Free meals for school kids do improve performance, but value for money questions remain
This article took some of the conclusions from the evaluation of the pilot scheme for free school meals and extrapolated them to say how effective the current policy would be. Following discussions with the authors of the pilot scheme report, we no longer felt these conclusions were fully reflective of the research findings. We have a newer piece on the topic which goes into this in more detail.
"The evidence, and this has been exhaustively analysed, piloted, examined, is that giving a healthy hot meal at lunchtime is as, if not more, effective than many of the, say, literacy and numeracy initiatives which have been undertaken in the past in the classroom. It has a dramatic effect." - Nick Clegg, speaking on the Today programme
It's true that the performance of children in schools where universal free meals were piloted improved as much, or more, than under two policies specifically targeted at improving literacy according to the Government's impact assessment of the policy.
The free school meals policy launched today by the Deputy Prime Minister applies to pupils aged 4-7. Under the pilot scheme, an extra 1.9% of children achieved the expected level of reading at Key Stage 1 (age 4-7) and an extra 4% at Key Stage 2 (age 7 to 11). The figures for maths were 2.2% and 5.5% respectively.
That said, some of the other educational initiatives that the free hot lunch scheme were compared to were found to have produced better value for money.
The free lunch pilot improved the performance of children, but other schemes have achieved the same for less
For the pilot all children at primary schools in Newham and Durham were made eligible for free school meals between September 2009 to July 2011. Their diets and performance in class were compared to children in neighboring areas to get an idea of the impact of the policy. In turn, this impact was compared to the impact of three other schools policies which ran for similar time periods and measured achievement in similar ways.
For Key Stage 1, the recent free school meals pilot was compared to the 'Every Child a Reader' scheme, which provided one-to-one support to infants who were struggling with literacy.
For literacy the pilot was estimated to have cost £235 per percentage point improvement compared to £295 for the Every Child a Reader scheme. The Every Child a Reader scheme had no significant impact on maths performance.
For Key Stage 2, the pilot was compared to Jamie Oliver's 'Feed Me Better' campaign, which involved him improving the quality of meals provided in schools and to the Literacy Hour, an hour-long period set aside for developing reading skills.
For literacy the pilot cost £112 per percentage point improvement. That compared to £16 for the Feed Me Better campaign, and £14 for the Literacy Hour. For maths, it cost £81 per percentage point compared to £15 for the Feed Me Better campaign. There wasn't any data for the Literacy Hour.
As the assessment points out, there are a couple of things to bear in mind when making these comparisons.
The free school meals pilot isn't just aimed at improving exam performance; it's also aimed at improving the health of children, unlike the Literacy Hour and Every Child a Reader. The assessment found that an extra 26% of children at participating schools ate vegetables, 20% drank water, and 13% fewer reported eating crisps at least once a day. There was no evidence of an effect on the Body Mass Index (BMI) of pupils.
And the Feed Me Better campaign had what it called a "potentially unquantifiable benefit associated with Oliver's involvement in the campaign", meaning the benefits to children's performance might not have been entirely down to the policy itself.
This article was edited to make clear that the percentages used in the report refer to percentage point increases. For example, a rise in the number of children studying French from 30% to 60% would be a 100% increase in absolute terms, but an increase of 30 percentage points. Or in other words, an increase of 30% of the total number of children.