The debate over grouping students by attainment vs mixed-ability is a long-standing one. In my department, we transitioned away from mixed-attainment, towards setting all year groups. Whilst initially a controversial decision, a clear rationale and open dialogue between staff ultimately secured my decision. Here, I will explore how we implemented it, as well as a critical review of the current research.
Before implementing any new initiative, we should always consult the research. However, the evidence in this area is, well, awful. Many studies are difficult to apply to varying contexts, often completed on small cohorts, or are based in America. Meaning that yes, the headlines say one thing, but in reality, it is difficult to apply in our classrooms.
Looking specifically at the EEF, their guidance reports that setting has an average impact of zero months and specifically cautions that whilst yes, it can boost higher attainment students, it negatively impacts lower attainers. This is important because, if set by mixed-attainment, we could entrench the very issues we are working so hard to erase.
Looking more into EEF’s review, I had some specific issues with the research quoted. For example, two of the most heavily weighted students concern only reading interventions, not typical classroom teaching. Several of the historical studies, such as those by Green (1936) and Moorhouse (1964), again their focus is on the teaching of reading. Another key reference, the Ingram study, is based on a small sample of 63 unassessed primary school students in the US, again difficult to apply in my own school.
It is only with the more recent Roy (2018) paper that the EEF’s evidence base offers research directly relevant to our context. This study provides a valuable overview of setting in the UK where only one third of schools inspected by Ofsted use mixed-attainment setting. This is similar to the completely non-scientific survey I conducted on Twitter, where the majority of those who responded said they set by attainment.
Roy’s research highlights a significant paradox: the prevalence of setting in schools stands in stark contrast to the body of research that questions its effectiveness. So, why is nobody trusting the research that is being produced?
Obviously, one could point to the quality of the research but I actually feel that it is the lack of concrete quantitative data even giving an illusion of a correlation between the setting and outcomes. Given the sheer amount of data we produce in schools in the UK, I find it bewildering how much of the research in this area is purely qualitative and based off the ‘feelings‘ of teachers and students rather than any concrete conclusions.
Therefore, many look at the research, see it doesn’t make that much of a difference, and continue, usually, with what they have always done. Below, I write about how I set by attainment and also provide my reasoning as to why I do it. Is this concrete quantitative data? No. However, many are aware of the context of my own school so therefore, I feel there are some conclusions that can be made for others to follow.
How I Set Using Attainment
All year groups sit an assessment just before Christmas, from which the entire cohort is ranked by their score in the test, and new classes are created at the beginning of the Spring term. I strategically create smaller classes for out lower-attaining students to provide more support, whilst our higher-attaining sets are larger. Probably the most controversial aspect of this these sets are purely based on attainment and nothing else. We do not consider student behaviour when setting the students into different classes ensuring the focus is purely academic.
The reason why I do not take into account behaviour is because, adding one student into a class where their attainment is not close to their peers, means they may often be much further ahead then others in the same class, which means they will complete work and sit there able to distract others, or they may be significantly far behind their peers and therefore struggle much more than necessary.
However, these are not set in stone for the whole year. A second assessment at Easter allows us to make small adjustments and the small number of students who have progressed are moved accordingly. This ensures that the sets remain accurate. This is then repeated again in the summer ready for the sets at the start of the next academic year. I wrote about how our assessment is organised in a previous post, which you can read here.
What is my rationale?
When implementing anything, I always provide my rationale for why I am doing this to senior leaders and teaching staff. So, despite attainment setting being against the research, why do I do it?
A core part of our department is that all students access the same ambitious centralised curriculum. The key difference is not in what is being taught, but in how. This is far more manageable than in a mixed-attainment setting, where, as mentioned previously, a single task can leave high-attainers disengaged and low-attainers overwhelmed. Setting allows us to pitch learning at a level that appropriately challenges the majority of students in front of us.
For example, I teach a middle set Year 9, they complete the exact same work as students in Set 1, but it may take us a little longer with more difficult topics, I may have to model more and I may need to give many more examples, or take two lessons to teach a concept rather than one. But again, they complete the exact same work because doing anything different would offering students a different experience based purely on their academic achievement. This is unfair.
This approach also reduces the cognitive load on the teachers. In a classroom with a vast range or prior attainment, a teacher is often forced to manage multiple abilities at once. In my observations of mixed-ability grouping, this often dilutes the quality of instruction where teachers can often simplify a concept so much that higher-attaining students are not being pushed into hard thinking.
Additionally, when I have observed mixed-attainment classes, I noticed that sometimes teachers sit students who are high attaining next to students who are low attaining. This often leads to high attaining students to help the students next to them. When this happens, it is lovely to see, but I don’t like it.
All students should have access to an expert teacher to explain to them what they don’t understand. If we have offloaded this to high-attaining students, we are doing both a disservice. The high-attainer should have access to work that occupies all of their time to complete, not finished really quickly. Similarly, low-attainers should have access to good modelling and explanation and the help from the person next to them often results in them just giving them the answers rather than how to get the answer. This then makes it more difficult for the teacher to actually assess the classroom as it seems as if more students understand, when in reality, they do not.
We also actively work against the ‘bottom set’ self-fulfilling prophecy by ensuring there is high mobility between our sets. A student’s placement is never permanent. Instead, it’s regularly reviewed based on their progress and performance. When students demonstrate improvement, they are moved up, which creates a positive and aspirational learning environment. This constant potential for movement fosters a healthy sense of competition and powerfully reinforces the direct link between a student’s effort and their academic outcome, motivating them to believe that their hard work will be recognised and rewarded.
Finally, attainment setting also allows for strategic deployment of teaching staff. Every teacher has differing skills. Some excel at building confidence and securing foundational concepts, while others are adept at stretching the most able students. By matching these specific teaching strengths to the particular needs of each group, we can ensure every student receives the most effective and impactful instruction. This tailored approach guarantees that all learners, regardless of their set, are supported and challenged by an expert who is best placed to maximise their outcomes.
The decision to set by attainment required navigating a contradictory research landscape. The limitations of the poor research on this area forced me to create a system that worked well for our staff and students. Sometimes, when looking at research, we can often focus on the headlines or summaries when in reality, they would not work in another school with a differing context.
As I have mentioned here, this is clearly prevalent with setting, but let’s be honest, this is common in many areas in education.
inquestion. 