On Political Books

May/ June 2013 Profs in the Cloud

The perils and promise of online learning.

By Richard D. Kahlenberg

Higher Education in the Digital Age
by William G. Bowen
Princeton University Press, 200 pp.

William G. Bowen, the former president of Princeton University, is a giant in the field of higher education, having written ground-breaking books in recent decades on affirmative action at selective colleges, class inequality in higher education, the factors that improve college completion, and the role of sports at universities. He has now turned his sights on two of the most important issues facing colleges: the problem of exploding costs and advances in technology as a possible cure.

Higher Education in the Digital Age, a slim and highly readable volume, is built around Bowen’s Tanner Lectures at Stanford University and includes reactions from four astute observers, Stanford President John Hennessy, Harvard education professor Howard Gardner, Columbia humanities scholar Andrew Delbanco, and Daphne Koller, president of the for-profit online education company Coursera. The collection of voices provides a thoughtful and provocative discussion of the emergence of online education, which Hennessy says is hitting colleges and universities with the force of a “tsunami.”

Supporters of online learning argue that it has the potential to pull off a higher education hat trick: reduce costs, raise learning outcomes, and reduce inequalities. Bowen, an early skeptic, now declares himself a “convert,” though one who adopts a measured tone. “I regard the prospects as promising, but also challenging,” he suggests.

Early in his career, Bowen helped identify the “cost disease” facing labor-intensive industries such as higher education and the performing arts, which have found it difficult to raise productivity. Bowen invokes Cornell economist Robert Frank’s observation: “While productivity gains have made it possible to assemble cars with only a tiny fraction of the labor that was once required, it still takes four musicians nine minutes to perform Beethoven’s String Quartet No. 4 in C minor, just as it did in the 19th century.” To make matters worse, the wage premium for highly educated workers, like professors, has grown.

These increased costs are then passed on to students. And to add insult to injury, states have simultaneously been cutting back on the funding of public higher education, shifting the burden to students. In 1980, state appropriations accounted for 32 percent of revenue at public colleges and universities, but by 2009 the share had fallen to 18 percent. Taken together, the College Board reports, in the last thirty years in-state tuition and fees at public four-year institutions have increased a whopping 257 percent, after adjusting for inflation.

No wonder, then, that colleges and universities are looking to online education for a technological fix to the cost disease. If a professor can tape a series of lectures once and doesn’t need to physically show up twice a week to give essentially the same performances over and over again, she can teach additional subjects. And if thousands of students can watch lectures in their dorm room or at Starbucks, the need for new auditoriums and classrooms is reduced. Given the potential cost savings, it’s not surprising that universities are shifting to digital course offerings. The number of students taking at least one online class has increased from one in ten in 2002 to one in three today.

But will learning outcomes suffer as universities move to a greater reliance on online learning? Bowen argues that if online learning options are executed in the right way, outcomes won’t suffer, and, indeed, should improve. Particularly promising, he suggests, is the “hybrid” approach, which mixes online lectures with face-to-face “active learning” sessions. This method, called “flipping the classroom,” was found by one poll to receive 69 percent support from professors.

According to Bowen, the most rigorous study to date of online learning involves a randomized trial experiment of a “hybrid” statistics class designed by Carnegie Mellon University, taught mostly online with a once-a-week face-to-face question-and-answer session. The experiment involved 600 students, some of whom took the more expensive class in a traditional setting while others took less costly hybrid classes. The differences in completion rates and on tests were not statistically significant—and low-income students succeeded at comparable levels in the two sets of classes.

Moreover, Bowen suggests that once professors become accustomed to a new way of educating students, outcomes could actually improve. For one thing, online approaches allow for customized learning, in which students who provide the wrong answer to a question can receive immediate feedback and automated guidance on how to find the correct answer. Likewise, educators can receive feedback to improve their craft. As Coursera’s Koller notes, “We can now collect every click from tens of thousands of students and start analyzing what in our classes is working and what is not; what our students are confused about, and when they figure it out, what they did to reach that understanding.”

Finally, Bowen and other supporters suggest that digital learning can promote greater equality of opportunity in higher education. Any effort to reduce costs without reducing quality should be an enormous boon to low-income and working-class students, who have suffered under what Bowen calls the twin scissor blades of rising college prices and stagnant wages. And in theory, online learning could democratize access to star professors, who until recently were available only to students at elite institutions.

Online learning could also deal with the capacity problems at underfunded two-year institutions, providing access, for example, to the nearly 500,000 California community college students recently placed on course waiting lists. Students who work full-time can take advantage of the flexible scheduling that many online courses offer. Online classes can also open up colleges to students from across the world, supporters argue, noting that two-thirds of students in online classes are not in the U.S.

But skeptics—including Delbanco and Gardner—wonder if the hype is too good to be true. In education, there is often a trade-off between quality and cost. Just as an expensive ten-person seminar usually provides more meaningful learning opportunities than a 300-person lecture class, learning may also suffer in cheaper online classes designed for thousands. A 2011 Columbia University study of 51,000 Washington State Community College students, for example, found that completion rates were 8 percentage points lower for students in online classes than for those in traditional classes. Delbanco notes that while the six-year graduation rate at public four-year colleges is an unimpressive 58 percent, the completion rate for students in what are called massive open online courses, or MOOCs, is a dismal 10 percent.

The low rate of completion for MOOCs surely reflects the fact that students lack skin in the game (MOOCs are usually free and not for credit), and Bowen is careful to advocate the hybrid approach. But the greater the proportion of professor time required, the greater the cost. As Bowen notes, the flipped classroom model may not save resources “and can even lead to higher costs.”

Richard D. Kahlenberg , a senior fellow at the Century Foundation, is the author of "All Together Now: Creating Middle-Class Schools Through Public School Choice" and the editor of "The Future of School Integration: Socioeconomic Diversity as an Education Reform Strategy."


  • Mike on May 31, 2013 9:20 AM:

    I teach mathematics at a liberal arts college in the Midwest. Our precalculus courses are self-paced: students use on online text and do mastery drills online, but must pass all of the proctored exams before the end of the term to pass the course. They can attend optional lectures and have access to tutors if they have questions. These students seem to do as well as students in traditional classes, but most of them do not take calculus. In my upper-level courses ( calculus, linear algebra, topology, game theory, etc.) I poll my students about whether or not they believe the course could be taught online and their responses (given anonymously) are almost 100% in the negative. Their comments are telling: they do not believe they could learn higher-level mathematical reasoning simply by reading the text or viewing a lecture; they need the professor there to answer their questions at the juncture of a difficult computation or proof, or to have a discussion that addresses the particular concerns they have at that moment. As the guy at the blackboard I have to be able to look at their expressions and body language in order to sense when they might be getting lost, or to tell whether or not my answers to their questions were adequate to bring them to understanding. I do not believe that higher-order thinking - even in a subject with only "one correct answer" - can be taught in any but a face-to- face setting.