Evidence-based practice in schools.

 The physician, academic and science writer, Ben Goldacre, posted this welcome news on his blog last Friday

Here’s my paper on evidence and teaching for the education minister.

March 15th, 2013 by Ben Goldacre in evidence based policy |

I was asked by Michael Gove (Secretary of State for Education) and the Department for Education to look at how to improve the use of evidence in schools. I think there are huge, positive opportunities for teachers here, that go way beyond just doing a few more trials. Pasted below is the briefing note from DfE press office, and then the text of a paper I wrote for them, which came out this week. You can also download a PDF from the DfE website here.

If you’re interested, there’s more on evidence based policy in this BBC Radio 4 documentary I did here, and in this Cabinet Office paper on trials in government that I co-authored here, as well as zillions more posts.

There’s a response to my DfE paper from the Education Endowment Foundation here (they’re running over 50 trials in 1400 schools), and a blog post from the Institute of Education here, I’ll post up more when I get a chance.

Hope you like it!

 In an article in today’s edition of Education Guardian Goldacre argues that if  teachers want politicians to base policy on evidence, they need to accept that randomised trials –very much much like those which used when evaluating various treatments in Medicine – are the way to show what works.

Medicine, in just a few decades, has leapt forward with evidence-based practice. By conducting “randomised trials” – fair tests, comparing one treatment against another – we’ve been able to find out what works best. Outcomes for patients have improved, through thousands of tiny steps.

There are many differences between medicine and teaching, but they have much in common. Both involve craft and personal expertise, learned through experience; but both can be informed by the experience of others. Every child is different, and every patient, too; but we’re all similar enough that good-quality research can show which interventions work best.

It seems to me that this approach might yeild better results than we have seen from other approaches.