In this edited version of remarks at the High Tech High GSE Speaker Series, the co-author of Disrupting Class discusses the thinking behind the book, his current work, and recent developments in the field. The interviewer is Larry Rosenstock of High Tech High. A video of the event is available at //gse.hightechhigh.org or via the tag above.
How did you get started on Disrupting Class?
In the year 2000, some of the people who started the charter school movement in Minnesota came to Clayton Christiansen and said, “We think if you studied schools through the prism of your theories, what you’d see is a lot of the struggles that are at the core of innovation. And rather than study schools from the inside, if you could show them through these lenses, we think you’d learn a lot about how we might innovate out of the problems that we face.” He took them up on it five years later, and that was when I got involved.
So you came out with the book about a year ago, and the world has changed since then. If you were rewriting it now, what would you change?
There’s a ton of stuff we would change. We projected that by 2019, 50% of all high school courses would be online. I think that misses the point. We want a more student-centric system, whether they’re online or offline. But that projection drew a lot of headlines, and for the first six months people said, “You guys are nuts. This is never going to happen by 2019.” Now, everywhere we go people say, “You guys are nuts. You are way too conservative with this projection.“ Things are happening a lot faster than we realized.
A second thing that has changed is that online learning started as a distance-learning phenomenon, but increasingly it’s getting snapped into hybrid models. Most of the involvements right now are not at a distance, but in physical places.
Bill Gates said that online learning will feature synchronous activity and asynchronous content. His point was, why try to duplicate a lot of asynchronous lectures that are already so strong? It’s the study group function that nobody has really figured out yet. Have you seen anything interesting in terms of online, synchronous study group activity?
Elluminate and others are making platforms for study groups to convene formally or informally. Online learning companies are introducing a whole new conception of teaching, where they’re doing a lot of team teaching, where teachers take different roles in students’ lives, and where there’s rapid interaction in peer networks. People are re-imagining what helpful study groups could look like, as well as understanding that, hey, they may not be helpful all the time. What you want is a flexible atmosphere.
How can we enhance what’s already going on? In mathematics, for example, you have the opportunity for massive customization in terms of feedback online. I assume that writing is something that requires somebody to be reading and reflecting with you, and that you’re not going to be an online teacher reading 500 students’ papers. That’s got to be a challenge.
There’s an interesting service, called SmartThinking, that basically does two things: they built a huge call center, where students can get help within two minutes by phone or online, and they get an edited paper back within 14 hours from highly skilled people. One of the founders had this insight that the problem with a lot of education is that kids’ feedback is limited by their teacher’s time. We have it backwards. What we really want is that whenever the kids need help, they’d be able to get it. That’s why we’re going to start seeing some innovative team teaching methods, using human resources very differently. We have to throw out our assumptions about student-teacher ratios and just focus on mastery, where students show that they’re learning, and then that dictates what those ratios are over time.
Let’s swing between the promise and the challenge. With respect to scaling, for example, there is the challenge of seat time requirements, as codified in state regulations.
Online learning is going to happen one way or another. Whether it becomes student-centric is an open question. A lot of the online providers are saying, don’t just give us funds for having a butt in a seat. Give us the funds if we show performance. Who in education has ever said that before? So we’ve got to push on the policy makers, saying “You’ve got to move to what matters.”
Let’s talk about what’s going on at the Federal level. You wrote a piece for The Atlantic, in which you named four big issues. Let me ask you about each one. The first one is, “Develop internationally benchmarked standards and assessments.”
It turns out that well-codified standards are the key factor driving customization in every industry. So standards are important, but not as we’ve traditionally thought of them—not standards that drive standardization. I would defer to others about what those standards ought to look like, but the design principles are really important. We want standards that are descriptive, not prescriptive. Give people different ways to show mastery.
I once worked on Federal legislation and discovered that you can’t regulate to innovate. Likewise, I would submit that you can’t standardize to innovate. How do you create standards yet not inhibit innovation?
This happens all the time in other fields, where standards bodies come together to enhance innovation. The DVD is a great example. It relies on video compression technology, and there are hundreds and hundreds of patents, and many different patent holders. There was no standard in place to codify this, and there was no innovation whatsoever, no movement. A licensing body came together and said this will be the standard, and innovation took off. That’s a good kind of standards-based process.
Onward to your issue number two: “Improve the effectiveness of teachers and principals.”
It’s become gospel that an effective teacher is the best benchmark of whether kids will succeed. But when you ask what an effective teacher looks like, the answer is that you know it when you see it, like pornography. What we said in The Atlantic piece was that online learning is a big opportunity to get effective teachers to the right people. If you look at the state of Georgia, for example, where they don’t have many teachers who are qualified to teach physics, then limiting yourself to teachers within the state is not good.
Nicholas Kristof wrote in the New York Times that the poorest kids have the worst teachers and the middle-class kids have the best teachers. He may be wrong on both counts, but he never closes the loop by assuming that poor and middle-class kids should be in the same class at the same time. The country isn’t thinking that way, either—and onlinedness could change that, as in the New Yorker cartoon where the two dogs are there and one’s at the computer and says to the other one, “You know what’s really cool? No one knows you’re a dog!”
That’s exactly right, because you’re all in the same place, and you may be working on different things at different times, it totally obviates this fixation.
Your third issue is, “Use data to inform decisions.” Steve Hamilton at Cornell once famously said that “the use of data in public education is considered to be an unnatural act.” What’s the upside and the downside of data?
If you have your standards right, then data can make a lot of sense. At High Tech High you’re using data all the time—perhaps differently than a lot of other people, but you’re using it. The downside, obviously, is if you’re using the wrong data to drive you to do things not in the interest of anyone.
At High Tech High we’ve thought about what single piece of data might offer the best indicator of school quality, by which one could compare schools in very different circumstances. Our “wild card” would be: of all the free-and-reduced-lunch students (a national standard, $28,000 for a family of four) who enter your school in the ninth grade, how many graduate from a four-year college?
That’s a standard that’s really output-focused, and it’s something we can agree on, and it’s about performance. I imagine there are a few others we could come up with that are that generalizable. That’s a commonsense way of thinking about it.
Your issue #4 is, “Turn around the lowest performing schools.” What do you think about that idea?
It’s the wrong question, because it focuses on the schools rather than on the students. Online learning obviates that problem.
Computers are getting smaller and smartphones are getting bigger. What about the prospect of using telephonic devices to get to ubiquity?
It’s a huge opportunity. Netbooks will drive down the cost for low-income families and students. But I think we could say that netbooks are driving backwards into the future. They’re basically little laptops, and then Apple has come along saying, wait a minute, if we put this tablet out there we’ll really blow apart this new medium.
The following questions (marked Q) were generated from the audience.
What recent online innovation has been the most effective, and what’s the most glaring need or gap in online tools?
I hesitate to single out one innovation. One group has just launched a full videogame- based American history course where you run ten missions to rescue American history from being corrupted. That engine can be adapted for courses in literacy, civics, science, math, and other subjects. Another innovation is the platforms that are coming out, which are much more interesting in the long run than the content factories. The next generation learning management systems will be fascinating in that they’ll say, if you’ve got to master this content piece or this standard or whatever else, here are 15 ways of doing it, and they’ll figure out what the student is interested in, and the teacher will help out with that process. This won’t be like a Blackboard, which is what we often see now.
Do you think we might be heading, in terms of platforms, towards a sort of unitary dispatcher, in the way Amazon has come to the fore?
That’s precisely what I think. And the network effects will pay off there because when you think about a good use of data, Amazon has it down. They help you find what you want more quickly, they recommend things for you, and it’s useful.
I took an online class and strongly disliked it because there was no sense of community and social interaction with peers. The chat room felt forced. How would you address this concern?
The need for community and social interaction is why I suspect most of the enrollment moving forward will be in bricks-and-clicks places, but I don’t know that. You can talk to some parents who say, “Thank God you’ve pulled my kid out of that destructive school environment.” So it depends a bit on where you sit, but the capacity for social networking and building community will improve over the years.
It’s hard to believe that many people still don’t have access to the internet. How do we achieve equity in the allocation of resources?
This is a place where public policy can really make a difference. You can’t reach the potential of online learning unless you make it robust. The Obama administration is thinking strongly about a national broadband plan, making sure the pipe gets to schools, and I would also say we ought to think about it more creatively. At Qualcomm they’re wondering, do we need bigger pipes, or can we use wireless in many more creative ways to reach whole communities, with maybe the school being the focal point? There’s a huge movement in one-to-one computing—a lot of kids have laptops, or are going to have netbooks or Apple iPads, because the cost is coming down. For the kids who can’t afford it, that’s a great place for public policy to step in and pay for that.
What does a typical day look like for an online educator?
It depends. At Florida Virtual School teachers work from home, and they’re required to be online from like 8 am to 8 p.m. and on weekends for certain hours. At Advanced Academics they work in centers, so it’s a totally different experience. They do a lot more team teaching, and they have staff available 24 hours-a-day because it turns out that kids like to work sometimes at odd hours. A lot of these are design choices that affect the cost and look of programs, and that are going to be wrestled out over the coming years as we figure out what makes sense, where.
What are you working on next?
We’ve set up the Innosight Institute to do more research. We’re diving into some case studies where online learning is happening—first, so that policy makers can better understand it, and second, for people to see options. Rather than focus on one “best practice” for online learning, they’ll have a range of examples to choose from, based on their circumstances. I think there will be a second edition of Disrupting Class, with a new chapter involving a jobs-to-be-done theory—the idea that the products or services that a company or organization thinks it’s offering are not the ones that the customer is actually buying. Our hypothesis is that education actually is not a job that students are trying to get done. Their job-to-be-done is something like “help me feel successful” or “help me have fun with friends,” and education is in fact competing against video games, gangs, athletics, and on and on. If you understood it from that perspective you might design experiences very differently to have kids demand education, and I think that’s behind what High Tech High does.
To learn more about the Innosight Institute and Michael Horn’s work, visit www.innosightinstitute.org
In this edited version of remarks at the High Tech High GSE Speaker Series, the co-author of Disrupting Class discusses the thinking behind the book, his current work, and recent developments in the field. The interviewer is Larry Rosenstock of High Tech High. A video of the event is available at //gse.hightechhigh.org or via the tag above.
How did you get started on Disrupting Class?
In the year 2000, some of the people who started the charter school movement in Minnesota came to Clayton Christiansen and said, “We think if you studied schools through the prism of your theories, what you’d see is a lot of the struggles that are at the core of innovation. And rather than study schools from the inside, if you could show them through these lenses, we think you’d learn a lot about how we might innovate out of the problems that we face.” He took them up on it five years later, and that was when I got involved.
So you came out with the book about a year ago, and the world has changed since then. If you were rewriting it now, what would you change?
There’s a ton of stuff we would change. We projected that by 2019, 50% of all high school courses would be online. I think that misses the point. We want a more student-centric system, whether they’re online or offline. But that projection drew a lot of headlines, and for the first six months people said, “You guys are nuts. This is never going to happen by 2019.” Now, everywhere we go people say, “You guys are nuts. You are way too conservative with this projection.“ Things are happening a lot faster than we realized.
A second thing that has changed is that online learning started as a distance-learning phenomenon, but increasingly it’s getting snapped into hybrid models. Most of the involvements right now are not at a distance, but in physical places.
Bill Gates said that online learning will feature synchronous activity and asynchronous content. His point was, why try to duplicate a lot of asynchronous lectures that are already so strong? It’s the study group function that nobody has really figured out yet. Have you seen anything interesting in terms of online, synchronous study group activity?
Elluminate and others are making platforms for study groups to convene formally or informally. Online learning companies are introducing a whole new conception of teaching, where they’re doing a lot of team teaching, where teachers take different roles in students’ lives, and where there’s rapid interaction in peer networks. People are re-imagining what helpful study groups could look like, as well as understanding that, hey, they may not be helpful all the time. What you want is a flexible atmosphere.
How can we enhance what’s already going on? In mathematics, for example, you have the opportunity for massive customization in terms of feedback online. I assume that writing is something that requires somebody to be reading and reflecting with you, and that you’re not going to be an online teacher reading 500 students’ papers. That’s got to be a challenge.
There’s an interesting service, called SmartThinking, that basically does two things: they built a huge call center, where students can get help within two minutes by phone or online, and they get an edited paper back within 14 hours from highly skilled people. One of the founders had this insight that the problem with a lot of education is that kids’ feedback is limited by their teacher’s time. We have it backwards. What we really want is that whenever the kids need help, they’d be able to get it. That’s why we’re going to start seeing some innovative team teaching methods, using human resources very differently. We have to throw out our assumptions about student-teacher ratios and just focus on mastery, where students show that they’re learning, and then that dictates what those ratios are over time.
Let’s swing between the promise and the challenge. With respect to scaling, for example, there is the challenge of seat time requirements, as codified in state regulations.
Online learning is going to happen one way or another. Whether it becomes student-centric is an open question. A lot of the online providers are saying, don’t just give us funds for having a butt in a seat. Give us the funds if we show performance. Who in education has ever said that before? So we’ve got to push on the policy makers, saying “You’ve got to move to what matters.”
Let’s talk about what’s going on at the Federal level. You wrote a piece for The Atlantic, in which you named four big issues. Let me ask you about each one. The first one is, “Develop internationally benchmarked standards and assessments.”
It turns out that well-codified standards are the key factor driving customization in every industry. So standards are important, but not as we’ve traditionally thought of them—not standards that drive standardization. I would defer to others about what those standards ought to look like, but the design principles are really important. We want standards that are descriptive, not prescriptive. Give people different ways to show mastery.
I once worked on Federal legislation and discovered that you can’t regulate to innovate. Likewise, I would submit that you can’t standardize to innovate. How do you create standards yet not inhibit innovation?
This happens all the time in other fields, where standards bodies come together to enhance innovation. The DVD is a great example. It relies on video compression technology, and there are hundreds and hundreds of patents, and many different patent holders. There was no standard in place to codify this, and there was no innovation whatsoever, no movement. A licensing body came together and said this will be the standard, and innovation took off. That’s a good kind of standards-based process.
Onward to your issue number two: “Improve the effectiveness of teachers and principals.”
It’s become gospel that an effective teacher is the best benchmark of whether kids will succeed. But when you ask what an effective teacher looks like, the answer is that you know it when you see it, like pornography. What we said in The Atlantic piece was that online learning is a big opportunity to get effective teachers to the right people. If you look at the state of Georgia, for example, where they don’t have many teachers who are qualified to teach physics, then limiting yourself to teachers within the state is not good.
Nicholas Kristof wrote in the New York Times that the poorest kids have the worst teachers and the middle-class kids have the best teachers. He may be wrong on both counts, but he never closes the loop by assuming that poor and middle-class kids should be in the same class at the same time. The country isn’t thinking that way, either—and onlinedness could change that, as in the New Yorker cartoon where the two dogs are there and one’s at the computer and says to the other one, “You know what’s really cool? No one knows you’re a dog!”
That’s exactly right, because you’re all in the same place, and you may be working on different things at different times, it totally obviates this fixation.
Your third issue is, “Use data to inform decisions.” Steve Hamilton at Cornell once famously said that “the use of data in public education is considered to be an unnatural act.” What’s the upside and the downside of data?
If you have your standards right, then data can make a lot of sense. At High Tech High you’re using data all the time—perhaps differently than a lot of other people, but you’re using it. The downside, obviously, is if you’re using the wrong data to drive you to do things not in the interest of anyone.
At High Tech High we’ve thought about what single piece of data might offer the best indicator of school quality, by which one could compare schools in very different circumstances. Our “wild card” would be: of all the free-and-reduced-lunch students (a national standard, $28,000 for a family of four) who enter your school in the ninth grade, how many graduate from a four-year college?
That’s a standard that’s really output-focused, and it’s something we can agree on, and it’s about performance. I imagine there are a few others we could come up with that are that generalizable. That’s a commonsense way of thinking about it.
Your issue #4 is, “Turn around the lowest performing schools.” What do you think about that idea?
It’s the wrong question, because it focuses on the schools rather than on the students. Online learning obviates that problem.
Computers are getting smaller and smartphones are getting bigger. What about the prospect of using telephonic devices to get to ubiquity?
It’s a huge opportunity. Netbooks will drive down the cost for low-income families and students. But I think we could say that netbooks are driving backwards into the future. They’re basically little laptops, and then Apple has come along saying, wait a minute, if we put this tablet out there we’ll really blow apart this new medium.
The following questions (marked Q) were generated from the audience.
What recent online innovation has been the most effective, and what’s the most glaring need or gap in online tools?
I hesitate to single out one innovation. One group has just launched a full videogame- based American history course where you run ten missions to rescue American history from being corrupted. That engine can be adapted for courses in literacy, civics, science, math, and other subjects. Another innovation is the platforms that are coming out, which are much more interesting in the long run than the content factories. The next generation learning management systems will be fascinating in that they’ll say, if you’ve got to master this content piece or this standard or whatever else, here are 15 ways of doing it, and they’ll figure out what the student is interested in, and the teacher will help out with that process. This won’t be like a Blackboard, which is what we often see now.
Do you think we might be heading, in terms of platforms, towards a sort of unitary dispatcher, in the way Amazon has come to the fore?
That’s precisely what I think. And the network effects will pay off there because when you think about a good use of data, Amazon has it down. They help you find what you want more quickly, they recommend things for you, and it’s useful.
I took an online class and strongly disliked it because there was no sense of community and social interaction with peers. The chat room felt forced. How would you address this concern?
The need for community and social interaction is why I suspect most of the enrollment moving forward will be in bricks-and-clicks places, but I don’t know that. You can talk to some parents who say, “Thank God you’ve pulled my kid out of that destructive school environment.” So it depends a bit on where you sit, but the capacity for social networking and building community will improve over the years.
It’s hard to believe that many people still don’t have access to the internet. How do we achieve equity in the allocation of resources?
This is a place where public policy can really make a difference. You can’t reach the potential of online learning unless you make it robust. The Obama administration is thinking strongly about a national broadband plan, making sure the pipe gets to schools, and I would also say we ought to think about it more creatively. At Qualcomm they’re wondering, do we need bigger pipes, or can we use wireless in many more creative ways to reach whole communities, with maybe the school being the focal point? There’s a huge movement in one-to-one computing—a lot of kids have laptops, or are going to have netbooks or Apple iPads, because the cost is coming down. For the kids who can’t afford it, that’s a great place for public policy to step in and pay for that.
What does a typical day look like for an online educator?
It depends. At Florida Virtual School teachers work from home, and they’re required to be online from like 8 am to 8 p.m. and on weekends for certain hours. At Advanced Academics they work in centers, so it’s a totally different experience. They do a lot more team teaching, and they have staff available 24 hours-a-day because it turns out that kids like to work sometimes at odd hours. A lot of these are design choices that affect the cost and look of programs, and that are going to be wrestled out over the coming years as we figure out what makes sense, where.
What are you working on next?
We’ve set up the Innosight Institute to do more research. We’re diving into some case studies where online learning is happening—first, so that policy makers can better understand it, and second, for people to see options. Rather than focus on one “best practice” for online learning, they’ll have a range of examples to choose from, based on their circumstances. I think there will be a second edition of Disrupting Class, with a new chapter involving a jobs-to-be-done theory—the idea that the products or services that a company or organization thinks it’s offering are not the ones that the customer is actually buying. Our hypothesis is that education actually is not a job that students are trying to get done. Their job-to-be-done is something like “help me feel successful” or “help me have fun with friends,” and education is in fact competing against video games, gangs, athletics, and on and on. If you understood it from that perspective you might design experiences very differently to have kids demand education, and I think that’s behind what High Tech High does.
To learn more about the Innosight Institute and Michael Horn’s work, visit www.innosightinstitute.org
