The child experiment

Tim Harford of the FT reckons that the evidence is in favour of charter schools proving education. Looking at the mixed evidence from the Swedish and American experiences, he concludes: “This suggests that more choice can raise standards in British schools. The Conservative policy is well worth trying.”

But is it more choice that causes the positive effects that Harford spots? Elsewhere in the column, he suggests that it’s not “choice” that makes the difference – rather it’s one quite specific educational approach:

The most credible research studies a particular subset of schools, typically in New York and Massachusetts. They are typically “no excuses” schools that emphasise discipline, long hours and short holidays. They are oversubscribed. The disadvantage here is that researchers are looking only at charter schools that parents already reckon are succeeding. But the advantage is that places are allocated by lottery and so researchers can compare lottery winners and losers. The encouraging conclusion is that such charter schools can be dramatically effective, especially for poor children and ethnic minorities.

Is there anything that makes that educational approach specific to charter schools? I don’t think so – and that makes charter schools (whose appeal is that they can work to any educational ethos the founders prefer) a pretty inefficient way to roll out this one potentially beneficial aspect. The Tories seem to be adopting only the part of the experiment that suits their own ideology of choice, rather than the part that actually works.

There’s one more point made by Harford about the way parents will tend to exercise that choice: “Parents do like schools with low poverty rates – which might push towards segregation…” On Twitter, he added that economic streaming was “already such a severe problem, not clear whether it would be made worse or better”. But if this is already a severe problem, why aren’t we moving towards a system designed to fix it – rather than the one aimed at the not-terribly-urgent issue of maximising the performance of wealthier children who are already doing pretty well? It’s a terrifyingly unjust way to reform the education system. Under this government, the poor are deemed lost from infancy, and in the choice experiment, they’re simply expendable.

Text © Sarah Ditum, 2010; photo by Squiggle, used under Creative Commons

5 thoughts on “The child experiment

  1. What system can “fix” economic streaming, which is indeed the underlying problem? In Bristol there’s a direct correlation between quality of school and the cost of housing in the schools catchment area.

    If you fall outside one of the “quality” catchment areas, your “choices” are pretty stark. I think on that basis many people are prepared to look at free/charter school models.

  2. Implementing successful policy rather than in self-selecting areas maybe? Choice means very little if you don’t have the means to exercise that choice – it does take money to travel further to reach that good school, say, and if parents prefer schools with a well-off intake, the schools in poor areas aren’t going to improve much.

  3. It is unclear what the theory is behind the idea that choice, in itself, improves educational outcomes. In economic theory, choice helps to reveal what customer preference is: if supermarkets only stocked plain white wheat flour no-one would know that there is a demand for a wide range of other flours. In education the problem is different: there is a demand for better education as measured by some well-known indicators and there is no need for a process to reveal that. The problem is how to operationalise better education on a large scale.

    If School A starts to attract more applicants, School B may attract less applicants but the barriers faced by School B to adopting innovations that will help it to compete against School A may be too great to overcome. The usual outcome is that School B continues on a spiral of decline and eventually closes, not that it is spurred to adopt improvements. If there are innovations that are shown to improve education, the State should be helping to spread them to those schools that need them most, not expecting that schools that have most difficulty will be able to adopt them spontaneously.

  4. I cannot understand such a proposal in times of economic uncertainty. In a purely financial sense, money saving is to be achieved via economy of scale; it makes more sense to merge smaller schools to share resources and services than to fracture the system to accommodate smaller schools championing the educational fashion-de-jour.

    At the moment division is widened by parental choice. If exam results fall, school A may see a fall in roll as parents opt for school B. Over time school A will experience further falls as parents with more reason or ability to choose move away from it. Very quickly there is a rich/poor divide. The same figures based success criteria discourage good SEN and EAL provision. A purely geographical selection system would not eliminate division, but would at least provide a reliable year-on-year performance from schools and stop non-selective schools cherry picking.

Comments are closed.