The Education Visionary

Khan Academy founder Salman Khan on the future of learning


"There's a lot more demand for people who want to just improve themselves than anyone would have guessed," says Salman Khan, founder of the wildly popular free educational video series that bears his name and author of the new book The One World Schoolhouse: Education Reimagined (Twelve).

Khan, a 36-year-old Bangladeshi American, first put together a couple of video tutorials in 2004 to help his young cousins learn math. The videos proved so popular on YouTube that two years later he launched the nonprofit Khan Academy to offer free online lectures and tutorials that are now used by more than 6 million students each month. More than 3,000 individual videos, covering mathematics, physics, history, economics, and other subjects, have drawn more than 200 million views, generating significant funding from both the Bill & Melinda Gates Foundation and Google. Khan Academy is one of the best-known names in online education and has grown to include not just tutorials but complete course syllabi and a platform to track student progress.

Reason TV Editor in Chief Nick Gillespie sat down with Khan in October to discuss how American education can be radically transformed, why technology is so widely misused in K?12, and how massive amounts of taxpayer money never make it inside conventional public classrooms.

reason: Talk a little bit about the videos and the enormous growth in their audience during the last few years. 

Salman Khan: People who look at the videos will see someone writing on a digital blackboard. And you'll hear a voice. For a lot of the videos it'll be my voice, working through things, thinking through things'"very conversational. It started with me making it for my cousins. It soon became clear that people who were not my cousins were watching them. They just kind of took on a life of their own. 

Many things have surprised me over the last several years. The biggest thing is that when I made these things I assumed these were for my cousins; they were pretty motivated students. I made them for what I would have wanted if I were 12 years old or 13 or 18 years old. I said: Well, maybe this will be for the subset of people who are really motivated, whatever that means. They'll actually seek out knowledge on the Internet, and then they'll find it useful. 

It didn't take long to realize that the feedback we were getting was from people who were not the traditionally motivated: kids who were about to fail classes, kids who were thinking about dropping out, people who were going back to school. And they were saying [the videos] make me understand the intuition, the big picture, and I'm starting to get excited about math. So the big realization is'"and I think this surprised frankly everybody'"there's a lot more demand from people who just want to improve themselves than anyone would have guessed. 

reason: In the book you mention that New York state spends about $18,000 per public school student per year. Clearly New York state is not known for great schools or great outcomes. We're spending $18,000 a year for flat results over the past 40 years for public schools. What's wrong with the status quo?

Khan: The reason I highlighted that in the book is that a
lot of times people make it sound like it's a money issue. The problem is you can never say you're spending too much on education. It's such an important thing; if you can get a dollar of value in education, it's worth it. 

reason: Although that's not what's been happening. 

Khan: Exactly. And when you look at the $18,000 number (or even in the lower districts that spend less, $8,000 or $9,000), and you multiply that by how many students are in a classroom'"someplace between 20 and 30′"you get a fairly large number. You get something [in the range of] $300,000, $400,000, $500,000. When you do that very simple back-of-the-envelope calculation, you realize how little of that money is actually touching the student. Very little of that is going to the teacher. Very little is going directly for the facilities. Most of that is going for layers of administration. We can actually professionalize teachers as they are, turn it into a career that pays as well as doctors. The money is there. There just has to be major restructuring in how you spend that money. 

(Interview continues below video.)

reason: Is there any reason to believe that if we tripled what we pay teachers we would have teachers that are 200 percent better?

Khan: I don't know. I think the general sense is that there's a lot of lip service being given to teachers: Oh, we need to respect you. We want the best of the best to be doing this. But society's not sending that economic signal. In engineering I used to say: How come more people are going into finance than engineering? Well, look at the salaries, and you get a very clear picture of why. Now that's actually changing in engineering. Engineers can do just as well as or better than people in finance. I think that has to happen in teaching. 

We are already getting a lot of great talent in teaching, but we'll get even more people who aspire to do this. And it will change the dynamic in the classroom to where the students say, I wish I had a chance of becoming that person who I have the privilege to be with in this room. That completely changes the dynamic of the classroom. I think that's possible. 

A lot of the excuses'"oh, we can't have technology; it's too expensive'"those are a round-off error compared to the amount of money that's being spent even on things like textbooks and whatever else. 

reason: One of the things you emphasize is that there are multiple ways and multiple sites of education. Talk a bit about how we have to start reimagining education so it's not something that happens eight-and-a-half months a year in a brick building with bad air conditioning. 

Khan: That's what a lot of people don't realize. We all grew up in this education system. The education system looks fairly similar anywhere you go in the world. All of us just assume this is what school is. What I write a lot about in the book is that no, this is actually a 200-year-old artifact. It comes from the Prussians;
they don't exist anymore, but they were kind of the cornerstone of Germany now. They said: We want to have public education, which is a very egalitarian idea. But how do you do that in a scalable way, in a mass way? Well, it's the beginnings of the industrial revolution. How do we do anything in scale? We put things in batches. 

In schools, [the batches are] age-based cohorts. We have a bell ringing every shift. They go at a set pace; at every station you try to apply something to it and at some point you sift the product: "This is the good product. This is the bad product. That's destined for the supermarket. That orange is going to be juiced." There's the exact same thing with kids. 

The U.S., in the middle of the 1800s, said: We want to do public education too. They said: Well, the Prussians have got a model and we're going to do the exact same thing. That's how we're going to scale. In 1892 a few people realized you had this Prussian model already all over the country, but it wasn't standardized. In Massachusetts it would be different from what's happening in Georgia. So you had this committee of 10 that literally sat down, headed by the president of Harvard, and decided physics will be their last year of high school, chemistry the year before. You're going to do two to three years of foreign languages. You're going to do geometry. It hasn't changed since 1892. 

reason: Education is one of the last places to experience the revolt against the industrial revolution mind-set of standardized parts and processes. Why is personalization so late in coming to education? 

Khan: Hopefully in the whole sweep of history it won't seem like it's come so late. Almost everything I write about in the book, they're ideas that have been around a long time: mastery-based, self-based learning. There are examples of this being experimented with in the 1920s and in the 1970s, and they actually saw really good results. They were studied, and [two things] made them very difficult to scale: Everyone else was indoctrinated in something else, and they just assumed that that's what education was. The other thing is that in 1920 or in 1970 to do this type of thing, where every student is learning at their own pace and mastering concepts, was a huge effort on the part of the teacher to just coordinate. Hey, you're doing something different [from what] they're doing; how do I keep track of it? They would have to run around with a worksheet and print out things and grade 30 times more things that they would have otherwise had to do. 

Now you have informational technology'"which has been around for 20 years, so you could say it's late, but 20 years isn't that long in the whole sweep of history'"that can now coordinate information. That, coupled with the idea that the barriers to consuming the information have gone to pretty much zero and the barriers to producing the information have gone close to zero. 

Before, if someone like me wanted to go out and make lessons for kids in schools, I would have to go pitch some publishing company. Then we'd go through some process, pitch it to some top-down bureaucracy. Eventually, maybe, I'd have to argue it with people. "Are you teaching it the right way? Are you not?" What gets down to the students would be this weird, watered-down thing. 

Now you can go straight to students. You can go straight to teachers. You can go to parents. All of these things have come together. 

reason: Some of the ideas that you talk about, like flipping the school day, are not even things that would require any real change other than a change in attitude. 

Khan: I write a lot about this, and we've become somewhat associated with this notion of flipping the classroom. One thing I point out is that it's not my idea; it's not Khan Academy's idea. In 2007, 2008, teachers starting emailing saying: Hey, you made some reasonable videos on completing the square or factoring polynomials. I don't think the most valuable use of my time as a teacher is to give lectures anymore. I can have students do that at their own time and pace. And then in class we'll make it interactive. We'll do the problem solving. What used to be homework is now in class. What used to be classwork, or lectures, is now at home. 

reason: The idea is that teachers can help kids work through problems and explain to the whole class, and then later in the day when they're at home or whatever, students can soak in the lectures at their own pace. 

Khan: That's right. That's one model which gets closer to the right idea. But then you can keep running with that. You don't have to say it has to happen at home or it has to happen in school'"you can do problem solving wherever. But you should have some point in the day where humans do get together, where they can help each other. It gives the teacher real-time information on how the kids are doing. And it allows you to go to the next level. Any time you're lecturing, it has to be one pace fits all. Now everyone can learn at their own pace, learn the stuff that matters, and you can start leveraging peer to peer.

reason: One of the worst things that happens now is when people say, "OK, we've got $1 million. Let's throw iPads at everybody." What's the difference between a stupid use of technology and an enlightened use?

Khan: The stupid use of technology'"I use a more euphemistic, more polite way of referring to it'"has been around since technology has been around. Every time a new device comes out people say, "Oh, we'll just use it in the classroom." When I was a kid, [it was] PCs. The computer lab was empty, and we'd go in there and play video games and then leave. Or we'd learn how to put in a diskette or whatever. It wasn't really integrated with the curriculum. You see the same thing happening now. An administrator gets excited about an iPad; all of a sudden we've got to get iPads in the classroom. And they get them there, but what are the kids doing? They're maybe playing games, checking their email, who knows? 

The key is: How do you leverage that thing so that it actually can affect learning? And not only affect learning but also transform what the classroom is all about. Right now when people talk about improving education, they keep trying to stuff more things in there; they keep adding more structure on it. And then they bring this technology, but it's just going to be used superficially: "OK, maybe I'll do a little exercise on this thing." 

What we say is: No, now that you have the technology, now that you can keep track of students in terms of what they know, what they don't know, let's completely rethink the model from scratch. Class time should be about interaction, self-paced. Do we have to separate classrooms anymore? Do we have to separate physics from calculus from chemistry? Do we have to separate Ms. Green's class from Mr. Smith's class? Can we have them happening together, and now Ms. Green and Mr. Smith can tag-team teach? They can teach to each other's strengths, and they can mentor each other. That's what I would call the more enlightened use of technology: actually rethinking the model.

reason: What happens to credentialing in your system? Completing K?12 is signaling to the next level of education that you've made the cut. College even more so. Where are your degrees from? They're not from rinky-dink schools.

Khan: I have several degrees from MIT and another one from Harvard Business School. 

reason: So what happens to credentials down the line when we get into a more individualized, personalized educational experience?

Khan: College is a confusing, muddled concept. There's a learning part, a socialization part, and a credentialing part. The students and parents appreciate the experiential, the socialization parts, but they are paying that significant amount, if you really ask them, for the credential. If you went to students graduating at Harvard and said: "Look, I'll refund all your tuition'"you get all the experiences, all the friendships, all the learning'"but you can never tell anyone that you went to Harvard University." Would they do it? I suspect most will not do it. Which tells you that they were paying for the credential. The experience was kind of gravy on top of that. The universities think that the credential is nice but the main thing they're giving is this experience. So that's a huge transaction'"a huge part of someone's total lifetime income'"where the person buying is buying something different from what the person selling [thinks he is selling]. 

What I believe should happen, and what I believe is happening, is you're going to have decoupling of the learning experience from the credentials. So regardless of whether you went to Harvard or whether you went to the local community college, if you feel like you know something you could go to a third party'"well-established, rigorous assessment, better than what happens at any school'"and prove that you know that thing. And you might have learned it on the job, at community college, on Khan Academy, on EdEx, who knows what it might be. 

If you do that, it also clarifies things for the university. You won't have the strange thing where kids are trying to make sure they get an A+ in a philosophy class so that they can get an interview at Goldman Sachs or at McKinsey or Facebook. They'll be there to learn. If they really care about [getting a] job, there's another route that's somewhat orthogonal to the first one. 

reason: Change is coming to the educational establishment. The conventional public school monopoly is clearly breaking down. Charter schools are breaking out all over the place. People are opting out of the system. They're doing homeschooling. What are the main impediments to increasing the pace of change? To what extent do they reside in the formal structures of school, and to what extent are they shackles in the minds of students and parents?

Khan: I think they are primarily in the mind. There are structural things too. I've talked to teachers, administrators who believe everything I say. They believe it's the way. They look at other schools that are self-paced. But they're like, "I got a state assessment test on this. And I got this calendar that the state has told me to do." So they have to do this in-between thing, this
little dance where they pay lip service to the state, but they recognize that the kids have to learn at their own pace. 

It's this weird transition state. I've actually found very few people disagree with the principles. They say, "This is common sense. This isn't really even under debate." There's a lot of cynicism about change. And there's a lot of inertia and bureaucracy and all of that. But I actually think it's happening far, far faster than people realize. In five or 10 years, it's going to be completely mainstream. And for something as systemically important as education, I think that's shockingly fast.