools say that they prefer a mixture of teaching methods, though more mention the importance of “phonics” than “look and say” or “real books” when describing their main or favoured approach.

One of the most fascinating aspects of the project has been observing lessons throughout the year. We tried to identify teachers who were thought by their local authorities - or by their headteachers - to be likely to improve their children’s reading performance during the year. We observed a large number of lessons, gave reading tests to 316 pupils in 13 of the 35 case study classes and selected six “target” pupils in each class, three girls and three boys spread across a high, medium and low ability range, as well as two pupils who seemed to be making especially good progress (see the box on the three improvers).

In terms of the reading test scores, these predictions seem to have been generally accurate. When given a standardised test - such as the National Foundation for Educational Research group reading test - in September and again in the following June, a class that has upheld its position, compared with other children of the same age, should obtain the same standardised score. The actual test scores children obtain are converted to standardised scores according to their age, so an imaginary “national average” class would obtain a score of 100 on both occasions.

The average standardised score in the 13 classes tested went up by more than four points, using a regression technique to avoid the problem of it being easy for low achievers to “improve”, but harder for high performers. Four points may not seem a great deal, but it is a very significant gain for one single year. If improvements on this scale happened every year in a primary school, we would soon have a nation of geniuses, and the tests would have to be recalibrated annually.

What is more, the large majority of children (62 per cent) improved their relative position, making progress beyond what was expected, compared with 28 per cent who made less progress than expected and 10 per cent who conformed to the national average.

It would be tempting to report one single neat pattern beneath all these improving teachers and pupils. Unfortunately, the reality of classroom life is that no single pattern of success seems to emerge. The different routes to higher achievement are often as different as the people who travel them. Effective teachers find the most suitable strategies for their circumstances.

From studying teachers and pupils at close range, four features did stand out: * Most successful teachers were skilful class managers, securing high levels of “on task” engagement from their pupils, in some cases amongst the highest we have ever recorded in a research project.

* All teachers made reading very important, but by different means. Some celebrated each success, others labelled everything. Most well regarded primary teachers display their pupils’ work prominently, but two of the most successful teachers in our sample had little on their walls.

* Many of the highest pupil improvers had teachers who tailored their programme very closely to their needs, the “bespoke” rather than the “ready-to-wear” approach.

* Most of the effective teachers not only gave support, but fostered independence, equipping children with the skills to decode words they did not know and the confidence to monitor and decide about their own progress.

There were also contradictions. Many teachers deliberately invited strong parental backing, but some spectacular pupil improvers had virtually no help with reading at home. “Paired reading” worked well for many pupils, but one five-year-old, read to by an eight-year-old who was not a brilliant reader himself, said, “The trouble is, he just goes ‘Buzzy buzzy buzzy, gabble gabble gabble’.”

Some of the most significant gains were made in the classrooms of analytical and deliberately reflective teachers who could describe in detail, often using technical language, what they were aiming to achieve and by what means. There were others who operated much more intuitively, however, and claimed to be baffled about their success.

We still have a great deal of case study material to analyse, but it is clear that there are many different individual and group patterns of improvement. Some teachers’ classes improve across the whole ability range. Others manage particularly big strides with the better or poorer readers.

At the individual level, “horses for courses” is an important message. If children’s reading is to improve, then there are many ways in which teachers can learn to pick up the varying signals and messages from their pupils, experiment with different approaches, and match their strategies more effectively to individuals and groups - as did many of the most successful practitioners in our sample.

One further important conclusion is that there is still a place for personal inspiration and leadership, as has been shown by Tim Brighouse in Birmingham, and by many heads and teachers in our case studies.

The Leverhulme Primary Improvement Project at Exeter University is directed by Professor Ted Wragg. The research assistants are Dr Rosemary Chamberlain, Gill Haynes and Dr Caroline Wragg. The books detailing results of the project will be published by Routledge next year. Children and teachers names have been changed.

Ten major responsibilities of language co-ordinators

PERCENTAGE OF SCHOOLS MENTIONING

* Monitoring and evaluating teaching 84% * Providing advice to staff on a regular basis 81% * Advising on reading materials 68% * Leading in-service training 66% * Ordering resources 60% u Organising the school library 42% u Running meetings for parents 35% u Devising language resources 35% * Working with teachers in their classroom 35% * Assessing children’s reading problems 19%

BIRMINGHAM

Can local education authorities make a difference? Birmingham is well known for its high profile, and Tim Brighouse, the chief education officer, has invested a great deal in promoting literacy. Primary school budgets have been increased and various well-publicised events like the “Year of Reading” have taken place. Test results have improved.

We were in a unique position to identify some possible influences, as we had 155 questionnaire responses from Birmingham schools which we could compare with the 1,240 from all other LEAs. We had also carried out a number of intensive case studies.

The evidence is circumstantial, rather than conclusive, but Birmingham schools do show several highly significant differences compared with other LEAs, as the percentage of positive replies to certain questions show (see table) By contrast, Birmingham shows less involvement of parents in the classroom than in the other regions, which teachers find a constraint. The main reason being that as so many parents do not themselves speak English as a first language, they do not feel confident to help.

One headteacher told us that many parents were even more diffident, as they were not literate in their mother tongue either.

A noticeable difference at key stage 1 is the extensive use of “big books” with jumbo print in Birmingham schools. One teacher enthused: “We’re going to have a big book focus every day ... looking at the shape of words, trying to feel the sounds in the words.”

Perhaps the most striking feature in Birmingham schools is the impact of Tim Brighouse. Almost every teacher and head mentioned him, citing a note of encouragement, a visit, an address to staff, as examples of personal inspiration.

PERCENTAGE OF POSITIVE REPLIES...........Birmingham....... Others

Language co-ordinator has an allowance 86 49

Language co-ordinator has non-contact time 84 60

LEA involved in language policy 36 28

Current reading initiatives in school 83 62

Baseline assessment carried out 95 75

Use of target setting for whole school 66 24

School aware of current initiatives by their LEA 94 46

THREE IMPROVERS

Adam

Six-year-old Adam entered Emma Foster’s class when he could hardly read. Her reading corner is a mass of books, tapes, poetry, invitations like “Come and read” or “Do you have a favourite story?”, and a board for children to write messages to the teacher and each other. Emma taught Adam to use picture clues and to break down words. She kept a careful record of key words and sounds learnt. Adam grew in confidence and progressed well through the Oxford Reading Tree. His parents and grandparents were encouraged to help at home. When he got stuck on a word, they helped him sound it out, and his granddad bought him comics every week. By the end of the year he could read with expression, making excellent use of punctuation cues, and using the sense of the story to correct his errors. He decoded the text faultlessly and was able to recount the story in detail. Through skilful teaching, regular practice, growth in confidence, seeing the point of reading, and help at home, his standardised test score had improved from 78 to 98 points.

Charlotte

When Charlotte started Reception class a month before her fifth birthday, she obtained one of the lowest scores in the sample on a language test. Her mother said in interview: “We never read to her before she started school. I missed a lot of school myself, like, when I was a girl”. Soon Charlotte became excited at everything to do with language. “This word says ‘cow’, c-o-w, cow”, she would volunteer in a non-stop torrent when I observed herI “I can write my name“I “Do you know what this book says?” By the end of the year she was the second best reader in the class. I asked her why she thought she had done so well. “Because Daniel (the best reader in the class, who sat next to her) gives me all the words,” she smiled, proud at having signed up her own free private tutor. Charlotte is a smart girl who needed school to discover her talent in language and personal relationships. By the end of the year she was asking her mother to spell words she did not know and showing her how to split difficult words into two. She will go far.

David

At the beginning of the year nine-year old David was adamant that he “absolutely hated reading. It’s boring. I just don’t like it”. Only his Liverpool Football Club comic was of any interest. There was no one to read to him at home and he was a seasoned procrastinator in class, able to turn finding a pencil into a five-minute task. By the following June he was transformed. “Reading? I really like it.” Asked why the change of heart, he was equally unambiguous: “It’s Mrs Jackson. She’s given me some really good books, adventures and that sort of stuff, and books that made me laugh”. Patrick Burson’s The Funfair of Evil had been particularly enjoyed. Mrs Jackson’s personalised approach increased the time and effort he spent on reading. Although his test score had only improved slightly, his attitude and the number and range of books he had read had soared.