This paper, by Professor Becky Francis and her research team at the IOE, which described setting of pupils in English schools as being a form of ‘symbolic violence’ against those who find themselves in the bottom sets, has provoked much discussion on my edu-Twitter timeline. I like to think, despite inevitable cognitive biases, that I follow a good range of people across the great Prog-Trad divide and so get a relatively balanced picture of discussions and debates when they arise.
From what I have seen, views on the paper have conformed broadly to type: those who I would characterise as progressives leapt on the paper’s conclusions, using it as further evidence of the inequity of ability setting and arguing for the virtues of mixed ability teaching; those of a more traditionalist bent, broadly supportive of ability grouping across the curriculum, have questioned the use of the hyperbolic phrase ‘symbolic violence’, as well the validity and reliability of the research, which seems to have largely drawn on interviews with a small number of pupils about their feelings about being in bottom sets.
So far, so standard. (One really interesting and typically nuanced response came from Michael Fordham, who pointed out that a discussion of the merits or otherwise of different types of pupil groupings has to take into account differences between subject disciplines to be worthwhile and precise. I’ll return to this later.)
In recent months I have read and thought a lot about class in education. I feel that much of what I have read and listened to has been better at diagnosing the difficulties facing working class children and the problems created for them by the education system than in prescribing remedies and solutions. One recurrent theme in these discussions is that of ability setting, which is largely viewed similarly to the way Francis and her team view it: as pernicious for working class children, who disproportionately find themselves in the lower sets; reproducing and reinforcing social inequalities and breeding in the pupils themselves a sense of inferiority and hopelessness.
Many commentators have pointed out that pupils’ views are of course only part of the full picture we need when weighing the evidence on the effectiveness or otherwise of ability grouping, and extrapolating from such a small sample as the IOE researchers have raises big questions about validity and reliability. I agree with these criticisms.
And yet. In all of my experience over a decade and a half in schools and education more widely it is clear to see that bottom sets get the worst of, well, everything. Their target grades are low because of their KS2 results; they tend to get the weakest teachers, or supply teachers; behaviour tends to be a challenge, and so often teachers resign themselves to babysitting and crowd control at the expense of education; they have bowdlerised content in the name of ‘differentiation’ that leaves them with huge gaps in their knowledge.
I am no expert in the research evidence on setting. However, from what I have picked up from following discussions involving those more learned that me is that the patchy evidence we have shows setting is a net positive for those at the top end of the ability range and a net negative for those at the bottom end. However, the conclusions from such research are flimsy because studies rarely take into account the factors I mentioned above: quality of teachers, quality of instruction and quality of curriculum
And it is the last of those that I think is the key to a traditionalist argument against setting. I don’t think we can repeat enough the mantra of ‘behaviour and curriculum, behaviour and curriculum’ as being the two key foci for school leaders if they want to provide the best possible education to all the young people they serve, and in doing so shift the bell curve to the right and close attainment gaps. At the risk of being simplistic, assuming that schools can secure excellent behaviour in bottom sets because they establish a culture of high expectations and enforce them (and we know this can be done), then curriculum is where we need to have a laser like focus. And if we have pupils set by ability then we will almost inevitably have what Christine Counsell has termed ‘differentiation by curricular input’.
The argument I have been grappling with but struggling to articulate, which I think Christine’s phrase beautifully encapsulates, is that bottom sets will almost inevitably be given a dumbed-down curriculum of less, and less powerful, content. The great contribution educational traditionalism, allied to emerging understandings of cognitive science, has made in the past decade to improving what happens in schools is to argue for the centrality of knowledge, or the curricular ‘what’ rather than the pedagogical ‘how’.
So if we are serious that all pupils can and should learn a coherent body of knowledge in subject disciplines, then why do we need ability grouping in subjects in which content is organised cumulatively? We can deliver the content to all pupils. In my subject of history, that means ensuring all pupils know key dates and people from the past, as well as developing sophisticated understandings of substantive concepts such as ‘monarchy’ and ‘The Church’ through repeated exposure to such terms in multiple contexts across their years of study. (NB: I do not apply this argument to maths and MFL, in which content is organised differently and hierarchically, as pointed out by Michael Fordham, and which in my understanding makes mixed ability teaching very much harder – there may be other subjects for which this is true).
If we also follow Daisy Christodoulou’s insights about assessment in cumulative subjects, there is a further argument that ability grouping is not necessary. We can use ongoing, formative assessment to find out how well pupils retain the micro-bits of content we provide to them, before seeing differentiation come through more extended summative assessments, which will almost inevitably produce a bell-curve of attainment when judged comparatively.
In the case of both of these arguments – delivering the same curriculum to all pupils and using an assessment model that complements such an aim – they seem to me to negate the upsides of ability grouping (differentiation in easier, content can be targeted more carefully), while tackling the problems (pupils viewing being in bottom sets negatively, the tendency to dumb down).
I am painfully aware that these arguments may be naïve, or just downright wrong. There are probably many issues and themes I have not considered. There are also caveats, for example what about pupils who cannot read being in the same class as those who are preternaturally bright? I don’t have all the answers but I wanted to make the argument and invite further discussion. Please join in!