Online Learning Is Best in a Supporting Role

Online Learning Should Return to a Supporting Role

By DAVID DEMING

The New York Times, April 9, 2020

As the coronavirus pandemic forces schools and college campuses to go online, the delivery model of education — largely unchanged for centuries — has suddenly been disrupted.

This may seem like the acceleration of a permanent shift toward online learning, but I have my doubts. In fact, economics tells us that technology will make in-person education more valuable than ever.

At the moment, teachers from kindergarten through graduate school are struggling to take their classes online, and the initial results are, understandably, spotty. But the longer this mass experiment continues, the more familiar remote learning will become. And, has been predicted for many years, online performances by superstars are increasingly likely to replace more pedestrian in-person lectures.

This can go only so far, because other important aspects of education are best done by teachers in more intimate settings. Educators will increasingly be tutors, mentors and role models, and economics also tells us that these features of a great education will not scale up.

Therefore, I worry not about the future of teachers but of students. I fear that on-campus learning will become an increasingly important quality differentiator, a luxury good that only students with means can afford.

Consider that online education has been around a lot longer than Covid-19. According to the latest estimates from the Department of Education, 35 percent of college students took at least one course online before the pandemic, and this share has been growing steadily for more than a decade.

This spring, schools and universities had to move courses online with only a few weeks’ notice, and the results have often been ugly. Students face significant challenges, such as spotty access to the internet or an unstable living environment.

Yet the long-term prospects for online learning are good — up to a point. Many universities already offered high-quality lectures online before this crisis, sometimes through partnerships with organization like edX and Coursera. Khan Academy has offered free courses for younger learners. The increased flexibility of online learning has been especially important when students need to balance burdens like jobs or, right now, to care for themselves or relatives who have fallen ill.

After this crisis ends, online lectures will still be increasingly valuable, because they are known in economics as “nonrival goods,” meaning they are not used up as more and more people view them. For this reason, the very best lecturers can teach everyone at the same time. This could make lesser lecturers obsolete and should, at least to some degree, generate much-needed productivity growth in education.

This seems grim for teachers, but I don’t think it will make us obsolete, for two reasons.

First, demand for education is a moving target, and as people become more prosperous they typically want better education, not worse.

So while cost is important, it’s not everything. Bending the higher-education cost curve through online lectures may seem appealing, but the point isn’t to enable everyone to learn on the cheap. Rather, people will want better education for the money, and online lectures alone won’t do it.

This explains why massive open online courses, known as MOOCs, have largely failed to disrupt traditional education despite the hype. Lectures are part of education, but they are not the best part.

Second, as online lectures become better and cheaper, the other essential components of education will take more time and energy.

Within economics this is known as unbalanced growth: the tendency for resources to shift toward parts of the economy where productivity growth is lowest. It is partly why the bulk of U.S. employment has moved away from manufacturing and into the service sector and, in education, why tuition and salaries keep rising. Precisely because they are personal, services are hard to scale up — few people are interested in mass-produced child care, for example.

The personal services provided by educators include tutoring, individualized feedback and mentoring, and numerous studies, as well as countless individual experiences, show that such services are essential for learning.

Good teachers work with students individually or in small groups to diagnose and remedy specific learning gaps. A survey of nearly 200 educational experiments found that “high dosage” tutoring — defined as groups of no more than six students meeting at least four times per week — was one of the most effective ways to improve learning. High-frequency individual feedback also greatly improves student performance.

Teachers are critically important as mentors and role models as well, the studies show. Students are more likely to complete a college degree when teachers have high expectations of them. A female instructor greatly increases the performance of women in math and science courses and their subsequent interest in STEM (science, technology, engineering and mathematics) careers.

Furthermore, racial gaps in course performance are smaller in classes taught by professors from underrepresented groups. Yet the implications of this research extend even beyond race and gender. Mentors matter for everyone, and they can have a powerful impact on students’ life choices and career success. There is simply no technological substitute for these aspects of great teaching.

Because of unbalanced growth, efficiency gains in online instruction will cause educators to shift toward more personal forms of education. Moreover, what economists call “cost disease” tells us that the price of tutoring, mentoring and direct personal intervention will rise, even as lectures are provided more efficiently online.

If these trends continue unchecked, on-campus learning and intensive interaction between teachers and students may eventually become unaffordable for all but the wealthiest institutions and, probably, the wealthiest families.

Two changes are necessary to avoid this tragedy.

First, we must broaden access to institutions that can afford a high-quality on-campus experience. Second, universities under budgetary pressure should resist the temptation to think of online learning technology only as a means of cost reduction.

It is wonderful that technology has enabled millions of students to keep learning even when direct contact is impossible. But once this crisis ends, we will be better off if technology frees up precious class time so that educators and students can engage deeply with each other and build personal connections that will last a lifetime.

David Deming is the director of the Malcolm Wiener Center for Social Policy at the Harvard Kennedy School.