Schools are wasting millions of pounds of pupil premium funding by spending it on projects that show little evidence of providing value for money, a report released today reveals.

A survey of nearly 1,700 teachers has shown that schools are squandering funds for pupils in receipt of free school meals on activities that are not cost-effective, such as cutting class sizes and hiring more teaching assistants.

The research comes just months after Deputy Prime Minister Nick Clegg warned heads that Ofsted would be looking “forensically” at how well pupil premium pupils performed in school.

At the start of the academic year, the Department for Education doubled the amount of pupil premium funding it intended to give schools to pound;1.25 billion - the equivalent of about pound;630 per pupil on free school meals.

But according to a survey by the National Foundation for Educational Research on behalf of the Sutton Trust, a charity that promotes social mobility, one of the top choices among schools was to spend their pupil premiums on reducing class sizes, despite evidence showing that it has no impact on attainment unless sizes are cut to fewer than 15 pupils.

One in six teachers surveyed said their school was paying to reduce class sizes. Eight per cent said their school was using the money to hire more teaching assistants, even though there is little proof that such tactics boost attainment.

Lee Elliot Major, a director at the Sutton Trust, said the most cost- effective way of spending the premium would be on more effective feedback in the classroom, but only 2 per cent of respondents said this was how their school was using the money.

The Sutton Trust has produced a toolkit that suggests how schools might spend the pupil premium more wisely - on peer-to-peer tutoring, for example.

“Evidence shows that bringing in extra teaching assistants has zero impact on improving attainment, and that while one-to-one tuition is very effective, it is also very expensive. Reducing class sizes only works when the number is cut to fewer than 15,” Mr Elliot Major said.

“We are not saying, `You should do this’,” he added. “What we are saying is you should not worry too much about trying to reduce your class size by two pupils, and instead look at providing better feedback to pupils, as this can have a major impact for a lot less money.”

In May, the deputy prime minister said he would be receiving regular reports from Ofsted on how well the pupil premium was being implemented. “Schools need to know that, in assessing their performance, Ofsted will be looking forensically at how well their pupil premium pupils do,” Mr Clegg said. “Inspectors are already being instructed to look closely at how schools are spending the money and to what effect, with plans to publish a survey early next year.”

By the end of this Parliament, in 2015, the pupil premium will have doubled again to pound;2.5 billion, so it is expected that schools will be held even more closely to account on how the money is being spent.

Commenting on the Sutton Trust’s survey, Russell Hobby, general secretary of the NAHT heads’ union, agreed that spending the money on reducing class sizes and one-to-one tuition “was not always a wise use of the premium”.

“I am not surprised that schools are choosing these methods, particularly one-to-one tuition, because they are the easiest ways for heads to demonstrate they are using the money to help pupils who need it most,” Mr Hobby said. “There is a concern about sanctions being placed on schools on how they spend their money, but there is no doubt it is best to look at the evidence that is out there.”

Dylan Wiliam, emeritus professor at the Institute of Education, University of London, said that improving the feedback given to pupils was a good method, but one that was very difficult to achieve. “I have been trying for 20 years to make teachers better at giving feedback, but it’s very hard,” he said. “If you have spare cash, use it to create time for teachers to plan their lessons better.”

Original headline: `Zero impact’: survey finds flaws in pupil premium use