By Peter Sacks
In my 1996 book Generation X Goes to College, I predicted that virtually anyone with a computer and a modem would have access to the storehouse of human knowledge. As a result, higher education as we know would become an anachronism, if not obsolete. The university's status would diminish because it would lose its competitive advantage in disseminating information.
The recent emergence of MOOCs (Massive Open Online Courses), however, raises obvious questions. Are these new teaching methods as effective, in terms of student performance, as real-life classrooms? Can these new technologies bring down higher education costs? Former Princeton president William G. Bowen takes on these questions and others in his new book Higher-Ed in the Digital Age. Once a skeptic, Bowen now concludes that online learning programs will reduce the cost of higher education without harming student learning outcomes.
The Promise of Online Education
His conversion is inspired by the findings of ITHACA, a non-profit organization that conducted "the most rigorous assessment to date" on the economics of online learning technology. That study demonstrated that student learning outcomes, as measured by standardized tests, are no worse in online courses than in traditional classes. Not better, just not worse. Though these results might sound unimpressive, Bowen asserts that they are "very important" because they disprove the common prediction that online education will harm students.
Moreover, he argues, the long-run cost savings of labor-saving technologies could be profound. One reason is that online courses can help institutions fill the gap between the large demand for critical courses and the limited ability of cash-strapped public institutions to satisfy it. Improvements here could reduce the average time it takes to complete a degree, making colleges more productive, affordable and efficient. Bowen acknowledges that the ITHACA study could not answer cost-savings questions because most savings occur over time. Still, Bowen notes that early evidence from various simulations suggest that the long-term instructional savings could be significant. Compared to a traditional course with multiple sections, savings in teaching labor costs alone ranged from 36 percent to 57 percent in the simulations.
Despite this promising data, it is unclear whether most university administrators will embrace this technology. According to the most recent (January 2013) report of the Babson Survey Research Group, 46 percent of whom say that teaching online courses requires more time and effort from the faculty than traditional courses. That percentage was 44 percent the year before. Regardless, the number of students who had taken at least one online course rose by 570,000 in one year, for a new total of nearly 7 million students. That's a growth rate "far in excess of those of overall higher education," the report says. Some 32 percent of all college students had taken at least one online course, the highest percentage on record.
Online learning therefore seems an inevitable part of the future higher education landscape. But like so many technical advancements, we often fail to consider the technology's far-reaching implications. Bowen, who focuses exclusively on online technology's potential to reduce costs, is no exception.
The Cost of Standardization
My major concern is the increasing standardization of the college experience. In order to make online learning worth the cost of development, institutions must achieve economies of scale so as to spread its costs over a large number of students. But achieving these economies of scale means losing certain intangible aspects of the classroom environment; indeed, online education makes no room for the interpersonal interactions that are an essential part of an authentic education. MOOCs in particular lack a human element. For example, the leading MOOC provider EdX (founded by Harvard and MIT) is poised to deploy artificial intelligence software for grading student essays. As someone who has taught college writing, I am most skeptical that a computer program is capable of differentiating an average essay -- though containing all the textbook components of an essay -- from a brilliant one. Machines will reward mediocrity because mediocrity is what machines can be taught to understand.
Rest assured that I am not a Luddite. Indeed, I concede that educators ought to rethink the role of the university. However, they must identify the university's comparative advantage, and given the advent of online technologies, the dissemination of information is not sufficient. Instead, academic leaders should recognize that colleges are uniquely suited to nurture imagination and creativity. No other institution is capable of creating centers of innovation that persist for generations.
To that end, we should let fast and cheap educational programs provide students with basic skills and have the universities provide the real education. Faculty will then take on a new role: Instead of lecturing large classes, they will become expert consultants who guide learners in the application of information for solving, creating and inventing. David Brooks recently cited one professor's prediction that universities will eventually tell students to take certain college courses online, "and then, when you're done, you will come to campus and that's when our job will begin."
My second concern is that cost-saving technologies will have different consequences for rich and poor institutions and for rich and poor students. Public institutions have faced decreased taxpayer subsidies for years and feel acute pressure to reduce costs through standardization. In contrast, wealthy private universities have little incentive to standardize and cheapen their learning environments.
The evidence suggests that online learning programs primarily cater to students from lower socioeconomic backgrounds. To address this question, I used two U.S. Department of Education longitudinal databases to assess the backgrounds of students who were enrolled exclusively in online education programs in 2007-2008, the most recent data available for analysis. Overall, an average of 82 percent of students was enrolled in traditional programs, while some 18 percent were enrolled exclusively in online programs. My findings:
- Test scores. Among the highest-scoring students on admissions tests, virtually all (95 percent) were enrolled in traditional education programs. Low-scoring students were more than twice as likely as high-scoring students to enroll in online programs.
- Selectivity. Just 8 percent of students at colleges with very selective admissions criteria were enrolled in an online program. By contrast, 13 percent of students at moderately selective institutions, 16 percent of those at minimally selective colleges, and 23 percent of students at open admissions colleges, were enrolled in online programs.
- Economic background. Students who enrolled in online programs were more likely to be first generation college students from lower income families than students enrolled in traditional settings. Some 20 percent of students considered low-income and first-generation college-goers were enrolled in online college programs. By comparison, just 14 percent of students considered not low-income and not first- generation college-goers were enrolled in such programs.
- Parental education. Students whose parents were relatively uneducated were more likely than students with highly educated parents to enroll in online college programs. About 21 percent of students whose fathers completed no more than a high school diploma were enrolled in online programs. That compares to just 9 percent and 12 percent of students whose fathers had attained a professional degree or a doctorate, respectively, going to college online.
Thus, it seems likely that lower-income and budget-strapped students will make the most use of online learning technologies. This is all well and good to the extent that more students will have access to higher education. Still, online college programs could further stratify our higher education system, dividing those educated at an "authentic" full-fare university and those who received their degrees from online programs.
We can therefore anticipate the formation of three distinct groups of students. Well-off students will attend the few colleges and universities that are wealthy enough to eschew standardization and automation. They alone will have real relationships with great faculty. A second, less wealthy group of students will use online courses for their general education and attend "authentic" institutions for a short while. For poorer students, online learning could well become the main course. They will attend institutions that, strictly speaking, grant post-high school credentials to the coach class.