MOOC, or massive open online course, is a hot topic these days. There is a great deal of excitement over what some see as the future of higher education, and for good reasons. It is a new yet familiar chapter of the Internet’s crusade against the old, the inefficient, and the undeserved monopolies of the pre-digital economy.
We see in the music industry that digital technology tilts the balance of power towards the consumer. Pink Floyd’s former manager Peter Jenner attacks the debundling of music albums as turning “a £10 product, the album, into a £1.60 product, the two singles that are worth buying.” Yet, from the consumers’ perspective, there is no good argument for why we should be forced to pay for the other ten songs we don’t like.
In a similar fashion, MOOCs threaten to debundle higher education. A kid in India today can take a Stanford course on cryptography on Coursera without having to move to Palo Alto and fork out sixty grand a year in tuition. Sure, she doesn’t get to enjoy the green grass and palm trees, but then again, why should she be forced to pay for the whole package if all she wants is the knowledge?
It seems that if greater access to education is good (and it is), then we should embrace MOOCs. So why are they not universally loved? I have a couple of thoughts on this question.
Benefits of MOOC
The Department of Education conducted a meta-analysis of research on online learning dating from 1996 to 2008 and the findings suggest that “the positive effects associated with blended learning should not be attributed to the media, per se.” This is because students generally received “additional learning time and instructional elements” not available in traditional controlled settings. In other words, the benefits of online learning came not from the medium of digital delivery itself but from the flexibility arising from its execution. It is therefore conceivable to introduce the same conditions back into offline classrooms and many education reform ideas, such as smaller classrooms and hands-on learning, attempt to do just that. But ultimately, the cost structure of online delivery may prove to be indispensable to the equation, much like how it would not be possible for the music industry to learn from iTunes by selling CDs with just a single song on them.
So, online education has at least one intrinsic advantage: cost. With cost comes flexibility, availability, and convenience — all the major selling points of MOOC. The question then is whether that leads to a better education for the student and whether it is better for society in general. I think the former is an easier question to answer and it is what most discussions tend to focus on, but the latter is what I am deeply concerned about.
The MOOC Debate
Boston Review recently published a series of articles that lend voices to both sides of the MOOC divide. One of them, titled Are MOOCs Good for Students?, written by Thomas Leddy, Professor of Philosophy at San Jose State University, makes a critical case against MOOCs.1 Californian public schools such as SJSU have bought into MOOCs in a big way largely because the state’s public school system is broke. Online education seems to be an easy fix, both as a tool to fulfill the needs of existing students and as a mean to generate income by widening the student body virtually, but people like Professor Leddy wonder if the schools are simply lowering the standards of education as a superficial solution to a deeper problem. On the issue of education outcome, his criticisms can be summarized in a few key points.
- MOOCs’ emphasis on multiple-choice quizzes fail to teach students how to write clearly and persuasively and to read critically.
- MOOCs reduce knowledge to piece-wise consumption and does not train the “ability to integrate and explore information creatively”.2 Leddy refers to this as “a symptom of our society’s degraded approach to knowledge itself.”
- MOOCs relegate instructors to the role of mere technicians and degrade teacher-student interactions. Students lose the opportunity to participate in the creation of knowledge because they are now further removed from the professors.
While entirely valid, these points are missing the long-term picture. These are substantial criticisms for why MOOCs in their present form are defective and cannot yet fully replace traditional classrooms. However, the keyword here is “yet”.
Companies like Coursera are fully cognizant of the present limitations in their execution and are constantly working to overcome them. Charles Severance, an Associate Professor at the University of Michigan School of Information, teaches a class on Coursera and remarks on his blog:
I love being part of watching the Coursera software and culture growing and improving under our feet. The engineers, course operations, and support folks at Coursera are a pure joy to work with. They come up with new ideas and put them into production so quickly – it amazes me.
Imagine the speed at which Google prototypes new ideas, and apply that to transforming education. Instead of focusing on how online education is limited today, it may prove more enlightening to consider what lies ahead. William G. Bowen, the former President of Princeton, makes a similar point in his Boston Review article The Potential of Online Learning.
So, are the problems raised by Leddy inherent to online education, or simply a result of its current implementation? My money is on the latter.
Online interactions are only going to become richer over time; we went from AOL dial-up to high definition video conferencing in a single generation. Upcoming technology, like Leap Motion and Google Glass, hints at how human-computer interactions will evolve in the near future. My feeling is that it is too early to write MOOCs off as lecture videos with multiple-choice questions. But I do agree with the argument that it might be premature at this stage of development for schools like SJSU to put all their eggs into the MOOC basket.
Even if we truly believe that online teaching will forever be an inferior alternative to traditional lectures for some purposes, there are clearly elements of a modern college education that are directly replaceable by MOOCs without any loss of content. Engineering students for example perform much of our learning by doing problem sets. Lectures are generally a non-interactive affair, unlike a history class where some form of debate or discussion can conceivably arise. Even with just the limited technology available to us today, one can imagine a MOOC system of online lectures where a small specialized teaching staff is required only to hold office hours and to grade parts of the problem sets that are not machine gradable (at least not with current natural language parsing).
It is easy to see why Computer Science professors tend to be the ones behind MOOC start-ups: many of them already run their own classes like one. At Stanford, it is common for students majoring in CS to watch all their lectures online and turn in their problem sets by e-mail. Attendance for recorded lectures easily dips below 30% after the first week of school. This often leads to comical situations where you bump into friends taking the same final exam as you, and neither of you had any idea that you were enrolled in the same class.
You can make the argument that these Stanford students are wasting their expensive tuition. The counter-argument to that invokes the Sunk Cost Principle and the fact that these students get to spend the time they free up on potentially more useful things, such as working on a personal project or networking with future Mark Zuckerbergs. But in either case, it is an inefficient system.
The New York Times Analogy
Until its recent price adjustment, it was significantly cheaper to subscribe to the physical New York Times paper and gain free complimentary access to the online version, than to pay for the online subscription as a standalone product. This marketing trick allowed the paper to artificially inflate its number of home subscribers, but the unwanted physical copies get delivered right to the trash can.
The real threat of MOOCs is not poorer education outcome because that is a solvable problem: deploy it first for subjects that lend themselves to remote learning, then work on improving the technology to cater to other fields of teaching.
The real threat is that MOOCs will debundle college education and force each component to explicitly demonstrate its true value to the tuition-paying consumers. To many people, the physical subscription of the New York Times was worth paying for only because the more expensive digital subscription is bundled with it. Ten years ago, we were willing to pay for an album for the two songs we like. Today, we purchase single songs off iTunes, or just stream them off Spotify because we recognize that even paying for the permanent rights to a single song is a level of commitment that most people do not need. Welcome to the world of piece-wise consumption.
The Debundling of College Education
The full college experience package costs tens of thousands of dollars a year. Learning through traditional classrooms is ostensibly a major component. If we carve that portion out and put it on the Internet for free, then what reason compels us to continue paying for college? Merely for the certification to prove that you received some form of education? Like the old music industry, this is a house of cards just waiting to be disrupted.
If the choice is between paying a hundred thousand dollars and not receiving a college education, then we can make a good case in terms of expected returns for why we should pay for education. But imagine when the choice becomes more granular: pay for a traditional college education experience or get the online version for next to free sans the frills. Maybe then we need to take a serious look at whether the remaining parts of the college bundle (social interactions, research opportunities, etc) is worth paying for individually, or if we can find other cheaper and more effective means of fulfilling those needs.
This change in mindset will have serious implications on the future viability of the university as an institution for promoting knowledge.
An Argument for Inefficiency
It is fairly obvious that New York Times’ attempt at bundling its subscription is a highly inefficient and wasteful system that led to newspapers being delivered to people who did not want them. But what is less obvious is the reason why these inefficiencies exist in the first place. In this case, it is because print advertising is far more profitable than online advertising and therefore it is financially beneficial for the newspaper to artificially inflate its print subscription count.
How much of today’s academia rely on such inefficiencies to survive? If students were not forced by bundling to contribute towards the maintenance of institutions of learning, would there be enough money left in the pot to encourage basic research, provide a venue for pooling knowledge, and keep professors gainfully employed?
If the New York Times cannot figure out a way to stay afloat after it gives up on bundling and traditional revenue models, then should professional journalism be allowed to simply vanish? What if, by laying bare the inefficiencies of the present model, we destroy the only (current) mean through which a public good is being generated, one that no individual would be willing to pay for if it were sold unbundled? MOOCs make consumption of knowledge cheaper, but who will pay for the creation of knowledge?
I am not making an argument in favor of preserving outdated models. It is a good thing that the music industry was dragged kicking and screaming into the new digital age, despite its unfounded doomsday prophecies of the death of music. The key is managing the transition. Before schools bet everything on MOOCs in a short-sighted attempt to boost immediate revenue, we need to consider the long-term effects, lest these institutions find themselves incapable of producing knowledge and academia as a whole is left poorer off. If our knowledge engines today are heavily dependent on the inefficient system that is college tuition to prosper, then we must break that dependency and find alternative ways of keeping those engines alive when the inefficiencies are eventually erased by new technology.
If record labels disappeared off the face of the planet today, it is hard to imagine that we would be far worse off for it. There is a wide variety of distribution channels and revenue models just waiting to step in, and the quality of music is unlikely to suffer as a result. The same cannot be said for professional journalism and knowledge creation. Inefficiencies can be symptoms of underlying problems. We need to go deeper.
On a final related note, I think it might be in the interest of those who champion the teaching of the humanities for software engineers to figure out an effective way to teach philosophy and history online. This sounds counter-intuitive, but just imagine a future where all kinds of STEM courses are readily available online and provide students with a learning experience that is not all that different from a college STEM degree. When the choice is between learning computer programming in a flexible online environment for free and paying tuition to get a degree in English literature in a traditional classroom, the pure economics of that near future will deal a terrible blow to the liberal arts.
- It is interesting to note that many of the professors in MOOC start-ups have Computer Science background, including Andrew Ng and Daphne Koller of Coursera and Sebastian Thrun of Udacity. Meanwhile, philosophy professors from SJSU penned an open letter to protest its school’s embrace of MOOCs. (back)
- This reminds me of Azuma Hiroki’s The Animalization of Otaku Culture paper describing what he sees as the fragmentation of culture, where cultural elements in modern media are consumed in discrete pieces instead of a coherent narrative. (back)