Two new studies appear to contradict each other. The Welsh Assembly has tracked key stage 3 and GCSE results across local education authorities and compared them with a model of what results should be when set against the number of children entitled to free school meals - a key indicator of social deprivation. Meanwhile, Stephen Gorard of York university, and Peter Tymms of Durham’s curriculum and evaluation management centre, released research into English secondary schools’ performance when family income and previous attainment are taken into account.

The Welsh statistics may suggest, at least according to one academic expert, that LEAs can make a significant impact at the extreme ends of the achievement table. But the English value-added study found the adjusted results in different schools varied by less than 1 per cent between different authorities.

Particularly startling in Wales are the performances of Merthyr Tydfil and Neath Port Talbot. The former outperformed Monmouthshire, which would have topped a table of raw results, with 54 per cent of 14-year-olds achieving the standard expected for their age, 8 percentage points more than predicted. NPT topped the GCSE value-added table, with 57 per cent of pupils attaining at least five or more A*-C passes, 10 percentage points above its predicted rate.

Is the greater difference in value-added achievements between LEAs in Wales compared to England something to be proud of? Not if you are in Wrexham, which did significantly worse (9 percentage points) than expected in GCSEs, or Denbighshire which achieved 7 percentage points worse than predicted at KS3. The results may, of course, reflect the fact that LEAs’ powers are less circumscribed in Wales. But raise a glass to schools in Merthyr. They have shown the difference that staff, schools and LEAs really can make.