In 2010, nearly four times as many children in Solihull reached the top grade - level 3 - in mathematics in key stage 1 tests than in Thurrock.

Solihull sits south-east of Birmingham and its average house prices are nearly double those of Thurrock, a unitary authority carved out of part of Essex in the 1990s. Other affluent areas such as Surrey and Richmond upon Thames have KS1 figures close to those of Solihull, while Rochdale, Leeds, North East Lincolnshire and Lambeth have results only marginally better than those in Thurrock.

Much of the difference is down to the arbitrary nature of local government boundaries across England. But widespread variations in educational outcomes have always been the case. No doubt many teachers will remember being shown the documentary Seven Up! during their training, highlighting the differences between children across the country.

Will the pupil premium change matters? Its aim must surely be to improve underperformance while challenging the better performers to continue to improve, rather than to sit back and wait for others to catch up. By doing so, it will, it is hoped, close the attainment gap.

At KS1, and especially where there are separate infant and junior schools, the use of free school meals (FSM) alone as an indicator for resource allocation may not be enough to channel the necessary resources. Indeed, the use of the FSM indicator as the basis for the pupil premium might spell the death knell of separate infant, nursery and junior schools.

More focus on the mathematical ability of those who seek to teach in the primary sector might be worth considering, as well as more work with parents. For just as in literacy, the offspring of affluent parents are likely to arrive at school understanding more about numbers than their counterparts from less well-off households.

John Howson is director of Education Data Surveys, part of TSL Education.