The North-South divide in terms of how well pupils do at age 5 virtually disappears once background factors are taken into account, new Department for Education research suggests.

The study into regional differences in attainment in the early years found significant variations - but the report says that few differences existed once factors such as ethnic composition and socioeconomic profile were taken into account.

“Overall, our findings show that, all things being equal, regional differences in attainment were generally small and not significant,” the report concludes.

The analysis, carried out by NatCen Social Research, found that in 2007 the regions with the highest proportion of children doing well in the early years foundation stage profile, which is a wide-ranging assessment carried out at the end of Reception year, were spread widely across England.

But by 2011, the North-South divide had become more prominent due to London performing more strongly.

And the report says that the data shows there is evidence that it was the social, economic and demographic changes in the regions which underpinned the differences in attainment between the regions.

Socioeconomic factors

It says that the children on free school meals in London and the South were more likely to achieve a good level of development at age 5 in 2015, compared with those in the North or Central England.

But when clusters of local authorities with similar levels of deprivation were compared, trends across areas which had similar characteristics were more stable over time.

The researchers, who also looked at children’s vocabulary and social skills at ages 3, 5 and 7 and key stage 1 test scores at age 7, concluded that where there were differences between regions, it was the differences in ethnicity and socio-demographic characteristics that largely explained the differences in attainment, although some “unexplained variance” remained.

The researchers said that some variance may be due to factors which are known to affect children’s development but were not included in the data they had access to.

 “One key factor that may be missing from our models was school-level information (school quality, teaching quality, class sizes, for example),” the report points out.

And the authors added that there was also no information on the quality of childcare included in the analysis.

The analysis used data from children’s early years foundation stage profile results, which are carried out at the end of Reception Year, and key stage 1 tests, which children take at age 6 or 7.

It also looked at data collected from the Millennium Cohort Study - a study which is following the lives of around 19,000 children who were born in 2000-01. This includes information on children’s development at nine months, their vocabulary at age 3, 5 and 7, and their social and emotional skills at age 3, 5 and 7.