Around 12.5 per cent of primary children are classified as having English as an additional language (EAL), and concern is growing that their achievement is lower than native English speakers.
The overlap between EAL and ethnicity categories makes it difficult to disentangle the effect on achievement. Some black and Asian children may have English as their first language, whereas some white children may not.
The official statistics show in 2007, there were 17,500 seven-year-olds who are white but not British or Irish: 75 per cent reached the expected level in reading compared to 85 per cent of white British children, some of them will be included in the 75,108 EAL pupils, of whom 77 per cent reached level 2.
But by the age of 16, EAL is no longer a significant factor in achievement. Whereas at age seven children who are not native English speakers are behind in reading, by the age of 16 there is no gap between EAL and non-EAL pupils.
Professors Mel Ainscow, Alan Dyson and Frances Gallannaugh of Manchester University, and Dr Jean Conteh of Leeds University argue that the very term EAL creates its own problems.
In a report for the independent Primary Review, they condemn the way the 2000 national curriculum statement on inclusion aligns EAL provision with that of special educational needs.
They say presenting EAL pupils as being "linguistic and social outsiders separate from the mono-lingual mainstream" runs counter to the experience of bilingual pupils, who do not see their multilingualism as a problem but as a natural and normal part of life.
They add that such attitudes lead to lower teacher expectations and may also mean teachers miss opportunities to use pupils' existing abilities to help them learn English.
A government report, "Ethnicity and Education" in 2005, found that the gap between teacher expectations and key stage test results was more marked for EAL pupils: there was a five percentage point difference between their KS2 English test results and teacher assessments, compared with only a two point difference for non-EAL pupils.
When the Toronto Board of Education reviewed research in its area, it found that pupils given support in their first language were likely to do better learning English.
Dr Charmian Kenner, of Goldsmiths College, University of London, also found that six-year-olds were quite capable of learning to write in two languages at once, and teachers could invite children to share their knowledge with their fellow pupils.
Last September, a centre for multlingualism in education was established at London Metropolitan University. Its director, Dr Tozun Issa, is Turkish-Cypriot and learned English when he was 17.
"I think there is a lot of good practice going around," he said. "We want to pinpoint what it is that makes one institution more successful than others and to disseminate that ethos."
His latest project, which he is carrying out with his colleague Alison Hatt, is to examine cultural indentity and language in a Lambeth nursery where around 20 languages are spoken.
So far, they have discovered that children as young as three clearly understand that English is for the nursery and Portuguese is spoken at home.
Crucially, if staff learn just a few words in the child's home language, it helps to create a link between them, which enriches the relationship and helps the child realise that other languages are welcome.