A Sense of Scale
The ups and downs of scaling in education with Gardner Campbell
We've all heard discussions of "scaling up" college and university courses and programs. But these are usually examinations of how to get the most bang for the buck through the widespread replication of high-profile courses.
Here, Gardner Campbell, an associate professor of English at Virginia Commonwealth University explores the notion of scale in education — scaling up and/or scaling down — and shares some of his own experience "playing with scale" in an English course at VCU.
Mary Grush: What is scale in the context of education? What are some of the keys to understanding it?
Gardner Campbell: Typically, scaling and sustainability are the two sister "S" words that get tossed around when curriculum and other education programs are discussed.
If a program is considered sustainable, that means you have the resources to run it; you have enough faculty to staff the classes for the foreseeable enrollment demand; if it's a course required for the major you can offer it often enough for a cohort to be able to take it; and so forth. Certainly, sustainability is important — but so are pilot programs and experiments!
Scaling, by contrast, is a question not only of resources but also of mission and priorities. So I've always been struck by the way discussions about scaling habitually veer toward how to replicate whatever the experience or program or course is for as many students as possible and at as low a cost to the institution as possible — usually calculated in terms of faculty labor. In other words, the question of scaling comes up in the context of mass production, which inevitably involves standardization and something like an assembly line.
The question of scaling comes up in the context of mass production, which inevitably involves standardization and something like an assembly line.
That has always seemed to me to be a one-way understanding of scaling that has some really severe unintended consequences. One is that you begin to look at every idea about learning, every design of every education program solely in terms of minimizing labor cost and maximizing profit by large-scale replication involving fewer faculty while keeping the tuition pricing more or less at the same level. Why? Because otherwise, universities are not making the money on these programs they want to make.
After I got into senior leadership — this was in the mid-2000s, running through roughly the mid-teens — I noticed I was hearing a lot of talk about how universities are businesses. Certainly, it's true. To be sustainable, universities need to be able to meet their budgets, have the money they need to keep the lights on and the physical plant in good repair, pay personnel… all the practical concerns of business.
However, stating that "universities are businesses," coupled with the term "scaling" in its current, standard usage, seems to leave us with a very flatly defined assertion.
Yes, universities cannot run at a deficit. They must make money to remain open. But a moment's reflection will bring to mind many different kinds of businesses — public libraries, mortuaries, veterinarians, churches. These institutions all have financial needs, but they're all very different in mission and social/cultural function. A nonprofit corporation is going to have a very different approach to budget, fundraising, and serving its mission than a Wall Street brokerage firm. And yet, the terms we use seem to get narrowed; rendered unidimensional.
As a result, scaling, as a topic of conversation, is almost always about scaling up. And the scaling up we talk about has, just as I've said, to do with presenting the educational experience to as many students as possible at as low a cost to the institution as possible — while, notably, usually not passing on any of those savings to the student, because otherwise the university is not going to make the money it wants to make.
Scaling, as a topic of conversation, is almost always about scaling up.
Grush: What happens when you don't assume that all scaling goes up?
Campbell: A great question! This past fall semester I taught a course that allowed me to experience a kind of radical scaling — both up and down, and in ways I hadn't imagined before. I thought, "This is scaling, but it's not all about scaling up!"
This past fall semester I taught a course that allowed me to experience a kind of radical scaling — both up and down, and in ways I hadn't imagined before.
The class was low enrollment, scaled down by pure circumstance: It was added fairly late in the registration cycle. It was also a special topics course, and those tend, especially in pre-1700 literature, which is my scholarly specialty, not to be heavily enrolled.
With this accidental scaling down, I asked myself what I might re-imagine in a scaled-down way. I chose to scale down the amount of rapid reading in the semester, while scaling up many class activities and different ways of encountering the basic text we were reading. We would spend three days on a part of the text instead of one. One of those additional days would be framed entirely in terms of students' online annotation using Hypothes.is, and another could be devoted to free-form discussion. So we were doing much more work on much less text. I didn't recognize it at first, but I was playing with scaling. And one consequence was that students felt more engaged and more accountable for their engagement.
I didn't recognize it at first, but I was playing with scaling.
Grush: Did you identify some other examples that take us out of the "one-way" understanding of scale?
Campbell: Yes, that radical scaling experience got me thinking more intensively, and importantly, more variously about what scaling really is, or can be. Why not scale down? Where might we scale down in a typical four-year baccalaureate program? I started thinking about how and why we might scale down there and scale up here… And I was questioning what that has to do with curriculum, what it means for pedagogy… And what does all that have to do with cost?
That radical scaling experience got me thinking more intensively, and importantly, more variously about what scaling really is, or can be.
Grush: What do other, external examples of scale show us — widely known education programs that address scale differently? I'd guess that some of those examples could represent fairly major trends… What can we learn from them?
Campbell: You don't hear the word MOOC very much anymore, the massively open online courses that at one time consumed so much of our attention. For a while, MOOC was the new darling, this idea that you could scale an educational experience up, not just substantially, but massively by using network effects in online learning.
Two main models of MOOCs emerged. There was the C-MOOC, built on connectivist ideas of learning that rely on network effects among the learners. And there was the X-MOOC, which tends to scale things up as massively as possible at as low a cost as possible, most often while charging tuition and structuring the course around learning management systems in familiar ways. And that became what Coursera, edX (now owned by 2U), and other kinds of X-MOOC-oriented businesses took up.
All this time, online learning became a much bigger sphere of interest and investment for colleges and universities. They wanted to be able to get in on what they imagined to be these massively profitable online courses.
Some of that initial energy around X-MOOCs morphed into the idea of so-called "evergreen" courses, in which a senior faculty member would design a course that could be more or less supervised by adjunct faculty or by term faculty who were not really involved in designing the course, but could "run it," so to speak. And more and more of the delivery of those courses became automated and readied for scaling up.
One professed goal for these evergreen course models was to increase what they called "student success," which was all too often defined only in terms of a grade of C or higher in the course, pushing people through the four-year program more rapidly and at greater completion rates, while minimizing labor costs. So with that, we're talking about content transfer and course certification. We're not talking anymore about ideas for learning. We're not talking anymore about a really rich or engaged or sophisticated student experience. And we're back to talking about maximizing income while minimizing labor costs. All this really does change the conversation when we think of the university experience, about what we are offering our learners. It's concerning.
Grush: That is concerning. What tools do we have that we can learn to use to support scaling up and/or down, while avoiding these pitfalls?
Campbell: You can scale up or down well if you employ network effects and develop a sense of scale to pair with that; an awareness of the impacts of scale on what you are doing. We should always remember that, as Doug Engelbart reminded us, scaling is not linear. You can't have a sixty-foot tall human being with the same proportions as a six-foot tall human being, because the six-footer's bone and limb structure cannot support a sixty-footer!
You can scale up or down well if you employ network effects and develop a sense of scale to pair with that; an awareness of the impacts of scale on what you are doing.
The rise of decentralized, non-algorithmically-driven social media might be a good example, if you think of it in terms of blogging, which is a great tool and was the word of the year at one point in the 2000s. Blogging is a way of employing network effects in which the value of the network increases the more people are blogging on it, because the potential for connections among the networked writers and readers becomes greater the more people who are on the network.
Another great example of network effects, right there in front of us, is Wikipedia, which has a really interesting infrastructure enabling focused, cumulative building in a decentralized network largely made up of volunteers. The result is the largest and most comprehensive reference work in the history of human civilization. As the old saying goes, "It'll never work in theory, only in practice." That's the kind of thinking, based on a rich understanding of the web and what it enables, that can transform one's sense of scale.
You can think about network effects in these examples as ways of scaling up. But what you're scaling up if you're thinking about connections is the opportunity for connection — actually supporting the potential scaling down to the interactions between as few as two people.
And in that, you're not scaling up the way you would as though you were asking, "How many pigs can we put in the pen so we can fill the trough and feed them all at once?" That to me is a disastrous way to think about learning! But with a more developed sense of scale, and tools like blogging put in the hands of people who care, we can avoid the pigpen problem.
Still, have we thought as hard about scaling as we could? Not yet. There's not a huge amount of that way of thinking in my experience of higher education.
Have we thought as hard about scaling as we could? Not yet.
And to try to do something as hard as scaling up a scaled-down experience — pick almost anything a small class enables — would require a great deal of ingenuity, considerable resources, and what I would call a kind of ferocious commitment to finding ways to do this. It might be disruptive. It would be uncertain.
But have such experiments really been tried? Have colleges and universities, public and private, but especially large public universities, made a serious effort to do this thinking about scale? Or have we been going for the low-hanging fruit of replicating massive learning experiences, usually, almost always, enabled with online learning in ways that will minimize labor costs while nevertheless charging students at the same tuition level that they would be charged anyway?
I don't think those experiments have happened. I don't think colleges and universities really understand how to employ network effects. And I don't think they've thought very hard about the way the web was designed: not as an information superhighway, but as "small pieces loosely joined" — more like Jane Jacobs's linked neighborhoods, and less like Robert Moses's brute-force expressways.
I'm always hopeful that there can be some inspired thinking that will move us beyond the barriers that constrain our better use of scale.
All of these questions were stimulated by my inadvertent, almost accidental seminar that happened in the fall. And these are only a few of the ideas that I explored as I played with scaling. I'm always hopeful that there can be some inspired thinking that will move us beyond the barriers that constrain our better use of scale. Maybe we'll eventually develop that sense of scale we can pair with network effects, ultimately to benefit learners.
[Editor's note: Image by AI, created with DALL-E]