Khan: Free learning, cheap credentials

Khan Academy founder Salman Khan talks about his new book, The One World Schoolhouse: Education Reimagined, in an interview with MIT Technology Review. Instead of the Prussian model — students march in lockstep through the curriculum — Khan believes technology will make “mastery learning” practical.

Everyone advances at his or her own pace. Don’t try algebra until you know your arithmetic. Spend less time in lectures and more in hands-on problem solving.

Most students can be motivated to learn, if they can go at their own pace, Khan says. “The core reason for students disengaging is that they are frustrated. They’re in algebra class but don’t have a good foundation in pre-algebra or arithmetic.”

Khan Academy is “investing heavily” in analytics, says Khan. “What does a student know? What does a student not know? How effective is the tutorial?” In elementary and middle schools using Khan in the classroom, teachers are very enthusiastic about the real-time learning assessments — more so than the videos.

Online learning will revolutionize higher education and liberate students from ever-rising college costs, Khan says.

Here’s what I think it could look like in five years: the learning side will be free, but if and when you want to prove what you know, and get a credential, you would go to a proctoring center [for an exam]. And that would cost something. Let’s say it costs $100 to administer that exam. I could see charging $150 for it. And then you have a $50 margin that you can reinvest on the free-learning side.

If students can earn credible credits by taking free online classes, the college cartel will be broken, writes Jeff Selingto at the Chronicle of Higher Education. Now universities often reject transfer credits, claiming the quality of instruction doesn’t match their own, he writes.

. . . what happens when students arrive at the registrars’ office with credit-bearing courses from professors at Stanford, Penn, and Princeton? What will the excuse be then to reject the credits—that the courses were free? Such an excuse might finally expose the true reason many colleges refuse to accept transfer credits: They want students to pay them tuition for a class . . .

It all depends on assessment. If there’s a credible, cost-effective way to measure learning, then everything changes.

About Joanne


  1. Your last paragraph really captures the key element that could change things. Arguably one area that has had some success with this is reading the law. In some states, such as California, it is possible to apprentice to a lawyer without going to law school, and then take and pass the bar exam.

  2. Roger Sweeny says:

    “It all depends on assessment. If there’s a credible, cost-effective way to measure learning, then everything changes.”

    But it is ed school dogma that there is no credible, cost-effective way of measuring learning.

  3. (Joanne): “It all depends on assessment. If there’s a credible, cost-effective way to measure learning, then everything changes.
    One additional critical element: reliable instant ID verification. Agencies that administer tests for credit or certification must have access to a mechanism that reliably identifies the test-taker.

  4. (Roger): “it is ed school dogma that there is no credible, cost-effective way of measuring learning.
    Let’s take them at their word. Then there’s no way to establish that they add any value to Education majors or to the children their graduates instruct. Therefore…

    • Therefore, ed schools don’t have to demonstrate that they add any value because they’re the gatekeepers. Don’t get the ed school stamp on your passport? You don’t get entry to Teacherland.

      That privilege would be harder to hang onto but the customers for the product ed schools produce – that would be school districts – are also indifferent to whether ed schools add any value. With apologies to the pithier original, to paraphrase a bitter joke from the Soviet Union, ed schools pretend to educate teachers and school districts pretend that’s important.

    • Roger Sweeny says:

      Ed school dogma also says that a successful year or two in an accredited ed school program by definition adds substantial value. It does not have to be measured.

  5. Crimson Wife says:

    I suspect that credentials that will actually be valued in the job marketplace will cost more than $150. The Chartered Financial Analyst exams cost ~$1100 for level 1, $1200 for level 2, and $1500 for level 3. Still way cheaper than a traditional MBA, though.

    • Florida resident says:

      Dear Crimson Wife !
      The main credential of a Harvard-educated person is
      _not_ that he (or she) graduated from it, but that this person was _admitted_ to Harvard in the first place, under the conditions of extreme _selectivity_.
      If on-line courses will remove this selectivity aspect, then the value of corresponding credentials will become very low.

      • Is Harvard really so selective? The diversity imperative undermines this assumption. Unless an employer wants what Harvard admissions officers want, why not use raw SAT scores or high school GPA that “selective” implies and tests of subject-area competence? Admission to Harvard may mean no more than a stoned high school experience, decent SATscores, a college B average in fuzzy studies (Did anyone ever say “I flunked Black Studies 335 so I switched majors to Physics”?), and a (claimed) exotic origin.

        • Yes, Harvard really is so selective. At least statistically.

          75% of the students at Harvard has SAT reading scores of 690 or above. 75% have SAT math scores of 700 or above. This works out to most Harvard students being 2 standard deviations above the average for IQ. And my guess is that the ones in the bottom quartile have other traits that Harvard likes. Some of these a company may not care for, but others might be nice to have.

          This filtering/binning is desirable :-)

          “why not use raw SAT scores … tests of subject-area competence?”

          Because the tests of subject-area competence are illegal :-)

          Graduating Harvard signals a few things:
          (1) On average, quite bright.
          (2) On average, willingness to work fairly hard
          (3) On average, willingness to play by the rules

          For lots of companies, these are very desirable traits. You do want to know that the kid got in, but you also want to know that they finished. And it *IS* legal to ask for the major. Decent SAT scores plus a physics degree from Harvard signal pretty well.

          Ivy League SAT Scores:

      • Crimson Wife says:

        Open-access online courses paired with rigorous exams to earn the credential would solve the selectivity issue. For example, only a tiny fraction of those who start the CFA program pass all 3 levels. Each exam has only about a 1/3 to 2/5 pass rate and they test progressively more difficult content. Any college graduate with long enough relevant work experience can sit for the exams but few manage to pass all 3 levels.

        • Yes, the CFA tests wash out a significant percentage. I understand that few of those who fail level II ever complete./pass the series. I have 3 kids with the credential and their employers value it enough to reimburse costs for those who pass. It is valuable even for those with MBAs from top-5 programs, too. BTW; all levels may be taken and passed, without the relevant work experience, but certification is not granted until the work experience is completed. That was the case for two of my kids.