Heads and their deputies are too quick to delegate responsibility for children with special needs to classroom teachers, according to the Office for Standards in Education.
OFSTED inspectors also found that while governors are frequently ill-informed about the purpose of the Code of Practice for Special Needs they are reluctant to become greatly involved in this area.
In 62 schools surveyed by OFSTED between September 1994 and October 1995, inspectors examined how schools of all types were interpreting the code and what problems they had implementing it.
The code of practice, which came into effect in September 1994, has no statutory force but schools must "have regard" to it. OFSTED's report, published this week, provides the most up-to-date picture of its impact.
OFSTED pays tribute to the efforts made by SEN co-ordinators (SENCOs) who have striven to implement the code without extra time or help and, usually, on top of a full complement of teaching duties. But SENCOs, say the inspectors, can be "sandwiched between senior managers who delegate too much and classroom teachers who are resistant to accept their full SEN responsibilities".
The allocation of time for SEN duties was found to vary widely between schools. In one large primary school the SENCO was given just one half-day a fortnight for fulfilling her role, and there was a general complaint about the excessive paperwork generated by the code.
At the annual conference of the National Association of Schoolmasters Union of Women Teachers last month, Education Secretary Gillian Shephard sounded sympathetic: "My door is always open for detailed discussions on how bureaucracy can be reduced," she said.
Most governors "take too limited a lead" and prefer to leave special needs matters to the staff, says the OFSTED report. "Very few" nominated SEN governors are well-informed about the funding of pupils with statements, few governors had read the code, and "on many occasions they defensively stressed the fact that their services were voluntary and unpaid".
In general, the weakest aspect was a vagueness about money, making it impossible to judge how well it was being spent. Most SENCOs and governors do not know the basis on which their school was funded for special needs.
Staff in grant-maintained schools were much less likely to have taken in-service training in special needs and were less knowledgeable about support structures than their counterparts in maintained schools.
While most schools included a reference to the SEN policy in their brochures - albeit brief - some were "extremely reluctant" to advertise the fact that they had a large number of SEN pupils. This was true even if their arrangements for these pupils were good, fearing that this would be interpreted negatively by prospective parents or that if the school developed a reputation for good SEN teaching, more pupils with special needs would arrive and drag the school down in the league tables.
OFSTED found that the idea of parents as partners tended to be a good intention rather than good practice, while parents themselves were sometimes alarmed by the idea of the "SEN register", confusing it with the Child Protection Register.
Because of parents' very different levels of knowledge and commitment about their children's problems, SENCOs are finding it difficult to strike a balance between making them over-anxious and tackling indifference or ignorance. Again, lack of time is identified as the main obstacle preventing SENCOs liaising with parents.
As for pupils, the code recommends that they should be involved in drawing up and reviewing their individual education plans, but this, say inspectors, is rare.
research focus, page 13 The Implementation of the Code of Practice, a report by OFSTED, is available from HMSO at Pounds 7.95