Plenty of groups these days are trying to reinvent college, and that’s not surprising at a time when higher ed is under fire for ever-rising costs, ballooning student debt and even questions about the value of a degree.
On one hand, entrepreneurs and foundations are rushing to offer higher education via new models and price points. Meanwhile, college researchers and innovators are diving into learning science and experimenting with new teaching methods. But those groups don’t always talk to each other, or even know what the others are working on.
This week for the podcast, we’re talking with someone who’s trying to build more “connective tissue” between academia and industry when it comes to reinventing college. That guest is Mitchell Stevens, a Stanford University professor who is the director of the Center for Advanced Research Through Online Learning. He describes the higher-ed landscape today as an entrepreneurial and philanthropic Wild West—exciting, but also full of tough challenges such as how to safeguard privacy as data and algorithms become central to learning.
EdSurge: Set the stage for us: How has higher ed changed as a result of online education and an influx of Silicon Valley players in the last 5 or 10 years?
Stevens: I would say there are three major things to think about over the last five to seven years:
One is the dramatic decline in the cost of computational capacity and computer memory—your smartphone’s ability to exist. Which makes access to these technologies, both for users but also for entrepreneurs, much more affordable than it was before.
The second is the cost escalation in the postsecondary sector, which has not abated at all. And given the current politics of the United States, the line between people who have college degrees and those who don’t has become much more controversial and contested than ever before. So there’s a kind of urgency to create educational opportunities for a wider variety of Americans at different price points, whether or not those things are called college.
And the third is that the current presidential administration has probably the most laissez faire policy approach to higher education delivery that we’ve seen since the existence of the Department of Education.
You put those three things together, and you get the conditions for a kind of entrepreneurial and philanthropic Wild West, which is what we’re sitting in the middle of this afternoon.
So it has gone beyond a simple conversation of whether online education is good or bad?
Good or bad is a moot point. I mean, we’re living in a world in which Harvard and Stanford have learning management systems in their classrooms right along with broad-access comprehensive universities. All of our students and all of our faculty across the whole sector are, in one way or another, online. And so the distinction between online versus face-to-face education has just become moot.
One of the things you’ve done from your perch has been to highlight issues of privacy and data. You’re drawing attention to this new reality, where places like Harvard and every other college has learning management systems scooping up all this data on student behaviors.
That is both an organizational and an ethical frontier. A learning management system has entered what we used to call my classroom as the faculty member. It was my classroom. I controlled it. When the door closed, that classroom was my purview [and no one knew what happened with the students there].
This week’s podcast sponsor is Emporia State University’s Instructional Design and Technology program: designed for those interested in creating dynamic, interactive learning environments in both public and private sectors, the master’s in IDT from ESU can be completed quickly and entirely online, preparing educators for the new age of the technology-driven learning environment. Learn more here.
Now I share that classroom with digitally mediated technologies like the one that’s carrying this conversation right now. So whose data are those? Are they the professor’s? Are they the student’s? Are they the learning management system’s? Are they the university’s? The answer is probably yes. All of those parties are somehow implicated in the joint venture that is education data. And what I and several others across the country have been trying to do is to have an ethical conversation that gets out in front of that domain so that when an inevitable data catastrophe happens, we won’t start the ethical conversation from zero.
What would be a potential data catastrophe that you’re describing?
Just a completely hypothetical situation: what if the son or daughter of a family that has been quite generous to a highly-selective private research university, her or his identity gets tied to the click-stream data describing her behavior in an intro level comparative literature class, which is on race and sexuality in 20th century America, for example. [And her clumsy statements about race leak out to social media.] That’s a pretty benign risk in a sense, right? That’s a small thing that could happen, but it could create a huge controversy.
So in other words, the systems are not built to have those things happen, but it ends up being a side effect potentially?
Yes. Charles Perrow, a sociologist at Yale University, calls them “normal accidents.” They are accidents that happen by virtue of the sheer complexity of a system. The Boeing air crashes are a perfect example. No matter how carefully you try to engineer something, the sheer complexity of the thing that you’re engineering creates risks. So that’s why I talk about inevitability.
So what do you do? Can you prevent a data catastrophe? Or are we just going to have to live with some trade-offs that are tough?
We had two convenings at the conference grounds in Pacific Grove, California, with colleagues at the Ithaka S+R group in New York, Massachusetts Institute of Technology and the University of Michigan. We modeled those convenings specifically after some other meetings that also occurred in the 1970s in advance of the sequencing of the human genome. Biomedical scientists came together to think about what ethical practice would look like in the wake of that scientific revolution. Did that mitigate the risks of human genome sequencing? No. Did it create a framework for making decisions about responsible behavior? Yes. That’s what we’ve been trying to do with this.
And we actually think that the sequencing of the human genome is a good metaphor for what might be happening with the instrumentation of teaching and learning. We can now instrument that phenomenon in ways that just weren’t possible until the computational revolution.
So for example, like my colleague the economist Raj Chetty at Harvard, let’s say we want to know about earnings returns for different kinds of college degrees. The way we’re going to have to do that is to integrate data from public and private sources in ways that they haven’t been integrated before. Is that necessary for scientifically-informed social policy? Absolutely. Does it bring risks of exposure to the men and women who are described by those data? Absolutely. But in my view, there’s no not going there as responsible educators.
We’re talking at the ASU GSV Summit, which is now in its 10th year. There are thousands of people here, but it’s a certain set of people, mainly investors and company folks. And it seems like you’re interested in having a conversation about who gets to build these tools, right?
Well yes. I mean, I think it’s no little irony that the primary meeting of educational researchers, the American Educational Research Association, is meeting in Toronto at exactly the same time as this meeting. We are on the other side of the moon from the educational research establishment. I think that tells you something both about education research and also about the education technology industry. These are two communities that have not seemed to have much use for each other.
It’s like both of them don’t realize the other conference is that day?
It’s as if they almost don’t realize that the other exists. That’s a very large problem, not only for the future of education in this country, but also for science, because the people who are doing something called science are not connected to the people who are doing something called the edtech industry. There’s only so much cumulative improvement you can have in any sector when there isn’t some sort of collective discourse about cumulative improvement. That’s what science is, essentially.
It seems like higher ed is a place where they should get this right?
I absolutely think we are the place that should get this right. And I have been working with colleagues at Stanford and at other institutions—Carnegie Mellon, Michigan, Harvard, and MIT, Berkeley, [UC] Irvine and elsewhere—to think about how we could build what a colleague of mine called “connective tissue” between academia and industry, in which both academia and industry are going to benefit from that association.
At present, a lot of the edtech sector is really organized around proprietary recipes that are tied to lots of promises that they’re selling to their clients. But if you get under the hood, there may not be very much science there. Similarly, educational researchers may not be asking the kinds of questions that are immediately relevant to the entrepreneurs in the sector.
So what my colleagues at Stanford and I are trying to do for education is very much what engineers did with industry during the Cold War, which is they cooperated with each other on the presumption that industry offered the scale that Washington needed to fight the Cold War. Academia offered the training and norms of disciplined peer review that really got the science going. And industry affiliate relationships in engineering are completely normal.
I notice you did not describe education researchers as the good guy and industry as the bad guy.
I would say that as a group, educational researchers have been highly skeptical of private capital. They have long presumed, I believe erroneously, that education is primarily a public good, that it should be supported by government and philanthropic resources. And they have been allergic to the notion that education is simultaneously a civic project and a business sector.
In my view, education in the United States has always had a joint venture quality. It’s always been a cooperative public-private partnership between government, philanthropy and industry. But the character of that partnership is changing at this moment in history, and frankly, government investment in education is steadily declining. And I don’t anticipate a world in which Americans are going to have the political will to raise the tax dollars that are going to be needed to provide the educational opportunities that our own people need for the next 25 years. That capital will come from the private sector.
Responsible educators should recognize that fact and then work with those sources of capital to think through what kind of disciplined educational sector we would like to have—one that would be both civic and commercial at the same time. What would that look like? How would it be responsibly governed? How would educational providers that are not conventional schools or colleges and universities demonstrate that they are responsible civic actors, that they deserve the trust of the federal government and Americans’ tax dollars? How will that kind of ecosystem get built?
And I’m ambitious enough to think that the norms of scientific peer review that have governed higher education and research within universities have a fighting chance at providing the framework for thinking through what a responsible education sector would look like.
What I frankly fear is that this entire sector will become defined essentially as an industry and not as an educational project, and we’ll wake up one morning, and it’ll all be governed by the Federal Trade Commission. In my view, that would be a civic disaster. If education in this country comes to be defined exclusively as a business, then we no longer have the value of the joint venture quality that has made American education so flexible and dynamic for 200 years.
If I hear you correctly, what it sounds like is: first there’s been this disinvestment by the public sector, and now there’s a chance of a disengagement of academics from the conventional college and university sector.
Yes, not only a disengagement of academics in the conventional college and university sector, but we’ll also have an education data science that’s essentially a proprietary science that says: “Trust us, it works.”
And that’s where I think having academic institutions that see themselves primarily as arbiters of truth and supporters of systematic research could make valuable contributions to the sector that other players can’t in the same way.