My second favorite speaker at TED this year — my first was David Brooks, naturally — was Salman Khan, a New Orleans native who quite accidentally came to realize that he was a gifted teacher. Kim Zetter has written a thorough post on the talk at Wired. While working in finance, Khan started uploading a series of YouTube videos offering basic math lessons to help his cousins do their homework. Pretty soon the videos took off. Khan’s story is a beautiful illustration of Clay Shirky’s concept of the “cognitive surplus.” As a brainy, ambitious MIT alum, Khan was always going to devote at least some of his spare brain capacity to interesting and demanding entertainments, like classical music or watching episodes of a dense, cerebral sci-fi epic like Battlestar Galactica. And as a decent, friendly guy, he was naturally going to help his young cousins. The presence of low-cost (i.e., free) sharing platforms allowed others to access this content he was happy to generate for free, and pretty soon he realized that he had a knack for doing this fun, engaging work.
Khan also realized that he hit upon a serious vulnerability of traditional teaching. As he explained to the TED audience, his cousins actually preferred learning from him via video, because it allowed them to pause and rewind — that’s hard to do in a live classroom. As Khan started getting feedback, he hit upon another key problem with traditional modes of instruction:
He began getting letters from teachers who said they were using his videos in class to great success. That’s when it hit Khan that the standard one-size-fits-all way of teaching that exists in most classrooms doesn’t work and leaves many students, whether they are smart or not, behind.
“In a traditional classroom you have homework, lecture, homework then you have a snapshot exam,” he said. “And whether you pass or not, the class moves on to the next lesson.”
Even the ones who get 95 percent of the lesson correct, still have 5 percent they didn’t grasp, and with each subsequent lesson, the percentage they don’t understand increases.
Because learning is cumulative, the failure to keep up completely can be fatal:
Good students suddenly begin failing algebra and calculus, despite being smart, because they have these small holes, like Swiss cheese, in their comprehension, and the holes keep increasing as the lessons get more difficult.
Khan Academy’s core principle is the following:
The Khan Academy, which he launched with a small team of software developers, has more than 2,000 videos on math, science and economics but now also offers accompanying online lessons that use an interactive approach to allow students to learn at their own pace and stick with problems they find difficult until they master them.
“The traditional model penalizes the student for experimentation and failure but does not expect mastery,” he said. “We encourage you to experiment. We encourage you to failure, But we do expect mastery.”
I find this idea particularly affecting because I was a student who fell behind in math around the 4th grade, and it took me many years to catch up. Khan has been working on how to introduce his methods into more traditional classrooms:
Khan and his team were invited to conduct a pilot project at a school in Los Altos, California using two 5th-grade classes and 7th-grade classes doing the Khan Academy online lessons for half of their time in class. The software tracks each student’s progress and logs that show teachers how long students worked on a particular lesson, where they got stuck, which videos they watched or paused and re-watched.
“When students are allowed to work at their own pace, you can see students who took a little extra time to get through one concept or another, once they get through the speed bump, they race ahead,” Khan said. “So the same kid you thought was slow six weeks ago, you now would think are gifted. And we’re seeing it over and over and over again.
The genius of this approach is that it frees instructors to work on the troublesome 5 percent, or rather to more effectively deploy limited human capital resources. It is easy to see how Khan’s approach could improve educational outcomes in a cost-effective way.
P.S. I should also note that MIT’s OpenCourseWare program, a pioneer in broadening access to specialized instruction, is about to undergo a massive expansion.