Low performing primaries that have been converted to sponsored academies are improving at a slower rate than their conventional state school counterparts, a new analysis of national test results claims.

The figures cast doubt on the effectiveness of the government’s key primary improvement policy of turning schools with poor results into academies.

They also undermine the Department for Education’s assertion that “brilliant” academy sponsors are leading to faster increases in test scores than those in non-academy state primary schools.

The analysis of government figures by the Local Schools Network (LSN) compares primaries starting from similar test scores, with similar proportions of disadvantaged pupils. It suggests that in every case non-academies are actually improving faster than their sponsored academy equivalents.

Heads leaders say the findings show that the government is wrong to focus on school structures when trying to improve results.

Russell Hobby, general secretary of the National Association of Head Teachers (NAHT), said: “Where the government has only one solution to school improvement - which is support for [academy] conversion - I don’t think that is a sophisticated enough approach. This just proves that point.”

Last month the DfE highlighted the performance of the 570 sponsored primary academies, describing them as “schools which had long records of under-performance as council-run schools but which are now finally getting the help they need”.

“In 2013 the proportion of pupils reaching the expected level in the 3Rs increased by three percentage points in sponsored academies compared to just one percentage point in council-run schools,” the department said.

But its comparison is between the relatively small number of sponsored academies - which by definition are starting from a low base - and more than 13,000 maintained primaries, including thousands with good test results.

Because it is much harder to achieve large increases in test results in a school where scores are already high, the comparison is likely to be heavily skewed in favour of the sponsored academies.

To achieve a more accurate measure of how sponsored academy status improves a primary’s test results, the LSN compared schools starting from roughly the same point in pupil attainment.

Its analysis divides the primaries into five bands according to the proportion of their 11-year-olds reaching the expected level in national tests in reading, writing and maths in 2012.

The LSN, which campaigns against academisation, then looked at how much the results of the different types of schools increased in the 2013 tests.

The biggest group of sponsored academies were the 104 with between 60-79 per cent of pupils reaching the expected level in the 3Rs in 2012. A total of 6,138 non-academy primaries were in the same bracket.

On average the proportion of pupils in the sponsored academies reaching the benchmark the following year fell by 0.8 percentage points, whereas the non-academies improved by 3 percentage points.

Non-academies did better in all other brackets as well, according to the analysis, which excluded academies that converted after January 2013.

Henry Stewart, a member of the LSN, said: “The key question is if a school is underperforming what do you do with it? What this data shows is that the worst thing you can do is make it a sponsored academy because that will slow its growth.”

The Department for Education did not contest the analysis but repeated its claim from last month.

The network’s breakdown shows there were 81 sponsored academies with results between 40-50 per cent in 2012 and 1,337 non academies. The following year the sponsored academies went up by 5.3 percentage points and the non-academies by 12 points.

Other brackets had much fewer schools but told the same story. Just 12 sponsored academies and 129 non academies had results between 20-39 per cent in 2012. The non-academies climbed by 23.6 percentage points and the sponsored academies by 12.3 points, the network found.  

Primaries below 20 per cent saw an average 41 point increase for sponsored academies and 42 points for non-academies. Above 80 per cent  - where there were just two sponsored academies - both categories of primaries saw an average drop of 4 percentage points in 2013.

Mr Stewart then cross referenced the analysis with percentages of pupils who were eligible for free school meals or looked after children (FSMCLA) to ensure an even fairer comparison. Again, this revealed that, in all cases, non-academies were achieving bigger increases in their 2013 test results.

The NAHT’s Mr Hobby added: “There no panacea to be found in school structures. It is all about the quality of leadership and teaching and you can find outstanding examples in maintained schools and in academies.

A Department for Education spokesperson said: “Our figures are based on analysis of quality assured data in published performance tables. Sponsored primary school academies have replaced some of the worst performing schools, and have improved three times as fast as local maintained schools between 2012 and 2013.”