The Classroom
How we integrate AVM into your school
At the core of AVM, i.e. the courses that we offer in afterschool programs of 8 weeks, is in-class, live, in-person instruction. On our website, we speak a fair amount about AVM in one of its aspects; a bespoke application we created to structure the learning process.
But the software, for us, is only a catalyst, while the classroom is the key to our program. What the software enables are interactions that we couldn’t implement using any standard learning management tools, so it is crucial to our method. But the classroom remains the central focus.
That’s where the learning dynamic between students emerges. In the abbreviated, marketing focus of a website, this is hard to get into the nitty gritty. But we take it very seriously, which is why this post was written. It will be linked from the related pages on the site; perhaps that’s how you got here. If so, and this is of interest, please consider subscribing.
You will discover on our website three things: that we teach using a paired learning model; that we assemble pairs into teams of three pairs, for six students; and typically our classrooms have 12 students – in other words, two teams, six pairs.
In this post, I just want to focus on the goals of this structure, and how it works in practice in the classroom. First an overview:
Pairing
Putting two people together to work out things by consensus is not a new idea. Our background is tightly connected to teaching people to code, which you can read about in another article, and pairing is one of the methods we adopt from the world of software.
Six Person Teams
The magic of a six-person team arises because each pair it contains is learning not only to advance their own abilities, but also to advance those of their teammates in the other two pairs. And each student relies on the diligence of the other pairs to communicate their understanding of new skills and techniques to them.
Twelve Student Class Size
Building and implementing tech bootcamps, as we have done over the past quarter century, the dynamics than arise in groups of people vying to learn everything they can in a short amount of time has been instructive and sometimes cautionary.
This experience has led us to favour groups of 12 as the ideal number for accelerated learning, balancing good group dynamics with equitable attention paid to keeping everyone on a solid learning track.
As an aside, we have experimented with classes of 18: that may be a feature of our classes in the future.
The Details
What follows digs a little deeper. If you’re interested, read on.
Staying on the Class Size theme: a 20-60-20 rule came across my radar around 2013. At that time I was working freelance to build other companies’ bootcamps.
There, large class sizes were common as a path to maximizing profit. Twenty percent of the class would cluster in a self-perceived “fast track” or “advanced” group; twenty percent would form the “slow down, I want to get what I’m paying (a lot) for” group. And the 60 percent in the middle would get on with learning, help each other, and provide the useful, positive feedback.
The general rule was, the two groups at either end of the spectrum mostly raised issues about the other. This wasn’t exactly a matter of animosity, but it could get edgy. Once people find their clan, as these groups did, they tended to use one another as a baseline for what a reasonable rate to roll out new material was. If you find yourself always sitting with, talking with, chatting with, and having lunch with the people who “totally get” your perception of a situation, the need to see that in the larger context diminishes. When you’re paying $20k, and wish the class would move a little quicker, dissatisfaction can arise – understandably.
And of course, the same is true of those who might cope differently with new concepts. After all, this is supposed to be stuff they don’t know; that’s why they signed up and paid the big money. Their mantra: Slow down, explain this to me.
The high cost of bootcamps like the ones we built, or that I helped develop for others justifies an expectation of value for money. Both the twenty percent groups shared this perception, but their standards for evaluation were different.
And of course, there are nuances; the perception that “we’ve learned all we need to here”, of the “fast learners” was ofter facile; the supposedly ‘slow” group often raised issues around extended implications of a feature of the subject. Why was this the case? Isnt’ there an edge-case where this problem would arise? Usually, this expanded outside the scope that we, as curriculum developers, had allotted to the subject.
Controlling the 20-60-20 Dynamic
In coding, the canonical concepts related to control structures (loops, binary decisions), data (variables, sets, arrays and hashes), and closures, parameters, return values, classes and data-typing, lend themselves to those who seize on abstractions and then apply them.
If you excel at abstraction, you’re in the 20 percent who become impatient.
What our courses are focused on is quite different: difficult-to-acquire skills based on complex, confusing software with lots of power, but less clarity. Teaching this kind of technical subject, where what is being learned usually doesn’t reflect crystalline abstractions, but rather more the arbitrary decisions of software programmers and user interface designers, favours a very different kind of mind. But a huge amount of the world of technology depends on exactly this.
But the best way to deal with a problem is to not have it. Twenty percent of 35 people is seven people; but for twelve people, it’s 2 (plus a fraction). Two people together don’t form a group, they form a pair. Seven people is a faction.
So since game technology as we teach it omits the abstraction-heavy coding component (another post is due on this), what we are left with is decidedly planted more firmly in the domain of the less abstract, more pragmatic, with a dollop of arbitrariness built in to contend with.
Would 20-60-20 be a problem in a learning environment focused on this particular set of characteristics? I don’t know, I haven’t got experience teaching 35 people this. But I’m not aiming to find out.
Either way, we structure our classes into 12 students, which is why the Team and Pairing aspects are important.
Pairs
In classic coding pairing, two people work to write a new code component. Typically, they sit together, but it can be the case that pairing happens remotely. Of course, remote work necessitates each person have their own computer, and in coding, this is how it’s done in person as well.
But our program is based on a shared workstation, where the two members of a pair have to work and sit together. That means sharing the keyboard and mouse. Not in the sense of one person having the mouse, the other typing, though anything’s possible.
Instead, a dynamic of passing control back and forth needs to be established. The AVM model makes each student and their partner a mutually-supporting unit.
This is important, because the dynamic of working through a Challenge isn’t just about learning a skill embedded in the Challenge. What is key is that while each pair sees the same succession of videos that demonstrate different aspects of the Challenge, each pair has a different Focus Statement. This directs them to select just one of the multiple videos as the best fit to the focus they’re assigned.
When each pair finds their solution, it’s the completion of a journey of their own; now, the members of that pair, both equipped with a new technique or model for understanding the subject, will split up to join and instruct one of the other two pairs in their team in what they have learned: and this will be reciprocated.
Each student is thus learning not only to advance their own abilities, but also to advance those of their other teammates. And each student relies on the diligence of the other pairs to communicate that understanding to them.
From Two to Six
The basic dynamic of the pair, which allows for focus to be on the screen enough to be effective, but demands engagement with your partner, is the mathematical basis for the group of six.
As we’ve mentioned, the two members of a pair, now possessing a newly acquired skill, ready to be imparted to others, predetermines the size of the team. The dynamic of course is that each pair’s partners go to the one of the other two pairs to instruct; so the movement of new knowledge and new ideas is rich and important in the completion of each new Challenge.
There’s a little more to it than this, though. Often, Challenges will involve a sequence of techniques that together add up to something more meaningful than the individual skills. For instance, to make an arm that moves realistically, with joints at the shoulder and the elbow, requres three separate techniques.
First, the components that make up the components of the limb need to be added in 3D space.
Next, the two components need to be associated in such a way that the end joint of the upper arm exactly coincides with the position of the top of the forearm.
Finally, the relationship that causes the forearm to move with the upper arm, but lets the forearm, while remaining attached to the upper arm, to move freely.
In many Challenges, each pair in a team will have a challenge that directs them to only one of the three different techniques like the ones above. So more than just instructing each other, the pairs in a team have to make sense of what they’ve learned in the context of what the other two pairs in their team have discovered.
In this case, the general Challenge might be, “create an arm that moves naturally”. But the pairs might see Focus statements like
create two arm segments
ensure two segments in 3D space are correctly positioned
create a connection that makes motion hierarchical.
Only the context of the videos makes sense of the general Challenge. And only close examination of each video in the continuously looping sequence allows them to be matched to the specific Focus.
This process demands a process of explanation, discovery, and development of a model for how the different techniques relate to each other. So a complex model of learning, where techniques become contextual, and not just stand-alone capabilities, further extends the sense of interdependence, and mutual responsibility for the completion of the learning process.
Why One Workstation Per Pair?
In early classes, our students, confronted with a single workstation, asked if they could, or should bring in their own laptops. Our response was that they could, but they didn’t need to.
Most did; and of course it only required half the class to bring theirs in for there to be as many computers in play as there were students.
But the utility of the personal computer, and the nature of the way Challenges are structured prevailed. Adding more computers didn’t assist students in solving Challenges, or progressing in the course.
Without being told not to, all the students abandoned lugging in their own machines, and instead focused on how to work together on one workstation.
One of the realities of teaching screen-based skills, is that the screen perforce becomes the centre of attention. But in AVM, the constant interaction with the partner quickly breaks that down.
Think about it: if you don’t have control of the input in that situation, you have to make your ideas heard, and that means ensuring that the axis of attention between the screen and whoever is running the mouse at that point isn’t allowed to overwhelm the person-to-person interactions.
What most people quickly realize is that the compulsion to fixate on the screen is almost non-existent when you don’t control the screen. We seek to interact with what we can get responses from, and when a student cedes control of the mouse and keyboard, where they get their responses from is – their partner.
Among the most interesting and revealing answers students in exit interviews gave, was in response to the question, “if this course were created around a computer for each student, would it improve, stay the same, or be made less effective?”
The answers were universally ‘Made Less Effective”. That seems conclusive to me.
The Instructor & The Learning Environment
The last element, which we’ll just touch on, is of course the instructor.
In the courses we provide on-site, we provide the instructor, along with the key technology (Mac workstations, 3D scanners, data glove digitizers, etc.).
When schools integrate our programs into their full-time curriculum (as is done in the Cambodians NGS [New Generation Schools] system), we also train instructors who are usually already staff at the client schools.
In the photos in this post, you can see that instruction is not based on the usual head-of-the class arrangement, with a desk that all students look towards to hear what the instructor will tell them next.
Instead, the instructor positions themselves behind the students. From there, they have a clear view of what’s on their screens. Rather than the student who needs assistance being required to draw attention from the instructor, the instructor spends their time monitoring the progress of the pairs, and anticipates need for help – but doesn’t necessarily provide it themselves.
Short words of encouragement, suggestions, but most of all, connecting one pair of students who have made progress on a Challenge with another who are feeling a bit lost, is easy because the teacher has far higher situational awareness of the state of the class as a whole, and consequently, each pair’s success with the Challenge.
Because there are two teams of six, for any Challenge, and for any pair with a particular Focus, there is another pair at the same time with the same Challenge and Focus. The physical separation of one team from another (generally the teams are on either side of the same room, with their backs to each other) cause the pairs to normally work independently.
But when obstacles arise, and often with facilitation by the instructor, a solution can be found by bringing the matched pairs together.
This is often done in the simplest way, when members of one pair walk over to their counterparts in the other team. But extensive use is made of Discord, a game-focused chat utility that allows the pairs to connect, share their screens, or swap files that can be key to getting the pair with less progress on track.
Again, we see the process of learning as optimal when students are driving it. This physical arrangement is part of reinforcing that. It might be noticed in the classroom photos that the desks at which the students sit are actually made in such a way to allow arrangement in an arc. This also contributes.
The arc means that each pair of students gains a line of sight to each of the other pairs working in their team. This makes it much easier to gain a reference to how the others are progressing, whether that means looking for pairs that have solved a Challenge, or those who may be looking for support or suggestions.
But additionally, a pair can use Discord to screen share with the corresponding pair in the other team; and in this way, cooperate in solving the more puzzling Challenges.
Custom arc-shaped desks aren’t strictly necessary; we can move furniture around to allow a close approximation in a standard computer lab. We’ll talk in another post about the process of coming in, setting up, and closing up a class; but this is what directly relates to the dynamic in the class.
I hope you found this interesting, and perhaps thought provoking. We’re always interested in hearing from schools considering the options that we provide: one way to do that is to comment on this post. If you received this by email, you can also reply to the email (I think, we’ll check)
Thanks for reading
Dan Donaldson, Founder
Apptessence Education and Technology Inc [Canada/Ontario]