This is where the rubber meets the road for transforming higher education. Community colleges are beset by waitlists (400,000 in California alone) and bottlenecks in important introductory courses, as well as low success rates. If scaled-up MIT-quality teaching can help with solve those problems, MOOCS could be truly revolutionary. Massachusetts Bay Community College president John O'Donnell calls edX an invention comparable to Gutenberg's printing press.
Online classes have been around for going-on two decades, so what's the big-deal about MOOCs? Scale.
So far, online courses have offered convenience, but they generally haven't scaled up any more easily than traditional ones; somebody still has to grade the papers, and answer students' questions. One study found 93 percent of institutions charge the same or more for online courses as for in-person ones. No solving the college cost crisis there.
Molly Broad, president of the American Council on Education, refers to the "iron triangle" of higher education: cost, access and quality. The assumption has always been it's a zero-sum game — you can improve any one of those only at the expense of the others. There's also the famous analogy of Princeton economists William Baumol and William Bowen from the 1960s, that college teaching is akin to a string quartet. No matter how technology improves, a string quartet simply can't be performed (well) by fewer people than in Beethoven's day. So the relative cost of college (and musical performance) will always rise, relative to other things where efficiency does improve.
If MOOCs solve the scale problem, they could upend those paradigms. But it isn't easy.
Take assessment. Multiple-choice online quizzes are simple enough, but on more open-ended assignments, MOOC students now are mostly grading each other's papers. When they have questions, they're mostly asking fellow students. "Crowd-sourced assessment" raises obvious questions. MOOC leaders are exploring artificial intelligence solutions but admit many aren't fully baked.
EdX's Agarwal even said his group is exploring a kind of rubric of "self-assessment." Asked if he had faith that, particularly in a course aspiring to credit-worthiness, students could really grade their own essays, he replied: "Faith? Yes. Certainty? No."
Cheating's another problem that suddenly matters with credit at stake. EdX is working with a testing company to arrange for proctored exams in centers around the world. Coursera says it will be easier for far-away students to let them wave an ID card and take a test in front of a webcam, proctored from afar. MOOCs won't offer those things for free. But they could cost much less than, well, the full string quartet.
Broad, of ACE, said MOOCs are promising, but her group will send faculty out to "kick the tires" and research whether online courses enrolling 150,000 can really be credit-worthy. They'll talk to both students who complete and those who drop out (at edX, 80 to 95 percent who sign up don't finish the work).
A likely outcome is more blended models like the Massachusetts experiment, where MOOCs provide the backbone and resources local institutions can't offer, but local institutions still handle the one-on-one and award the credit. Such models could be "the best of both worlds," said Coursera co-founder Daphne Koller. Versions are already in places as varied as San Jose State and the National University of Mongolia.
Struck, the Penn classicist, agrees courses like his will likely work best partnering with local institutions much closer to the students, at least when it comes to credit. Intro-level science classes are one thing, but it's just not feasible at a scale of 54,000 for a class like his.
Higher education involves both transmitting information and "experiential learning that changes a person," he said. For the latter, at least in his subject, the technology's not yet there.