Figure hugging;Leading Article;Opinion
The publication of council spending plans for schools this week could be a case in point. What amounts to a spending league table is predicated in part on the assumption that devolving more cash to schools is good, while spending by councils is bad. And yet in the same week (page 7) Office for Standards in Education reports on Buckinghamshire and Durham point to the importance of effective monitoring and support services for raising standards in schools.
Kensington and Chelsea spends more than twice the average for London authorities on its statutory and regulatory duties. Bureaucracy, if you like. It also delegates one of the lowest proportions of schools' budgets. And yet in The TES's own analysis of national test results (TES, January 29), pupils in the borough spectacularly outperformed those in authorities which have far fewer children on free school meals. Perhaps what is done with the held-back money is more important than how much is retained.
On page 22 this week we publish a fascinating analysis by Lancaster University economists which concludes that competition between schools has improved their technical efficiency. It compares certain inputs - children not eligible for free meals and numbers of qualified staff - with outputs (attendance and GCSE results). Competition was simply inferred from the proximity of another school.
The results seem directly to contradict an Open University study reported in The TES last year. But that was based on heads' assessment of the reality of competition locally. The Lancaster findings are also at odds with OFSTED pronouncements on class sizes and whether the lowest-performing schools are improving faster or more slowly than the best.
The fact is that the most highly sophisticated analyses are only as good as the assumptions and the data that underpin them. The quality of the information upon which practice is to be based is crucial, whether at the level of the self-improving school, the local authority, or the education service as a whole. If we truly want to find "what works" best, we need to collect better information, and focus more critically on the assumptions which underlie research.