If the Government wants to increase the cost-effectiveness of secondary school spending it should channel funds to maths and science rather than English, researchers have found.
It may also make sense to spend more on "average" pupils than those at either end of the ability spectrum, they say.
The highly-controversial recommendations have emerged from a study by researchers at the Institute of Education, University of London. Professor Rosalind Levacic and her team tracked 430,000 pupils between key stages 2 and 3 and compared their progress with what schools spent on them.
They found that spending an extra pound;100 per pupil raised KS3 maths and science scores by a small, but significant amount (0.04 of a level). But extra spending appeared to have no effect on English scores. "This may be due to the larger impact of home background and family linguistic ability on English attainment and the greater importance of the quality of formal learning for maths and science attainment," say the researchers, who were funded by the Department for Education and Skills. They also found cutting pupil-teacher ratios by one appeared to raise maths and science scores slightly, but had no effect in English.
Extra resources also had more impact on average pupils than the top and bottom 20 per cent of the ability range. "The impact of additional resources was also greater for higher-ability pupils eligible for free meals," the study said.
Increasing the number of teaching assistants also led to small improvements in maths and science but was less cost-effective than cutting pupil-teacher ratios.
Researchers have made many attempts to pinpoint the relationship between spending and performance but this has proved difficult because schools in disadvantaged areas get compensatory funding. It can consequently seem as if higher spending produces poorer results. However, the Institute of Education team adjusted for this and a range of other school and community factors that might have skewed their findings, which cover the years 20001 to 20023.
"Our study suggests that the impact of marginal increases in school resources is likely to be small," the researchers conclude. "But this does not mean that resourcing levels do not matter."
A second investigation into the link between spending and results has found that schools should invest in books rather than computers or extra staff if they want to improve performance.
Researchers who analysed data on 6,150 primaries concluded that schools could raise the average KS2 points score from 28 to 30 by spending an extra Pounds 100 a year per child on books. Spending on technology was only a tenth as effective, a team led by Steve Hurd of the Open University concluded.
"Estimating the Relationship Between School Resources and Pupil Attainment at Key Stage 3" is available from www.dfes.gov.ukresearch (click on published projects)