November 22, 2017

MOOC Mania: Debunking the hype around massive open online courses

Illustration by Jacob Thomas

In the fall of 2011, Stanford University offered three of its engineering courses—Artificial Intelligence, Machine Learning, and Introduction to Databases—for free online. Anyone with Internet access could sign up for them. As Sebastian Thrun, the director of Stanford’s Artificial Intelligence Laboratory, tells the story, he assumed just a handful of people would enroll in his graduate-level AI class. Instead, more than 160,000 students registered. A massive number.

That’s when the enormous hype began about massive open online courses, better known as “MOOCs.” Since then, Thrun and his fellow lab professors Daphne Koller and Andrew Ng have founded education organizations that offer free online classes. Thrun’s start-up is called Udacity (in part, a takeoff on the word “audacious”), and Koller and Ng’s is Coursera. In December 2011, in response to Stanford’s initiatives, MIT launched its own effort, called MITx (short for “Massachusetts Innovation & Technology Exchange”), and a few months later joined forces with Harvard, drolly changing the name of the organization to edX. A consortium of British universities has also created its own MOOC platform, Futurelearn. So far, more than 90 universities worldwide have teamed up with one or more of these MOOC providers, prompting the New York Times to crown 2012 as “The Year of the MOOC.”

Although it’s clear that there’s a flurry of interest in MOOCs among universities, higher-ed students, the tech industry, and pundits, these free online courses are also likely to have a significant impact on K–12 librarians and other educators. As Joyce Valenza, a teacher librarian at Springfield Township High School in Pennsylvania, pointed out on her SLJ blog, “Never Ending Search,” MOOCs “can reach tens of thousands of students of all ages, regardless of geography or social class. They have the potential to be equalizers. MOOCs have the potential to disrupt traditional education platforms. And experts predict they will.”

In fact, according to Education Week, MOOCs are already making waves among the schoolyard set: the University of Miami Global Academy (UMGA), a private online school for middle and high schoolers, is already exploring how to use MOOCs to serve its 143 students. Late last year, UMGA “launched what may be the first free MOOC for high school students—a three-week test prep class designed to get students ready for the College Board’s SAT subject test in biology,” reports Valenza.

The price of popularity

No doubt, the excitement over MOOCs stems from many of the thorny issues facing education today, including the argument that school curriculums don’t provide students with the knowledge and skills that they need to be successful. But the most powerful case for MOOCs’ rising popularity reflects the soaring cost of college tuition and the accompanying increase in student loan debt. By some estimates, the cost of obtaining a college degree has grown almost twelvefold over the last few decades, and in 2012, our nation’s student loan debt hit a record high, surpassing the $1 trillion mark—a figure larger than the country’s outstanding credit card debt.

Still, the allure of a cost-free education is only part of MOOCs’ appeal. Their open enrollment means that anyone—no matter their age, level of educational attainment, or geography—can (at least in theory) participate. And both the “massive” and the “online” components of these courses mean that they can handle the incredible and ever-growing global demand for higher education. According to UNESCO, within the next decade this demand will reach more than 200 million students a year, most of whom will come from developing countries. To meet that demand in a traditional way would require building thousands of new college campuses. The alternative? Simply take advantage of Internet technologies—something that MOOCs aim to do.

By framing the future of education like this—on campus or on the Internet, offline versus online—MOOCs have sparked much talk about “the end of the university as we know it.” New York Times columnist Thomas Friedman has described MOOCs as a “revolution,” and Stanford University President John Hennessy sees online education as a “tsunami coming.” Thrun himself believes that as a result of MOOCs, in 50 years there’ll be only 10 universities left on the entire planet.

Is this rhetoric or reality? It remains to be seen. But as MOOC mania continues to grow, accompanied by the belief that these courses will ultimately shape our schools’ future, it’s worth pausing to look more closely at what these massive online classes offer. Although tens of thousands of students register for MOOCs, relatively few actually complete the courses. For instance, roughly 14 percent of those who enrolled in Thrun’s 2011 AI course passed the class. And that figure is quite a bit higher than the average completion rate for MOOCs, which according to graduate student Katy Jordan’s research at Open University hovers at around 7.6 percent.

Take edX’s Circuits and Electronics class. More than 150,000 people signed up for it, but only 7,157—a measly five percent—completed the class. And just 69,221 registrants looked at the first problem set and a scant 13,569 took the midterm. Of the 10,262 students who showed up for the final exam, 5,800 people received a passing score.

Why is the attrition rate so high? And why does this pattern persist in most MOOCs? Which students are succeeding and which are failing? Despite a lot of speculation (ranging from students simply being too busy with other pursuits to the courses themselves being too hard or too easy), not much is known.

Part of the problem stems from the registration process for many of these MOOCs—just enter an email address and you’re in. As such, there aren’t a lot of demographic details about the students who are taking the classes. No questions are asked during the sign-up process about age. Gender. Race or ethnicity. Location. Educational background. These questions are sometimes asked in surveys (voluntarily completed), but the demographics of MOOC students aren’t routinely gathered. As it stands, much of what we do know comes from professors releasing their own data, rather than from any of the MOOC service providers themselves.

Georgia Tech’s Tucker Balch, an associate professor at the School of Interactive Computing, released the following information based on the survey of students who took part in his recent Coursera class, “Computational Investing.” Of the 2,535 students who completed the course (or 4.8 percent of those enrolled), 34 percent were from the United States and 27 percent came from non-OECD countries. The average age of participants was 35 (ranging from 17 to 74). Seventy percent were white. Ninety-two percent were male. And more than 50 percent of the students already had a master’s degree or a PhD. Clearly, this is hardly the “typical” undergraduate population (although it’s worth noting that “Computational Investing” isn’t really a “typical” or introductory class). Nonetheless, these figures do raise questions about who exactly is being served by today’s MOOCs: Is it “learners” from around the world? Or, for lack of a better word, is it “knowers” from the U.S.?

Behind the hype

Much of the hoopla surrounding MOOCs is that they increase access to higher education and liberate students (and professors and institutions) from the tyranny of a large lecture class. But a closer inspection—one that moves beyond the mania and scrutinizes the details—suggests that MOOCs may be falling short of many of their promises.

Most of Coursera’s and Udacity’s MOOCs still rely almost entirely on lectures—they’re just videotaped and delivered online, and occasionally divided into smaller, 5- to 20-minute chunks. It’s possible that the reliance on lectures—and by extension on the automatically graded multiple-choice quizzes scattered intermittently between the courses’ videos—reflects these MOOCs’ origins in the science and engineering departments. But it’s clear that this teaching format (on- or off-line) doesn’t really work for many academic disciplines or for many types of college courses, particularly for students more accustomed to active classrooms, hands-on experimentation, and student-professor discussions.

But “it’s early” insist proponents of MOOCs, as they emphasize the experimental nature of these courses and their ability to iterate rapidly. (The often-unspoken implication here is that traditional universities and their professors are inflexible and unchanging.) Despite the unanswered questions about who precisely these classes are serving, more and more universities are hopping on the MOOC bandwagon, seemingly because many fear, as former Princeton University President William Bowen recently quipped in a New York Times article that they might otherwise be “left behind.” And this impulse, almost more than any of the other arguments about access or cost, seems to be driving much of the narrative surrounding why MOOCs, why now.

Interestingly, many of these narratives about the history, structure, and rationale of MOOCs overlook the origins of massive open online courses. The term doesn’t come from the courses offered by Stanford AI professors in 2011, but rather it was coined in association with an open online course—“Connectivism and Connective Knowledge”—offered in 2008 by the University of Manitoba. This class—and the connectivist MOOCs like it that are still offered today—looked quite different from the lecture-based, learning-management, system-focused courses that Udacity, Coursera, edX, and the like offer. Students could take part by using the technologies of their choice—blogs, YouTube videos, synchronous video discussions, and so on—and this participation was aggregated via links and RSS feeds into a main site, one that was openly available on the Web.

But it wasn’t just the tools that were different, the academic backgrounds of the professors behind these innovations were different, too. They were largely from the field of education rather than computer science, and the emphasis was on building learning networks and communities—and helping learners think about how to negotiate online learning spaces—and not simply on replicating or scaling the content delivery of typical engineering courses.

It’s interesting that this origin of MOOCs remains largely ignored (and unfortunate as there are decades of experience from those who’ve taught online and taught with technology that are being left out of many of these discussions), but it’s not particularly surprising. MOOC mania taps into powerful narratives—both true and false—about the relevancy of the curriculum, the cost of college, and the adaptability of education institutions. Many institutions are joining MOOCs, hoping that the mania pans out and that these free online classes will, if nothing else, keep their brands up-to-date. But the questions about who exactly they’re serving with these classes will have to be answered sooner or later as having tens of thousands of students sign up for a class is hardly the right metric upon which to build the future of education.

As K–12 schools begin to investigate MOOCs—weighing their potential benefits and challenges—it will be crucial to ask questions about course completion and student success. While these online classes might offer a way to deliver online education to the masses, it will be just as important—if not more so—to think through how we can provide massive student support.

Freelance journalist Audrey Watters (@audreywatters) has written about education technology for “MindShift,” “Inside Higher Ed,” Edutopia’s blog, and her own site, Hack Education.