Kelvin Davis on improving education

When the government says that national standards, charter schools, league tables, performance pay, quality vs quantity of teachers will all raise achievement, they might be right.

That's because there are very few strategies that teachers (or governments) can implement that actually make students dumber. Teachers can rightly put their hands on their hearts and swear that what they do in class lifts achievement. Just about everything has some positive effect, but some have a large positive effect while others barely register. It would make sense to develop policy based on those strategies that have the greatest positive effect.

The much quoted Professor John Hattie's research lists, from most effective to least ffective, 138 different 'things' that may be implemented in education, and all but five have a positive effect on learning. The five strategies with a negative effect are: Summer vacation (-0.09), Welfare Policies (-0.12) Retention (Holding kids back a year, -0.16), Television (-0.18) and Mobility (-0.34). So unless we prescribe longer Christmas holidays, keep kids back a year or two, or force students to watch an extra 8 hours of TV a day, almost everything else will have SOME positive effect on learning.

The same goes for government policy - practically any educational policy will have some positive effect for some students.

In order to get the best achievement outcomes from any policy, the policy itself needs to be supported by research.

What does Hattie's research say?

Any 'strategy' with an effect size of 0.40 or less is practically pointless. Which makes sense.

In Hattie's list the strategy with an effect size of 0.40 (Reducing Anxiety) is exactly halfway through the list of possible strategies. Hattie is saying if any particular strategy is to be used it should at least be in the top 50% of strategies.

What does the research say about Charter Schools?

Charter Schools have an effect size of 0.20, or the 107th out of the 133 strategies that have some positive effect. Charter Schools are therefore an extremely pointless and expensive strategy.

There are still 40 strategies that are deemed pointless, but, are still more effective than Charter Schools.

What does the research say about League Tables and Performance Pay?

Nothing. They don't rate or feature in any way in Hattie's research.

What then is the basis for League Tables and Performance Pay if there is no research evidence to show these two 'things' will make a difference? How does the government know these two 'strategies' won't have to be included alongside the five already proven to make students dumber?

There are 106 'things' more effective than Charter Schools at improving learning, of which 66 are deemed to be very effective. 

It would make sense for the government to stick to what is proven by their guru's research to make a difference and really create the conditions where quality teachers can weave their magic.Only when the proven strategies are all implemented, should they pull out their ideological ideas. However, I suspect by then there would be no need.

By Kelvin Davis