For slightly tenuous reasons I’ve recently taught the whole of year 12 to juggle. I’ve been quite into juggling since I was an undergraduate, joining the juggling society then, later, as a postgraduate, starting my own (the image above shows a ‘big toss up’ – something that often marks the end of a juggling festival or convention – quite an experience!). I’ve started juggling clubs at schools, and taught a large number of various students and staff over the years. I think I’ve got pretty good at it, and my process for doing so has been gradually refined. Teaching a whole year group in six sessions across three days was really enjoyable, and a wonderful break from the exam revision that I feel like I’ve done endlessly this term. Juggling is a great example of how and why direct instruction is such a powerful tool, and so much more effective than getting students to figure it out themselves – see this post from Umes Shrestha for more on this. It really highlighted to me the process of modelling and live feedback, and made me reflect on how well I do or don’t do this in my regular teaching. It’s also made me reflect on the challenges in taking principles of teaching that we might observe or read about in other contexts, and translating to one’s own domain.
Teaching someone to juggle a basic three-ball cascade can be broken down into small, progressive steps. I start with just one ball and demonstrate first correct posture and hand position, the flight of the ball that they should be aiming for, and then show what a correct throw looks like. Then we all do it together, before letting them continue to practice on their own. I do, we do, you do. I’ve done this enough times to know pretty much exactly where they might go wrong and precisely what they need to do to fix this. Once they’re practicing then I can easily wander around giving specific, corrective feedback that improves what they are doing (“drop your shoulders” “move your hand like this so you throw across your body not forwards”). Once I deem that enough of them are comfortably performing this step correctly I pause them all – ensuring everyone is attending to what I’m doing, removing extraneous load by getting them to put their balls down while doing so – and then we do the next step. Same process – I do, we do, you do – with lots of live feedback. It works. I can’t claim that I can teach *anybody* to juggle but I’d say my success rate is pretty good. I think it’s a strong example of how to model a new skill, and really made more explicit in my own mind just what effective modelling looks like. I left these sessions feeling very satisfied and wondered why I don’t experience similar satisfaction in my normal teaching on such a regular basis. I’ve read a lot about modelling in teaching in various books, articles, blogs, as well as discussing it with colleagues in CPD sessions and department meetings, and online via Twitter. I think it’s something that is much easier to describe than do, and want to consider why it can be such a tricky thing to get right.
Prior knowledge
Typically in a group I’m teaching to juggle there might be one person who thinks they can already juggle. A quick demo usually reveals that while they might have the basic gist, there’s a lot to work on in terms of the fundamentals – posture, hand position, throw trajectory, timing etc. So it’s normally very helpful for them to go ‘back to basics’ and start from scratch with everyone else. If they really don’t need to do this, then I have some simple problems to give them to work on to move their juggling to the next level. Stretch and challenge for is easy! Having everyone else pretty much in the same place in terms of prior knowledge makes the modelling so much easier because you’re building their schemas from the ground up. This is far less common in the classroom where students bring so much of their own context and knowledge to the lesson and everyone is often at very different starting points. Even in my subject, Psychology, which most have never learned before, there are still huge differences in what each student brings to class. Some of this is their own general world knowledge which is quite important for understanding lots of psychology, some of it their domain specific knowledge from other subjects (for example if they already know a lot about the nervous system from studying biology) and some of it is their procedural knowledge (eg how to construct an effective piece of writing). Dealing with such a wide range of starting points means knowing how to pitch your modelling at the right level is tricky. Providing suitable scaffolding whilst at the same time giving appropriate stretch to those with most prior knowledge is probably one of the most difficult problems in teaching.
Task complexity
Although it may not seem like it to a complete beginner, the basic juggling pattern of a three ball cascade is actually quite simple. The mechanics of it are fairly easy to explain, although the gulf between knowing and doing can be quite significant for some! Teaching students how to formulate a good piece of writing – evaluating a psychological theory, say – is a much more complex beast. There are multiple moving parts and learning to handle all of the elements at once is difficult. Students need strong content knowledge, they need to be able to analyse precisely the demands of the task in front of them (which often involves unpicking and then using elements from a novel scenario in a question stem), and they need to know how to structure their writing both at a global level – what does an essay look like – and at the level of the sentence too. Unlike juggling, it’s not a linear process either, you can’t learn one thing and then add something new in a logical progression. At some point they have to be able to do it all at once – they need to have all three metaphorical balls in the air at the same time (ironically, something which you typically *don’t* need to do when starting out with actual juggling). In Cognitive Load Theory terms, element interactivity is high, massively increasing intrinsic load (Lovell, 2020)
Since reading Daisy Christodoulou’s Making Good Progress I’ve been much more conscious of the need to practice each of these things separately, giving plenty of time for modelling and practice of each one on its own, rather than just trying to set them off writing an essay from scratch. But even taking that approach, at some point they’ve got to learn to combine it altogether and getting that right in class is hard. Knowing at what point they have learned well enough each of the individual components to be confident enough to bring it all together coherently, particularly given the diversity in prior knowledge mentioned above. Balancing the time needed to ensure a good success rate – crucial for maintaining motivation and showing them that this is achievable – against the ever-present pressure of the need to deliver all of the content before we get to the exams.
Practice & Feedback
When someone’s juggling isn’t working I’m expert enough to be able to diagnose the problem fairly easily and offer very precise, specific feedback. “Your timing is off because your second throw is going much lower than your first; make them the same height.” Students can adjust that single thing and pretty much immediately see the impact. When a whole class of students are engaging in a piece of writing for themselves, giving that feedback is much more difficult and the impact can be less tangible. Sometimes it’s fairly straightforward – “add a concrete example to illustrate that concept” or “give a specific study here which supports this point” but often it’s a lot more nebulous, and it’s certainly a lot more variable. In a class of fifteen A level students it’s quite feasible, likely even, that there are at least ten different things on which I might need to give specific feedback. Quality of writing often comes down to very subtle choices around language use such as vocabulary or syntax, and explaining exactly why this word works better than that one, or why that paragraph structure could be more effective if they reordered those sections is tricky. Students may have to take more of a leap of faith here – trusting in you as the subject expert that this makes their writing better even if they don’t really see it for themselves; the impact is far less tangible because they don’t see the improvement in the same way. Sometimes you can only really appreciate the difference that something makes from the perspective of an expert, and understanding that students don’t have this privileged viewpoint is important.
The whole process is very demanding and places a lot of load on the teacher. You’ve got to quickly read what they’ve written, diagnose the most significant issue (no point giving them several things to think about if we want to avoid cognitive overload), formulate highly precise and specific feedback and then check they’ve understood your feedback to be able to enact it. All the while somehow keeping your eye on what is going on around the rest of the class, monitoring the time left in the lesson, and probably thinking about what you’re going to do next. No wonder this is so hard to do! Covid restrictions have made this much more challenging, for sure. Wandering around the classroom looking over their shoulder has been literally impossible for many months, and even now requires the tedious putting on of masks, and then having slightly awkward muffled conversations with students.
Motivation
When I’m teaching people to juggle most are usually interested enough to have a go. Juggling looks impressive (and I have a range of nice tricks up my sleeve to emphasise this if they think the basics are a little pedestrian) and most people want to try for themselves. The first steps are pretty straightforward and almost everyone gets to experience a bit of success to keep them going; motivation to continue is not really an issue. But once they get further down the line, it starts getting hard. Typically it’s the point at which they are trying to progress from a cycle of three throws and catches (known as a ‘flash’ for juggling aficionados) to four and beyond. At this point the type of feedback I give often changes. It’s less about the basic mechanics of their technique, and much more about just practicing. There are some tips I can offer to help with some specific issues, some just need to go back a step to gain better fluency before moving forward again, but for most people they’ve ‘got it’ but just need to keep doing it until they become more fluent. And at this point some will give up and some will keep going. The key thing here is that juggling doesn’t actually matter – there is nothing riding on whether they learn it or not. If they choose to continue it’s because of some form of intrinsic motivation, they are just enjoying the challenge of learning something. There’s no material gain for learning to juggle, no extrinsic motivation. Contrast that with learning how to write an exam-style essay, ensuring you know how to jump through the hoops of mark scheme rubrics, and you can see the difference.
This also means my feedback on juggling is much more authentic because it doesn’t matter to me how quickly or slowly they make progress. I can tell someone that what they really need to do is spend a lot of time practicing their throws because it makes no difference to me, or them, whether it takes them a day, a week or a month to learn (although obviously most would prefer quicker progress). Again, the difference with academic study is stark. The very real pressure of knowing students have a concrete end-goal, that they *have* to be ready by the time they take their exams, significantly impacts this feedback process. I literally can’t give them the time some of them need and the feedback I really want to give would be deemed unhelpful. Students often ask “what do I need to do to improve” and it’s a question we rightly encourage. But sometimes the answer is not what they want to hear because it’s not a quick fix, and sometimes they don’t want to engage because it’s hard. What I honestly think some of my students genuinely need to do is take a year or two off from academic study and just read lots of books or articles. They simply don’t have enough exposure to the variety of language and challenging texts that will help shape their own writing. They need to grapple with academic language and difficult concepts, expanding their vocabulary and their repertoire of writing techniques. Yes, of course, we can teach some of this, and provide scaffolds like word banks, sentence starters, writing frames or model paragraphs. But there’s a limit to how much time you can devote to this in a two year course when exam specifications are jam-packed full of content. So, obviously, this isn’t what we say to students. We offer more concrete suggestions, and try to give them as much of that scaffolding and support as we can to help them make progress. We try to give them as much encouragement as possible to keep them motivated, even though we may think, privately, that they aren’t going to make the progress that they are hoping for.
So what are the key takeaways here? Obviously, modelling and feedback are hugely useful techniques that form the bread and butter of much good teaching. But learning to do them well is hard.
- It takes significant time and practice to get this right, and will need persistence. Time spent with colleagues chunking and sequencing tasks, planning and scripting instructions, and practicing delivering in a ‘safe’ space are all hugely beneficial uses of departmental time.
- It requires real depth of subject knowledge, and proper understanding of the gap between your own expertise and that of the novices in your classroom. Watching someone modelling something in a lesson outside your specialism is instructive in terms of the process and mechanisms, but will need careful thought about the intricacies of your own subject knowledge to be able to apply it effectively in your own lessons.
- It needs acknowledging that modelling complex tasks will, at some point, feel messy and almost beyond the point of control. There will be, metaphorically, balls flying everywhere and plenty of them hitting the floor.
- It needs grounding in a strong classroom culture, relationships rooted in mutual respect and trust, and built on clear routines and expectations. There’s no way you can give live feedback to a whole range of learners at the same time if you haven’t got a handle on this.
- Finally, it needs recognizing that good modelling may well get everyone further along than when they started, but it won’t necessarily get everyone all the way, at least not in the timeframe we are often working to.
To paraphrase/bastardise Dylan Wiliam’s famous words, modelling might work everywhere, but will work better in some contexts than others.
References
Christodoulou, D (2017). Making Good Progress?: The future of Assessment for Learning. Oxford University Press.
Lovell, O (2020). Sweller’s Cognitive Load Theory in Action. John Catt Educational
A Psychology Teacher writes 