Compared with the potential minefields of the literacy debate, the task facing the architects of the National Numeracy Project has been relatively simple. Maths teaching in this country simply does not have the equivalent of the long-running phonics versus "real books" arguments, which have tended to make teachers feel under siege, clinging defensively to their preferred methods. The problem with maths, for many teachers, derives far more from a lack of confidence in the subject, resulting in an over-dependence on commercial schemes to tell the class what to do.
So it is perhaps not so surprising that the new ideas, directions and structures of the National Numeracy Project have received a warm welcome from teachers in the 12 participating local authorities. The rising star of the project is clearly mental maths, now featuring at the start of every maths lesson, in a tight-knit structure consisting of 10 minutes' oral work, 30 minutes' group activities (grouped by ability), and a 10-minute plenary at least four times a week.
Teachers have become much more the focus of the lesson, and the oral warm-up is a sort of performance, featuring various props and children coming up to the board to contribute. The idea is that the warm-up should, where possible, be related to the main part of the lesson - not simply times tables but decimals, fractions or place value.
"Teachers are finding it quite demanding, and need to be shown new strategies for these warm-ups," says Ruth Merttens, who is helping to train project teachers, and is herself co-author of the Abacus maths scheme (which emphasises whole-class teaching). "But they are gaining in confidence as the children make progress, and discovering how to fit their own personalities to the teaching. The children are enjoying it, too; they are even asking me for more of my 10-quick-question tests."
Sue Gifford, senior lecturer in maths education at the Roehampton Institute in London, believes the project framework is drawing on good research on children's mental calculation strategies. "In the past, mental arithmetic was about testing speed, and about public humiliation when you got the wrong answer. Now teachers are focusing more on method, and saying to children, 'Tell me how you did that', which is good for children's self-esteem and good for teachers' confidence and knowledge."
"Because of the shift to regular, daily practice of mental arithmetic, teachers are noticing a difference in children quite quickly," says Lynn Churchman, numeracy project manager for Lancashire. "The project is also encouraging a healthier use of commercial materials."
The schools finding it most difficult to adjust to the whole-class and group teaching of the National Numeracy Project, according to its national ( director, Anita Straker, are those that have been most reliant on commercial schemes: 30 children each working on a different page of a maths book, with the teacher rushing from one to the next as they get stuck. Some schools will need extra support, she says, particularly with the group work in the middle part of the lesson.
Mary O'Sullivan, headteacher of St Joseph's Lower School in Bedford, says the teacher's role in lessons has developed as a result of the project, and "although we were not quite prepared for how tightly planned it is, teachers have welcomed it". Lessons are very structured and much pacier now, she says, and the children more "on task".
Harry Weightman, headteacher of Easington Colliery Primary School, in east Durham, is enthusiastic about the oral component of the project, which has already improved his pupils' "number power" and won the support of parents. "I especially like the plenary session: it tidies up the lesson, and focuses on what the children have achieved in it."
But, as with the National Literacy Project, for many schools this shift of emphasis to whole-class work brings attendant worries about providing for children of different abilities. Next term, Ramridge Junior School, in Luton, begins its work on whole-class teaching with the project consultant, and Hew Wright, the headteacher, is apprehensive. "I see more positive things in the project than I did before, such as the planning framework, which is first class," he says. "But the one concern I have is, does it allow for differentiation, so that all children make the best progress they can?"
Setting children for maths is one way of reducing this difficulty, and David Panther, project consultant for Bedfordshire, says more schools have begun to set their 7 to 11-year-olds. But class size is also important for the success of the project, according to Ruth Merttens. Thirty is the upper limit for it to work, and classes on the continent, which teach maths in a similar way, are considerably smaller; classes of 35 or 36, however, pose a serious problem, with four or five children at either end of the ability spectrum not getting sufficient attention.
The complexity of the framework also provokes concern in some quarters. "The over-detail could be worrying," comments Sue Gifford. "A strictly sequential programme might not match all children's learning development, and there may be an assumption that once you have dealt with a topic the children have learnt it. Much will depend on the way the work is monitored."
"Teachers tell me it is prescriptive," says Anita Straker. "But when I ask, 'Is it too prescriptive?', they say, 'No, we still have some flexibility, but we have guidance and a structured programme to work around'.
"I hope this framework will be useful when the national curriculum is revised. My impression is that teachers would like clearer guidance for literacy and numeracy, and although the pruning down of the national curriculum was helpful in one sense, in terms of the core subjects it may have gone too far."