EEF Blog: Magic Breakfast - a case study in scaling evidence for impact
The Department for Education has this week announced it’s awarding £26 million to Magic Breakfast and Family Action to run morning clubs in over 1,770 schools across the country, focusing on the most disadvantaged areas. In this blog, we look at the EEF’s (continuing) role in this initiative and how it sits in our wider drive to improve how we support teachers and school leaders to scale evidence for impact…
The problem of scale
How do we scale evidence for impact, ensuring that children and young people from the most disadvantaged backgrounds get the maximum possible benefit from what we know about ‘what works’ in education?
It’s a key question for us at the EEF, given our explicit mission is to support teachers and senior leaders to improve outcomes in order to close the attainment gap.
We know how to generate evidence – we’ve so far funded over 160 projects, all independently and robustly evaluated, many with initially promising results.
But scaling evidence is the hardest challenge. ‘What works’, we know from countless examples in this country and beyond, often turns out not to work as it gets bigger. The programme developers become more distant, affecting the quality of delivery; schools make changes to key components, with unintended consequences.
Successful interventions can be quite fragile: stretched too far, they too often break. Programmes which not only raise attainment but are also capable of growing while sustaining that benefit are rare. Which is why, when we find them, we’re mustard keen to do all we can to scale that impact.
Magic Breakfast: our trial
A little over a year ago, we published the independent evaluation of EEF-funded project, Magic Breakfast. This found that its model of a free, universal, before-school breakfast club delivered an average of +2 months’ additional progress for pupils in reading, writing and maths.
Importantly for the EEF, the evaluation of Magic Breakfast was what’s termed an effectiveness trial. This means we were testing a scalable model of the intervention under everyday conditions in a large number of schools.
This is a tough test to set. Small-scale trials of targeted interventions with intensive developer involvement are much more likely to yield positive headline impact figures – but their practical use to most schools is limited by their reach. So Magic Breakfast’s achievement was especially notable.