A reflection on grouping pupils…


We know now that most pupils benefit more from mixed ability grouping, although as Francis et al (2017) have suggested, a better term for this is: ‘mixed attainment grouping,’ because of all the connotations regarding ‘ability’ within educational discourse.

It is true that higher attaining pupils achieve slightly better under setting, at least academically, but the overall affect on pupils and schools is more negative than positive. As the EEF Toolkit concludes, ‘setting or streaming is not an effective way to raise attainment for most pupils (2018).

In this vein, I would like to share how I’ve developed grouping in my setting with good results, both in terms of hard data and the soft data of classroom climate, pupil well-being and attitudes to learning for both pupils and parents. Without doubt, there are always tweaks and changes that may suit one cohort better than another, and I’m sure, there is still yet more to learn here, but this is where I’m up to.

No doubt for some, what I describe here will be utterly obvious, much of this is simply about good teaching regardless of grouping; however, the focus here is on what works best for mixed attainment grouping as a pedagogy, not least because this approach as been proven to be important in terms of social justice, with the understanding that forms of segregation exasperate social inequities in most circumstances.

Around ten years ago we started questioning ‘ability’ grouping (as we referred to it then)  for all the reasons it has been found to be detrimental for middle and lower attaining children (Francis et al 2017).

Like this, we found that most pupils stayed in the same groups throughout school, never gaining access to the tasks set for higher ability pupils, and never having the opportunity to catch up with them either. We also found that the lower attaining groups were almost always populated by more less affluent pupils, while the high attaining groups were filled with mainly pupils from middle class backgrounds.

In addition, ability grouping created a self-fulfilling prophesy for pupils and was generally divisive for pupils and parents, both social and emotionally. It certainly created unnecessary angst for many parents and pupils, creating a labelling culture which all too often defined expectations in all quarters.

Over time, we woke up to these issues and we moved to mixed attainment grouping, differentiating learning tasks rather than pupils and offering open ended choice of tasks for all pupils.

(Recently, for a lecture I gave at Goldsmith’s on grouping and ground-up policy change, I reflected on this journey with two esteemed colleagues – it’s a ten minute watch here)

Since then and over time, I have recognised key ways that have enhanced mixed attainment grouping as an effective strategy.  Underpinning these factors is the fundamental principle (and a seemingly obvious one) that the aim is always to meet all pupils’ learning needs within lessons through highly interactive, responsive teaching. This might mean scrapping what’s been planned, shelving carefully chosen learning tasks, or shifting seating arrangements on the spot. I say this, because when declaring a commitment to a methodology, rigid dogma can easily follow. It should not be ‘mixed attainment grouping come what may’; this could be just as detrimental for some pupils as rigid attainment grouping.  As with all things in education, effective learning is what matters, methodologies and ideologies should not supersede what’s best for pupils’ learning. Teachers need to be ideologists at their desks, but pragmatists using evidence of what works at the chalk face.

These are some of the key strategies that have helped make mixed grouping work well:

Planning – Plan progressive learning tasks that move pupils from high scaffolding to independent thinking in relation to the topic. This means getting underneath the topic and understanding the knowledge goals really clearly, as well as the cognitive process of embedding concepts then applying and developing them. This is  no different from what would be required for fixed grouping, except here, children are usually sat alongside other children who might be at a different stages along the progression. It might also be that a wider range of progression is needed to accommodate the variety of attainment; however, it is my experience that gaps narrow more easily over time with this approach.

Direct instruction & cognitive load etc – This is the same for any type of grouping, but remains a key teaching point for me. When introducing new, or returning to unfamiliar content, direct instruction works best. This means breaking down material, appreciating the capacity for working memory to process brand new information and cutting down extraneous information in presentations and within the classroom. This helps all learners. Ensuring the knowledge goals are explicit and clear to learners is vital – what does success look like? Don’t leave them guessing or wondering here. Providing worked examples to support pupils processing new material is vital, as is recognising when to draw back and shift pupils towards more independence.

However, this recognises that engagement should not be sought through objects or images that distract the learner from what they need to learn. As Willingham (2009) asserts, ‘memory is the residue of thought,’ so that trying to maximise attentional focus in pupils with things that ordinarily attract their attention during their leisure time simply replaces their thinking with these other things. Like this, we have to take care not to mistake behaviour with cognition, but understand how they relate to one another.

Consequently, a teacher’s first call is not ‘making lessons fun.’ It is however, their duty to make lessons meaningful and challenging enough for all pupils to feel the rewards of successful thinking. Once pupils start to get a kick out of ‘getting it,’ and we know they do, when we hear kids shout out loud gleefully, ‘now I get it!,’ we know that children are enjoying learning for the sake of it. Successful thinking feels good, build on this as a teacher and you’re really on to something.  We do not need external bells and whistles here – save these for other times in school.

Assessment for learning and all thatMixed grouping has the potential to require more of AfL technique if there is a wider range to assess, although good practice is good practice wherever.

Teachers will need to act like air traffic controllers, scanning the skies for pupils coming into land, the ‘tipping point’ for successful thinking. The point is, we need to know when pupils need more or to move on. As Willingham (2009) notes, the human brain prefers not to think, but to use memory if it can; therefore, if tasks don’t feed the brain with the incentive to think, if they are just too hard or just too easy, then we lose pupils.

Assessment on the hoof is the key to this and putting all those AfL techniques, like mini white boards and hinge questions to good use will support teachers knowing when some pupils need to move on from direct instruction and worked examples to more independent learning. Good planning should ensure that this is prepared and resourced for so that pupils can go somewhere with new learning, refining their own mental schemata. ‘Tick, tick, tick, now go read a book’, is OK very occasionally (we’re all human) but not a good basis for providing pupils with a love of learning for itself. Also – if this is the default end of the lesson for some pupils every time, what are they learning?  What are you teaching them about themselves as learners?

The starting line – Last year in Year 2  we trialled a change in how we provided learning tasks for mixed attainment groups.

In the past, we provided differentiated tasks, named, ‘MUST, SHOULD and COULD, and sometimes even MIGHT, where pupils assessed themselves (sometimes with guidance too) and started where they thought they needed to. Children sat in mixed groups and worked from various starting points depending on the subject and topic.

However, we found that with young children sometimes this meant some pupils were jumping into more abstract or independent learning too soon, mistakenly assuming that the knowledge for this was embedded. This revealed itself as gaps in knowledge during more independent, open tasks.

This situation suggested that fundamental factual and procedural knowledge was not properly embedded into long term memory and had not become automatic.  This is built on the idea that expertise come about when access to knowledge in the long-term memory is automatic, without ‘thinking’, rather like driving a car after a while.  When pupils have this automaticity, they have much more working memory capacity to problem solve successfully and organise mental schemata further. Remember, the brain prefers using memory rather than thinking so when learners have more to access in memory – their processing is optimised. As Willingham (2009) points out, expert problem solvers may look like they are skilfully calculating, when in actuality they are accessing long- term memory while using their working memory to make links and connections into new knowledge.

In this way, we decided to trial these young learners all starting in the same place, appreciating that young learners will have fairly rudimentary mental schemata and are not experts with large bodies of background knowledge yet. We provided worked examples, leading to less and less scaffolded tasks, but all started off with the same task. Of course, some children needed to stay with the worked example stage for longer before moving on.

This meant that while some pupils spent more time working on the first tasks, perhaps needing more support and direct instruction, some pupils were using these tasks as quick retrieval practice which, as Kirpicke & Grimaldi (2012) assert, that far from being simple recall of old learning, acts to embed and extend knowledge further.

Through this leveling of the starting line, we found that overall pupils’ learning stuck much more than when some pupils had the opportunity to miss the initial first ‘easy’ activities. Although – again, flexibility is everything, sometimes it was abundantly clear pupils could skip and we let them, but the general, ‘all start here,’ approach was better in that it appeared to solidify connections – making learning more automatic for many.

In addition, we also found that this created a more equable working environment between pupils; often lower attaining pupils had the chance to ask their higher attaining neighbours about the initial task as they began together; they could absorb strategies and learning behaviour from them also, which I will elaborate on next.

Seating arrangements – These need to be fluid and flexible. I found that there were pupils in the class who acted as  effective role models for successful learning. These pupils engaged positively with the learning procedures. They were self-motivated to use the feedback and the class systems to enhance their learning and actively sort out help when they couldn’t work things out for themselves.  Some pupils just aren’t ready for this stage yet, some of this might be to do with the development of their executive functions, the cognitive control of behaviour which enables pupils to select and successfully monitor what they are doing towards their learning goal. It is almost always not laziness!

In a class of thirty children there will be a huge range of cognitive, social and emotional development.  We know that working memory increases with age in children, so that some children will have a surprisingly limited cognitive capacity, while others a surprisingly voluminous one.  With a limited capacity to process new information, some children will find it hard to hold everything they need to do in mind. For example, by the time some children pick up their book, find their seat and a pencil, the instructions for the current learning task have gone, literally gone! Seating children strategically next to children who are able to process everything, can help to remind them about what’s going on, and in a second or two, get them back on track.

In more extreme cases of course, these children need tasks broken down further to accommodate their capacity and perhaps a bit more ‘preparation for learning’ built around them.

However, human beings tend to watch and mimic the successful behaviour of others (this is to do with evolution and survival). For example, when I go swimming, I keep an eye on the best front crawlers in the fast lane and try to do what they do, I don’t copy the Jurassic strokes laboured over in the slow lane.

Like this, those children who tune into the processes in class that make learning successful, like checking the board for examples, asking for help, reading the success criteria against what they are doing, these act as role models for others who have yet to do this. This is not to say that teachers should not be intervening when children are struggling, and this is not to say that we use the higher attaining children as glorified teaching assistants, this is not what I mean. We have a duty to all pupils, including those further ahead (there is an excellent ‘dos and don’ts’ list put together through research by UCL here which details this). However, if we know that humans learn from each other simply by watching and modelling themselves, we should use it.

Seating these ‘effective learners,’ in the vicinity of those still struggling to adopt effective learning procedures is really effective for the whole class dynamic, as is explicitly teaching these procedures to some children too. Still, I found that the higher attaining children don’t lose out by having their time taken up ‘helping’  others because they are already tuned into prioritising successful learning, so they tend to steam ahead while the child needing a little ‘va va voom’ next door, watch and learn.  There are also times when pupils further ahead really benefit from explaining concepts to others – after all to ‘teach it,’ is the final point of mastery.

As said, this all needs monitoring and fine tuning, if a child gets needy and disturbs another, I move them, often back to more direct instruction and guided learning.

Flexibility in grouping In understanding the needs of all learners, the differing cognitive capacities, we also understand that there are times when we need to pull groups of similar attaining pupils together in order to maximise the overall benefit of mixed attainment grouping.

In understanding that some children need multiple attempts at processing new knowledge, because they have insufficient background knowledge or limited cognitive capacity, there are plenty of times when we take a group and pre-teach or over-learn new content. This is where we depart from the mixed attainment grouping and I think we should. Here we give pupils who need it a leg up so that they can then go back and reap the benefits of mixed attainment grouping. The point is that these groups are temporary, not set or referred to with a label.

Capitalising on forgetting & remembering – We have also come to recognise that retrieval is a powerful learning tool in its own right and not simply for assessment (excellent article in TES here about this). Research by Karpicke and Grimaldi, challenges the idea that retrieving and reconstructing knowledge is a ‘neutral process,’ but rather found that, ‘every time learners retrieve knowledge, that knowledge is altered, and the ability to reconstruct that knowledge again in the future is enhanced,’(2012:404).

There was a fashion not so long ago for teachers to be seen to challenge pupils continually by moving them on to new knowledge, this meant pupils were not going back enough, resulting in shallow and transient learning.

With this in mind, retrieval practice has become a feature in my curriculum planning while also being recognised as an invaluable intervention to support some pupils to catch up within a mixed attainment setting. This kind of practice is not repetitive chanting or learning by rote, but rather activities which involve the active retrieval of knowledge which can be in the form of low-stakes quizzes, series of oral questions, quick multiple choice sets, card matching, true or false checks – there’s lots of ways and I’ve found that children always enjoy this aspect of learning – as long as we keep it low stakes and allow pupils to feel in charge of it, for example, by them marking/assessing their own.

To finish, this is simply a reflection on practice changes I’ve made. As said, it’s useful to me and I hope might be useful for others who might be trying a mixed attainment approach. Plus, it’s my belief that  schools should be places where social justice motivates us to examine what we do.

Lastly, it is worth remembering that school is not all about this. Schools, and especially primary schools, are about social and emotional learning and fun as much as the academic.  If memory is what the brain prefers best, then we need all sorts of good ones… not just the academic, semantic kind. So, here’s to messy, goal free mucking around in school too!


Francis, B., Archer, L.,  Hodgen, J. ,Pepper, D., Taylor,B. & Mary-Claire Travers, M. (2017) Exploring the relative lack of impact of research on ‘ability grouping’ in England: a discourse analytic account, Cambridge Journal of Education, 47:1, 1-17.

Karpicke, J.D. & Grimaldi, J.P. (2012). Retrieval-Based Learning: A Perspective for Enhancing Meaningful Learning. Educ Psychol Review. (2012) 24:401-418.

Willingham,D.T. (2009)  Why don’t students like school? Jossey-Bass.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s