Coursera CEO: Colleges will survive the online education revolution
(Poets&Quants) — When Rick Levin left Yale University two years ago as one of the Ivy League’s most successful presidents in history, he imagined taking a year-long sabbatical, writing on economics and higher education, and perhaps getting back into the classroom to teach.
After a 20-year run as Yale’s president, Levin certainly deserved some time for introspection. But a casual conversation at a New York City party would lead to something quite different than a return to the lecture hall.
After informing an investor of his admiration for a relatively new Silicon Valley startup, Coursera, the online learning platform and advocate of massive open online courses (MOOCs), Levin found himself speaking with founders Daphne Koller and Andrew Ng. Initially signed up to be a part-time advisor, Levin was soon offered the larger role of CEO, which he took on in April 2014.
His first year on the job has been a whirlwind, putting the 68-year-old educator at the forefront of a revolution in higher education. Just this week, Coursera announced the first MOOC-based MBA degree with the University of Illinois College of Business. In February, Coursera launched a series of “specializations” in which a school offers a sequence of courses along with a capstone project. Among the most compelling specializations was an offering of four core MBA courses from the University of Pennsylvania’s Wharton School.
Today, the Coursera platform boasts 1,027 courses from 119 partners, including the business schools at the Universities of Michigan and Virginia, IESE Business School in Spain, the Indian School of Business in India, and HEC Paris in France. More than 12.8 million students have registered for courses, including over 1.5 million for Wharton’s more than a dozen Coursera MOOCs.
Online learning was not foreign to Levin when he joined Coursera. In fact, as president of Yale, he was instrumental in launching the university’s foray into online education, first in partnership with Stanford and Oxford, and later with Open Yale Courses. Just before leaving his job, in May 2013, he played a key role in furthering Yale’s partnership with Coursera.
Poets&Quants recently sat down with Levin in a conference room at Coursera’s headquarters in Mountain View, Calif., where Google also makes its home. In our interview, Levin pooh-poohs predictions that free online courses are going to put many universities out of business anytime soon. More likely, he says, employers will begin to recognize the value of online credentials and reward employees for having them with promotions, pay increases, and new opportunities.
True disruption to higher education, believes Levin, will take many years and largely affect commuter colleges not known for deep engagement between students and faculty. For universities that sit on the sidelines, there could be significant consequences. Levin predicts that global rankings of universities are likely to take into account the number of people in the world touched by a university’s professors. That would make a global university’s status and prestige partly dependent on a school’s reach, which can be expanded significantly through online learning.
Poets&Quants: You’ve just celebrated your one-year anniversary as CEO. Were there any surprises when you first took on the role?
Rick Levin: One of the surprises was how few other universities had yet grasped the potential in the broad outreach that MOOCs make possible. We did them at Yale in 2007 and one of the obvious benefits was enhancing access to education. It was to give opportunity to tens of thousands of people who wouldn’t have had it. About 73% of our learners are outside the U.S. and about half of those are in emerging economies.
When I first came on here, I was surprised to hear the prevalent thinking that MOOCs was thought to enhance the quality of on-campus teaching. Many of my old buddies didn’t get it. The main deal is that millions of people can gain access to your courses. There has been a big change in this last year, partly because these are bright people and they are seeing the results. An instructor spends his or her life studying a subject and their mission in life is to advance knowledge in a field and disseminate it. Before, it entailed touching a few hundred lives a year. This is a huge multiplier of that.
I recently showed data to an official at Rice University on the number of people who have completed Rice courses on Coursera. And he told me, “that is more people than the total number of graduates in the history of Rice.” So it’s very cool.
And yet there have also been dire forecasts for the eventual impact on the business model of delivering higher education. Rich Lyons, dean of UC-Berkeley’s Haas School of Business, has publicly predicted that as many as half the business schools in the U.S. could shutter in as little as five years or as many as 10. How disruptive do you think this is going to be?
We are still at the stage where it is expanding the market rather than substituting for educational offerings. The biggest effect is in bringing new learners in. Three-quarters of our learners are over the age of 22. They are beyond secondary and college years. Most of them are working and they are using it primarily for career advancement or personal enrichment in equal proportions. That is not a hugely disruptive thing at this point. That is additive, an enhancement to what we provide.
The evidence is beginning to be very clear that, if properly managed and done strategically, online learning can be a net revenue enhancement. MOOCs can increase the visibility of a university, generate revenue from learners who buy certificates, and serve as lead generation to master’s and other programs.
Over time, there will be a tendency for institutions of higher education to want to use some of these high quality MOOCs in their own instructional programs. But it will take time and there will be institutional inertia that will make it slow. If we look over decades, sure, disruption will happen.
And when this does happen, which colleges are most vulnerable in your opinion?
What great residential universities and colleges offer we are a long way from being able to replicate. We can transmit the content really effectively, and in some circumstances more effectively, but we can’t offer the full educational package online. We are getting better and better at using the platform to develop critical thinking skills. That is a work in progress, but it is not at a level of the deep intellectual give-and-take in a seminar. The extreme polar cases are the Oxford and Cambridge tutorials, which take that to the limit.
Vulnerable are other schools that provide a less intense experience. The more they are commuter colleges without a residential component, the more at risk they are. And then there is the question of when people will use MOOCs as a substitute. If the choice is to go back to school for a degree in computer science or taking courses from Coursera, that is already a close call for a lot of people. But again, none of this is going to happen fast. Internet time moves at light speed but universities run on a very different time clock.
Of course, one of the things that makes this disruptive is the economic model behind MOOCs. Can you speak to that?
The per student cost to deliver a MOOC is a fraction of what it costs on campus. For the straight MOOC, it’s generally estimated that the university’s costs are on average $50,000 to create a course. If that course has 50,000 learners enrolled, it’s $1 a student or about $5 a student for those who might finish the course. That’s a lot cheaper than anything you can do offline.
The degree-granting authority of a university has long been a key asset of higher education’s business model. But with this week’s announcement of a MOOC-based MBA degree at the University of Illinois, that seems to have opened the door for more immediate disruption.
This is a really innovative and interesting way of combining MOOCs with a degree program. The University of Illinois will create eight specializations on Coursera. One of them, digital marketing, is already up and running. Learners can take them for free, or get a verified certificate by paying for them. And students who, either before taking them or having taken one or two, wish to apply to the university for an MBA degree may do so.
They are selective admissions but one of the interesting features will be that Illinois can judge an applicant’s capability based simply on the competence showed in those MOOCs. This is an avenue for people who would never have the paper record to get a first-class MBA to get one. The degree entails another level of engagement above the MOOCs. A student has to take six of the eight specializations and then on top of it you will get a layer of deeper engagement and activity that will be a higher-priced premium offering. At a total cost of roughly $20,000, the tuition is steeply discounted from what an MBA would cost a state resident in Illinois. The pioneering part is that this is built on the foundation of these free online courses. We love this idea. It is really creative.
The critics of MOOCs often note that a relatively small percentage of those who register to take a free online course actually complete it. The research shows it can be under 5%. Why is that so?
It’s similar to what you have at many universities. At Yale, undergraduates can shop their courses for only six days or whatever. You can shop much longer with MOOCs, and entering a course is like sampling it. The low completion rates are due to people who sign up and never show up or quit in the first week when life gets in the way.
If a learner does a week’s worth of work, then 30% to 50% of the people get to the end. And then there are people who get to the end and don’t do the assignments. I often watch lectures without completing a course or an assignment. There are many people who are getting something out of them who don’t complete them.
Completion rates go way up if you pay. Some people do make a pre-commitment, but a lot of people wait to the end to pay. Completion drives payment, though payment helps to incentivize completion.
Critics also worry about the rigor of some of these courses and the grading of them. Do you worry about that?
It’s up to the instructor how to set the grading. They can do it any way they want. We have assessment tools on the platform, a whole variety of machine-graded assessments that are not just multiple-choice questions. We can machine grade computer code, mathematical expressions, and short answers. And while it’s often surprising to academics, it turns out that peer grading works really well. We crowd source it. Instructors find the assessments made by a group of peer graders correlate very highly with their own assessments. This works really well when the professor has set a clear rubric.
How will online learning change the educational landscape in the near term, say over the next five years?
We are going to see people’s economic opportunities substantially enhanced as the labor market recognizes these credentials. That is the main near-term consequence. It is growing really rapidly and it serves a valuable social function. It is a new means of advancing yourself in the workplace. And it is working. We have done a study with researchers at a partner university of 60,000 to 80,000 learners who completed courses and asked them the reason they took a course. We will be publishing the results soon, but the answers strongly indicate that MOOCs are viewed as something that can bring significant career benefits.
In five to 10 years, online credentials will be very significant. Employers already are beginning to say that certificates would be valuable in a job application. Google did a blog post on this recently, listing certificates in computer science. LinkedIn has created an area to list online certificates under the field for education. Now you list your degrees and you list your certificates. Even though we are a startup, Coursera is the second most commonly listed certificate on LinkedIn after Microsoft and we are ahead of Cisco which has had a certificate program going for decades.
We will also see the use of these MOOCs as a foundation for in-house training programs by employers and government agencies. In Singapore, they are paying more than 1,000 people to take our data science sequence of courses. We’ll see more of that kind of thing because it is such an efficient way to give training.