Working through my response to the SEND reforms and attempting to contribute to the policy conversation by pointing out some ‘illusions’, so that we can support schools to be truly inclusive.
An important, but also frustrating, publication came out last half term, reminding us that there is not a firm evidence base for many of the most commonly delivered SEN interventions Common SEN (Mis)Interventions – An Evidence Summary . From ‘mindfulness in schools’ to ‘Zones of Regulation’, the evidence is thin. It is a stark reminder about the opportunity cost of delivering something additional. Making the decision to take students away from their mainstream teacher and peers, is a weighty moral choice.
The ‘separation effect’ is much discussed in education, first coined by Webster et. al in their landmark book on Maximising the Impact of Teaching Assistants (2012), in it they highlight that “The ‘separation’ effect is a situation where pupils with the highest level of SEN receive a different educational diet to their peers, characterized by high levels of individual attention from a TA, but at the cost of less attention from the teacher, and being separated from the teacher, peers and the mainstream curriculum.”(Russell, Blatchford, and Webster (2012), p. 1).
Aside from the social effects of being seen to be different and the realities of being taken away from subject experts the possibilities for ‘catch up’ can be thwarted by structural curriculum issues. In many subjects the curriculum is hierarchical, Maths being a good example of this. The reality is that if pupils are streamed, for example to focus on securing their understanding of the foundations of number (which the evidence is still shaky on for lower attaining pupils by the way[7]) the chances of them ‘catching up’ with their peers are very slim. We might prevent the gap from widening, and that may be worth it, but we have to acknowledge the trade-off that we are making and the ceiling that these decisions may place on a child later in their school journey.
If there is a primary need, be that their number skills, reading or emotional regulation, that is seriously getting in the way of a student accessing an inclusive classroom then that student does need some targeted or specialist support. The SEND Reforms reflect that, and that is a good thing, but they don’t give us any magic answers about what this targeted support should be.
If we are taking students out of class, we must know that we are doing the right things with them once they are out. Whilst there is lots of great practice in many of our schools, we have not yet reached a consensus as to what makes really impactful targeted intervention.
I think there are two reasons for this. One is that evaluating the impact of interventions is seen as the domain of the SENDCo (and in reality, if done at all is often delegated to TAs). Therefore, it is not given the time and scrutiny it deserves. Secondly, it’s just a really hard thing to do.
Whilst I’m glad to be able to wave a paper to justify my dislike of fidget spinners. My problem with the Steplab approach- aside from the fact that many of the things they label as ‘interventions’ are not that- is the reductively empiricist framework. This approach seems to be dominant in education right now. The challenge is this: if we rely on randomised control trials and meta studies alone; when it comes to highly personalised interventions we will always conclude that nothing works. For those of us who’ve been working in this area for a long time and have seen the impact with our own eyes and ears we just know that isn’t true.
However, there are also many illusions of progress, many shiny packages of interventions that appear to solve an issue but actually rely very heavily on proxies, such as ‘they completed the worksheet’ and show poor far transfer into real life. We have to attend to this challenge and get better at reviewing impact.
Inspired by the Steplab paper, I recently did my own evidence sift on programmes I’ve seen used well in both primary and secondary for years. Packages like the Anger Gremlin or Talkabout for Teenagers and Zones of Regulation are evidence informed in their design. The Gremlin series draws heavily on CBT, Talkabout is informed by common social communication activities that Speech and Language therapists deliver and even the Steplab paper conceded that the underlying aims or teaching emotional awareness and self-regulation, particularly through interoception[8] that Zones of Regulation uses are legitimate.
However, these programmes, are very often delivered by TAs, without the right systems and structures in place in the wider school to help pupils generalise the skills. And so they very often fall short on their aims.
With reading intervention, though much debate still exists we have reached a tentative consensus in the profession and as such we have the systems to ensure rigorous implementation and impact. In Maths it feels like we’re further off. When it comes to Social Emotional and Mental Health interventions I’d say we’ve hardly even started.
Given the increasing need and lack of services available in our communities, I don’t think this is good enough. I strongly believe it is not a schools place to replace the role of the family or clinical services. It’s primary role is educational and as a result if pupils are successful in the classroom they will feel like they belong and will be engaged (in the previous blog in this series I discuss the importance of the teacher as the intervention ). Nonetheless, the reality is that many pupils barriers often sits far outside the conventional bounds of the classroom and given thresholds and waiting times for external services schools cannot sit idly by.
So how might we develop a rigorous evaluation framework, when we are looking at interventions that are highly personalised? I don’t have all the answers, but I do think we need to build the evidence base from the ground up. That means having very clear goals for our students and the interventions that we are providing for them and tracking their progress towards these carefully and frequently. We need to find ways of capturing both the small step progress in the intervention and the more generalised far transfer into wider school life and beyond.
In order to do this we need more consistency in assessment and identification of SEND needs. Currently, two schools five minutes down the road from each other and taking very similar intakes can have vastly different SEND registers due to the differences in the systems, tools and thresholds for identifying SEND. We need to build consensus about what good identification looks like as much as what good progress looks like.
As with so many things, I think many of the answers exist between us. Currently there is a renewed focus on foundational skills. This needs to extend beyond the basics of literacy, oracy, handwriting and numerical operations to thinking about infant and child development in the broadest sense. This is where coalitions and cross fertilisation come in. Something I’ll explore further in the final chapter in this series.
streamlinedsendco.com 