Intrusive Scaffolding, Obstructed Learning (and MOOCs)

Training WheelsMy five-year-old son recently learned how to ride a bike. He mastered the essential components of cycling—balance, peddling, and steering—in roughly ten minutes. Without using training wheels, ever. That idyllic scene of a bent-over parent pushing an unsteady child on a bike, working up enough speed to let go? It never happened. At least not with him.

I’m not sentimental for that Norman Rockwell moment, because I had it several years earlier with my older son. I spent hours running behind him, steadying him, catching him. What made it so difficult for my older son to learn how to ride a bike? Precisely the thing that was supposed to teach him: training wheels.

The difference between the way my sons learned how to ride a bike was training wheels. My older son used them, and consequently learned how to ride only with difficulty. His younger brother used a balance bike (the Skuut in his case), a small light (often wooden) bike with two wheels and no pedals. As the child glides along, thrust forward by pushing off from the ground, he or she learns how to balance in a gradated way. A slight imbalance might be corrected by simply tipping a toe to the ground, or the child can put both feet on the ground to fully balance the bike. Or anything in between.

With a pedal-less bike you continually self-correct your balance, based on immediate feedback. I’m leaning too much to one side? Oooh, drag my foot a little there. Contrast this with training wheels. There’s no immediate feedback. In fact, there’s no need to balance at all. The training wheels do your balancing for you. Training wheels externalize the hardest part of riding a bike. If you’re a little kid and want to start riding a bike, training wheels are great. If you’re a little kid and want to start to learn how to ride a bike, training wheels will be your greatest obstacle.

If you think of riding a bike in terms of pedagogy, training wheels are what learning experts call scaffolding. Way back in 1991, Allan Collins, John Seely Brown, and Ann Holum wrote about a type of teaching called cognitive apprenticeship, and they used the term scaffolding to describe “the support the master gives apprentices in carrying out a task. This can range from doing almost the entire task for them to giving occasional hints as to what to do next.” As the student—the apprentice—becomes more competent, the teacher—the master—gradually backs away, in effect removing the scaffolding. It’s a process Collins, Brown, and Holum call “fading.” The problem with training wheels, then, is that fading is all but impossible. You either have training wheels, or you don’t.

Training wheels are a kind of scaffolding. But they are intrusive scaffolding, obstructive scaffolding. These bulky metal add-ons get in the way quite literally, but they also interfere pedagogically. Riding a bike with training wheels prepares a child for nothing more than riding a bike—with training wheels.

My oldest child, I said, learned how to ride a bike with training wheels. But that’s not exactly what happened. After weeks of struggle—and mounting frustration—he learned. But only because I removed the all-or-nothing training wheels and replaced them with his own body. I not only removed the training wheels from his bike, but I removed the pedals themselves. In essence, I made a balance bike out of a conventional bike. Only then did he learn to balance, the most fundamental aspect of bike-riding. I learned something too: when my younger son was ready to ride a bike we would skip the training wheels entirely.

scaffoldingMy kids’ differing experiences lead me to believe that we place too much value on scaffolding, or at least, on the wrong kind of scaffolding. And now I’m not talking simply about riding bikes. I’m thinking of my own university classroom—and beyond, to online learning. We insist upon intrusive scaffolding. We are so concerned about students not learning that we surround the learning problem with scaffolding. In the process we obscure what we had hoped to reveal. Like relying on training wheels, we create complicated interfaces to experiences rather than simplifying the experiences themselves. Just as the balance bike simplifies the experience of bike riding, stripping it down to its core processes, we need to winnow down overly complex learning activities.

We could call this removal of intrusive scaffolding something like “unscaffolding” or “descaffolding.” In either case, the idea is that we take away structure instead of adding to it. And perhaps more importantly, the descaffolding reinstates the body itself as the site—and means of—learning. Scaffolding not only obstructs learning, it turns learning into an abstraction, something that happens externally. The more scaffolding there is, the less embodied the learning will be. Take away the intrusive scaffolding, and like my son on his balance bike, the learner begins to use what he or she had all along, a physical body.

I’ve been thinking about embodied pedagogy lately in relation to MOOCs—massive open online courses. In the worse cases, MOOCs are essentially nothing but scaffolding. A typical Coursera course will include video lectures for each lesson, an online quiz, and a discussion board. All scaffolding. In a MOOC, where are the bodies? And what is the MOOC equivalent of a balance bike? I want to suggest that unless online teaching—and classroom teaching as well—begins to first, unscaffold learning problems and second, rediscover embodied pedagogy, we will obstruct learning rather than foster it. We will push students away from authentic learning experiences rather than draw them toward such experiences.

After all, remember the etymological root of pedagogy: paedo, as in child, and agogic, as in leading or guiding. Teachers guide learners. Scaffolding—the wrong kind—obstructs learning.

Sacred Heart Mission photograph courtesy of Fernando de Sousa / Creative Commons Licensed. Scaffolding photograph courtesy of Kevin Dooley / Creative Commons Licensed.

39 thoughts on “Intrusive Scaffolding, Obstructed Learning (and MOOCs)”

  1. Actually, the root is not pedo. From Wikipedia (and other dictionaries, too):

    “The word comes from the Greek παιδαγωγέω (paidagōgeō); in which παῖς (país, genitive παιδός, paidos) means “child” and άγω (ágō) means “lead”; so it literally means “to lead the child”.”

    As I used training wheels myself but don’t really recall the experience, I can’t say whether they obstructed my learning or not. But I wonder how the analogy holds across disciplines. For instance, what would embodiment mean for mathematics? And I also wonder, How much scaffolding is intrusive, or too much? Can a first grader begin learning calculus?

  2. One more thought. Removing the wheels also removes part of the complexity of riding a bicycle. It may be that training wheels don’t obstruct but rather the complexity of balancing while applying force to pedals (and an inconsistent force when first learning) may be what’s hindering the learning. In effect, the child scaffolds balancing with feet instead of pedals without needing to apply force to the pedals. Once balance has been achieved, then the child can move to more difficult task of balancing and pedaling at the same time.

  3. I like where you are going with this argument, but your definition of scaffolding is not congruent with the scaffolding in educational theory as derived from Vygotsky and Bandura, which is why you have to specify obstructive scaffolding when making the parallel to your training wheels narrative. Scaffolding is what is necessary to allow the novice to engage in practical application towards mastery of whatever the venture is. Inherent in the definition is the idea of growth, movement, and mastery. You hit the nail on the head that training wheels allow a child to ride a bike without teaching them to ride a bike…but that’s not scaffolding. That’s augmenting, shortcutting, de-enriching. The skuut sounds like a better parallel to educational scaffolding, whereas the training wheels are more didactic or behaviorist.

    A lot of the problem is in what is expected of teachers. Educational research shows better results in terms of student abilities with constructivist or cognitive pedagogical approaches…but those are tough for administrations to measure. It’s much easier to measure student outcomes in regards to a set rostrum of information or expectations. A scaffolded approach to computer programming would be to assign the students to program a set of commands, and as a teacher base your level of interaction with each student dependent on their needs…some might need help researching, others fixing typing errors, others visualizing the end result…the assistance is in direction, not in doing. A behaviorist approach to computer programming is to provide a list of instructions to achieve an end result, and then have the students follow the list. Or in terms of art education, scaffolding would be to draw a picture in the style of Monet, and as a teacher assist students with colors, techniques, or visualization based on their needs, whereas behaviorist would be to have students follow along in doing the steps you are doing, resulting in a created Monet. One system (scaffolding) involves mistakes, obstacles and could result in less than perfect product…but gives full agency to the student, providing them foundation for further development. The other (behaviorism) provides a glossy product, but replication in the future is likely dependent on more scaffolding, so results are rarely duplicated in the real world.

    Why do we teach like this? Administrators and parents like the gloss. It’s tough to tell a principal or parent that a student is struggling but getting there, and the teacher can be blamed for falling through the cracks. If the system is gamed to show progress, we look at progress in finished result of product rather than the methodology to get there.

    How this deals with MOOCs…well, there is no opportunity for teachers to utilize skill to scaffold and tailor to students. There is discussion of personalization, but that comes out of HCI, AI and maybe a touch of cognitive style, with very little (if any) attention to pedagogy. If statistics says it will show you how to run a regression, they will provide you the training wheels to do that. A few students will do well with that and get out their on their own to replicate, but most will still need the training wheels. And unlike with kids and bikes, students who have difficulty replicating in the real world are more likely to give up than keep trying.

  4. The part of the scaffolding metaphor that appeals to me is the idea that you construct scaffolding one level at a time. That is, you stand on the first level of your scaffolding in order to build the second level, you stand on the second level in order to build the third, and so on.

    In a teaching context, it’s helpful to consider what knowledge and skills students might develop in, say, the first week of the course that they can build upon in the second week. What they learn in the second week should then equip them to go further into the subject in the third week. This process continues, hopefully guiding students to the learning goals you have set out for the course.

    This form of pedagogical scaffolding is very useful, but it is sometimes (often?) missing from college courses, perhaps especially MOOCs. Students will, to some degree, naturally build on what they’ve learned earlier in a course, but this process isn’t always designed into the course by the instructor.

    Maybe the problem with training wheels on bicycles is that they don’t actually function as scaffolding in this sense. That is, you can’t build on the knowledge and skills you gain by using training wheels to become a better bicycle rider. Instead, you have to “unlearn” what you learn by using training wheels to actually make progress in your bike-riding ability.

    The balance bike works as scaffolding, however, since the skills you develop riding a balance bike are ones that you use when riding a normal bike. You take those skills and add to them others (like pedaling) to become an independent bike rider. Perhaps the training wheels approach is a little like building level two of one’s scaffolding (pedaling) before finishing level one (balance). Better to build the levels in order (balance, then pedaling).

  5. My only MOOC experience was a Coursera course on Programming in Scala, and it mirrors your points. The lectures were nice, but only went a little beyond the textbook on which they were based. The problem sets were (thankfully for me) heavily scaffolded so that the student’s results would be more elegant and efficient than if I had had to muddle through the thought process required to obtain an answer.

    Potentially the only unscaffolded part of this particular MOOC would have been the physical student-to-student meetups organized on the course site. I didn’t feel I had the time and energy to insert myself in these. Maybe next time.

  6. I bought a Skuut for my son when he was three. He was so excited that he got on it right away, fell over sideways and never got on it again. Sigh. So now we’re getting him a training wheels-enhanced bike and he’s 5. I agree that first he’ll learn how to ride a training-wheels bike, but there’s a huge amount of collateral learning that goes into that. Wearing a helmet, crossing the street, feeling the speed and feeling upright, falling and getting up, etc. So the essential “learning outcome” your post is getting at is balancing, and yes, that will only happen when he learns that particular piece–when the t-wheels come off. But all the other things he needs to learn he can learn with the t-wheels.

    It’s interesting to think of MOOCs as only scaffolding. But I think that the collateral learning in MOOCs may be very different from that of f2f learning. That’s my concern–that students take MOOCs the way my 3-year-old son got on a Skuut; if they fall off the first time, they’ll never get back on. If they’re like your son, they might do great. But many will just disappear.

  7. I completely agree with your observations and your analogy. However, what I fail to see is what you think the solution may be. Your directional statement for a solution as in – “unscaffold learning problems and second, rediscover embodied pedagogy” – is intriguing, but offers no tangible solution for classroom and MOOCs.

    Do you have more specific solution examples in mind?

  8. Love this idea and analogy. My eldest also learned quickly, on her own. She learned balancing from a scooter, and just hopped on a bike and got up to speed quickly.

    My question is, what happens when students aren’t successful? Can we be flexible enough to change the scaffolding? The same scaffolding won’t work for everyone, and what is intrusive to one will be just right to another. And then if we are talking about MOOCs, how do we build in that flexbility since we can’t tweak it for each student.

Comments are closed.