GCSE results published earlier this year contained the surprise statistic that pupils who speak English as an additional language (EAL) had outperformed native speakers for the first time. But new research warns that the overall improvements for EAL pupils mask sharp regional differences.
A London Metropolitan University study has found that, while non-native English-speaking pupils are doing very well in inner London, the same cannot be said for other regions in England.
A TES report in February revealed that EAL pupils had made a breakthrough, with 80.8 per cent achieving five A*-C GCSEs last year - more than the 80.4 per cent figure for their native English-speaking counterparts. Nationally, EAL pupils have also been steadily closing the gap on the five A*-C GCSEs including English and maths measure since 2008, although native speakers remain slightly ahead.
But, using GCSE scores that included English and maths, academics found that, while inner-London EAL pupils outperformed native English speakers by 4 percentage points last year, their counterparts in Yorkshire and the Humber were 10 points behind, with pupils in the South West lagging by 8 points.
The study bears out experts' cautious reaction to EAL pupils' breakthrough on the general GCSE measure. The National Association for Language Development in the Curriculum urged against complacency, stressing that EAL pupils are not a homogeneous group.
John Bangs, senior research associate at the University of Cambridge, said it was important not to neglect groups of pupils still "bumping along the bottom", such as Pakistani and black Caribbean boys. "The bad move would be to simply say that children with EAL were doing better than other kids and leave it at that," he said.
The London Metropolitan University study warned that "crude ethnic categories" used in Department for Education data can make it difficult to fully understand the achievement of different minority groups. "If we are to get any closer to understanding the role of languagebilingualism and multilingualism in children's relative attainment, we need better data," the report said.
The researchers noted that "many of the widest attainment gaps are present in local authorities with substantial Pakistani ethnic minority groups ... who tend to speak Urdu, Punjabi or Mirpuri and experience economic disadvantage".
The study also calls for further research into ethnic communities from Eastern Europe, because their needs tend to be obscured by the "white" or "white other" ethnic categories used for government statistics.
"Similarly, black African ethnic groups need to be specified...to gain a fuller picture of their educational achievements," the report said, particularly in the case of pupils from central and east African countries such as Congo, Angola and Zimbabwe.
The researchers noted an earlier study based on 2008 key stage 2 results in London, which found that the black African ethnic category contained some of the highest and lowest-achieving groups. For example, black African Lingala, French and Somali speakers tended to perform worse than the lowest attaining group overall - black Caribbean. But Igbo, Yoruba and English-speaking black Africans achieved as well as the white British group.