The Khan Academy uses sophisticated statistical techniques based on prediction of student performance in the near and further future on tasks related to those which they have already studied on the platform, using information on past performance.
“I think too much conversation about Khan Academy is about cute little videos," Khan said in an interview last week. “Most of our resources, almost two-thirds of [the staff], are engineers working on the exercises and analytics platform. That, I think, is what we’re most excited about.”
....Using math and computer science concepts decidedly more advanced than most of those in Khan’s video library, the Khan engineers have trained the website’s exercise platform how to predict, with startling accuracy, how likely it is that a student will correctly answer the next practice problem — and whether that student will be able to solve the same type of problem a week, two weeks, and a month later.
They do this by accounting for hundreds of data points that describe, in numbers, the entire history of the relationship between a learner and a concept. “If [a user is] logged in, then we have the entire history of every problem they’ve done, and how long it took them, and how they did,” says Ben Kamens, the lead developer at Khan Academy. “So whenever anybody does a problem, we see whether they got it right or wrong, how many tries it took them, what their guess was, what the problem was, how many hints they used, and how long they took between each hint.”
The Khan engineers are also working to tweak the exercise platform so it does not confuse genuine mastery with “pattern matching” — a method of problem-solving wherein a student mechanically rehashes the steps necessary to solve that type of problem without necessarily grasping, conceptually, what those steps represent.
By Steve Kolowich, Inside Higher Ed, December 7, 2011