We’ve just released the first part of our new research into the characteristics of high performing multi-academy trusts.
We wanted to understand whether there are any characteristics consistently associated with high performance and, unsurprisingly, we struggled to find any structural characteristics that consistently correlate. However, we did find some interesting patterns that highlight the diversity amongst MATs and raise questions for the way we research them.
One of a kind?
Our research looked at the size, geography, school types,
pupil characteristics and performance of MATs. We asked EPI to conduct a
cluster analysis to see whether MATs group together according to these features.
They found five clusters of MATs with different characteristics (described
fully in our report).
For me the most notable contrast was between the first and fifth cluster:
- Cluster 1: Small and medium sized MATs in tight geographical clusters and dominated by converter academies. These trusts have lower levels of disadvantage, EAL, low prior attainment, and SEN than other trusts. There is no clear pattern to their performance though they are more likely to demonstrate high performance on current performance than improvement. They are unlikely to have schools rated as inadequate or with high expenditure.
- Cluster 5: Medium and larger trusts (including system leader trusts) generally not in tight geographical clusters and dominated by sponsored academies. These trusts tend to have high levels of disadvantage and EAL. These trusts show a mix of results on measures of current performance – disproportionately high numbers below average in Key Stage 2 reading and Progress 8, with better performance in Key Stage 2 writing and mathematics – with a more balanced picture on improvement measures. These trusts do well for Pupil Premium Pupils at Key Stage 4. Over half have at least one school rated as inadequate and around half have schools with high expenditure.
These descriptions really stand at opposite ends of a spectrum: the newer, smaller MATs created out of groups of converter academies, contrasted with the larger, system leaders created through the early, pre-2010 sponsored academisations.
Previous research already shows that it’s only for the pre-2010 sponsored academies that there is clear evidence of academisation making a positive contribution to pupil outcomes. For me, our cluster analysis again highlights that the earliest trusts, which have now become many of our system leaders, may be considered a different ‘breed’ to the smaller, newer trusts arising out of academy conversion.
Much of what the sector currently knows about ‘effective MATs’ comes from focusing on the experiences of the early trailblazers. An important question for me is whether we can reasonably generalise from their experiences to the newer trusts growing, under very different conditions, in their wake.
What is ‘effective’ anyway?
The variation between different types of MATs doesn’t just
raise a question over the lessons we can generalise between them. It raises a
bigger question over the way we judge performance.
Even when we look at individual schools we can find multiple measures of ‘performance’. We tend to call them high performers if they have a good or outstanding Ofsted rating, and if attainment or progress is above average. But there are some schools where progress is amazing, but attainment is only average because of their intake – are they always recognised as high performing too?
When we look at MATs this picture becomes even more complicated. The original aim of academisation was to raise the performance of schools. So for the earliest MATs, containing mainly sponsored academies, (‘Cluster 5’) high performance meant significantly improving school outcomes and bringing schools out of special measures.
More recent converter academies, though, already have strong pupil outcomes and are very rarely rated below good. So for these small ‘Cluster 1’ MATs how should we define ‘high performance’? Is it sustaining their outcomes over time? Is it being rated outstanding? Is it collaborating with other schools beyond the MAT to share practice?
Again, which such variation in how we might define high performance, this raises the question – at what point are we comparing apples and pears?
"All good research throws up as many questions as answers"
All good research throws up as many questions as answers. In the case of this analysis, we’ve definitely shown that there aren’t simple formula for MAT success. However, we have opened up a point of discussion: how substantial are the differences between types of MAT and what might be the implications for the way that CEOs choose to lead them?
The exciting thing is that right now we’re analysing case studies and interviews with over forty MAT CEOs, from a broad range of contexts and sizes of MAT. This sort of large scale analysis will break new ground for the sector and will add to our understanding of the real complexities of leading a successful MAT.
This article originally appeared on the website of the Ambition School Leadership. In March 2019 the Institute for Teaching merged with Ambition School Leadership to form Ambition Institute.