Khan Academy redux

Khan Academy redux
Photo by Desola Lanre-Ologun / Unsplash

This is another throwback post, this time from February 5, 2013. It was originally posted at the old Casting Out Nines blog at the Chronicle. I didn't look at the date before I picked this one, but I have a knack for finding these throwback posts that were from almost the exact same date as today; this time almost exactly 10 years ago. The "redux" in the title refers to a resurfacing of an earlier post, mentioned in the first paragraph, that still is one of the most-read and most-responded-to items I have ever written. I picked this one today for a reason; go to the end to find out.


The last thing I expected to encounter this week was a resurgence in the Khan Academy Debates of this past summer. Those, if you remember, centered around this spoof video created by my colleagues John Golden and Dave Coffey. My own contribution to those debates remains the single most viewed post I’ve ever published in nearly ten years of blogging. But honestly, I hadn’t thought much about Khan Academy since then — until Monday afternoon.

Dave (Coffey) sent me a tweet alerting me to this whitepaper published by the Pacific Research Institute, a free-market think tank based in San Francisco. “Look at page 14,” Dave said. I did, and found that I was being used as a prime example of a Khan Skeptic. Actually I am the last in a list of skeptics whose skepticism the authors attempt to dispatch. I’m in good company, as Keith Devlin is the first on that list and Veritasium’s Derek Muller is in there as well.

The whitepaper itself seems to advocate a position that schools would be more effective, and students better served, if they were more free from government involvement — more free to innovate and reform themselves, with a flipped classroom approach being the foremost example of reform. I actually do not disagree with this idea. I am on record as being pro-school choice, and I am firmly right-libertarian on basically every political issue — although I loathe the dehumanizing influence of politics and choose not to discuss this here on the blog, or anywhere else — so in terms of the motivations of the authors, I don’t really have any big issues.

What I do have issues with, is the single-minded insistence in this paper that Khan Academy is the exact same thing as the flipped classroom. Throughout, the authors can’t seem to decide whether they are advocating “Khan-like” approaches to school or the Khan Academy itself. Competitors to the Khan Academy, of which there are a a growing number, are never mentioned — which is a strange thing to say about a whitepaper from a pro-free-market organization — and any suggestion that Khan Academy itself might be improved upon is dismissed as “ivory tower pontificating”, especially if the criticism comes from actual educators who, of course, are too steeped in the establishment to have any good ideas.

I have little to no interest in rekindling the Khan Debates of last summer and getting “You’re just jealous of Khan’s success”, etc. comments multiple times. But since my name was brought up in this whitepaper, I thought it would be appropriate to respond.

The section on Khan’s critics starts on page 10 with the sentence: “There is an old saying that no good deed goes unpunished, and so it is with Khan Academy.” This should let you know what you are in for. The entire section is worth reading in its entirety, especially if you’ve been thinking you need more straw-man arguments in your life, but I will focus on the part where I show up on page 14.

The authors start by correctly quoting some of the nice things I had to say about KA in my “Trouble with Khan Academy” post. Then they say:

However, Talbert says the Khan Academy can never replace an actual class on mathematics. The program does not offer a live teacher or human interaction. He further argues that the Khan Academy does not have a real curriculum for effectively teaching students.

The third point is not entirely right. What I actually said was (emphases in the original):

[KA] is not a coherent curriculum of study that engages students at all the cognitive levels at which they need to be engaged. It’s OK that it’s not these things. […] Khan Academy is a great resource for the niche in which it was designed to work. But when you try to extend it out of that niche — as Bill Gates and others would very much like to do — all kinds of things go wrong.

My point in the original post was about KA trying to be a curriculum — a complete one-stop educational resource. The whitepaper authors, instead, think I am talking about having a curriculum. The difference is more than merely semantic. My daughter’s elementary school has a curriculum — a focused course of study that is implemented by the teachers in the school. But the school itself is just an organization. It would be absurd to say that her elementary school is a curriculum.

Khan Academy wants to be a curriculum, and therein lies the problem. The authors of the whitepaper seem to pick up on this and offer, in Khan’s defense, the suggestion that Khan never said he wants to be a complete educational resource:

Khan never says that he wants to replace actual classes on mathematics. He simply wants to restructure them so that students are able to advance at their own pace and receive more individualized assistance. By advocating a switch to a flipped-classroom model, he wants to enhance teacher interaction with students, not minimize it.

But this is either plain wrong or a significant reversal of Khan’s earlier objectives. In the long feature article in Time magazine on Khan Academy from July 9, 2012, it says (emphasis added):

Khan is using the money [from donations from Google, etc.] to transform the academy from his own personal YouTube channel into an educational nonprofit with Silicon Valley start-up DNA. The goal: to create a complete educational approach–with video lectures, online exercises, badges to reward student progress, an analytics dashboard for teachers to track that progress and more–that can be integrated into existing classrooms or serve as a stand-alone virtual school for anyone wanting to learn something new.

I find it hard to square this very public statement of KA’s goals with what the authors of the whitepaper want those goals to be, unless Khan has backpedalled from this ambition since July.

The authors move on to quote from my blog post with some my criticisms of Khan Academy: that good videos are not enough to constitute good education, that what else is needed is the sort of high-level cognitive processing that (at least for now) only takes place through human-to-human interaction with an instructor. They get this right, and I stand by those criticisms. I actually think Khan believes this too, and the authors offer this in Khan’s defense:

Khan stresses, however, that he wants students to interact with their teachers so that higher-order learning takes place. “The promise of technology,” he says, “is to liberate teachers from those largely mechanical chores so that they have more time for human interactions.”

I have said the exact same thing before and even acknowledged that Khan and I are on the same page with this. So there should be no “however” in the authors’ statements. Case in point: As readers of this blog know, I expended over 100 hours of my own time last semester making over 14 hours of video content for my transition-to-proof class to help implement this vision locally with my own students. I did this while maintaining a full-time tenure track faculty position with teaching, service, and research obligations, not to mention daily contact with actual in-real-life students and their educational struggles, and without any external funding. I am not a critic of the idea of the flipped classroom. What do I criticize? I’ll get to that in a moment.

But first it’s worth noting that the authors end with two very strange things to say. First:

Ordinarily, teachers only interact with students 10 to 20 percent of class time in the traditional classroom. Using technology, if that figure went up to 90 to 100 percent of class time, “The student-to-time-with-teacher ratio would improve by a factor of five or ten.”

Did you know that 94.23% of all statistics are totally made up on the spot with no supporting evidence? And then,

Overall, the arguments of critics remain unconvincing when compared to Khan’s documented impact and what Khan truly believes.

Not nearly as unconvincing as arguments based on anecdote and personal feelings, which is what most of the authors’ defense of Khan seems to consist of. I have plenty of anecdotes and personal convictions of my own that go in the opposite direction.

Let me summarize just exactly what it is I am criticizing: It’s the videos themselves and the idea that Khan Academy thinks that a collection of videos plus web-based rote mechanics exercises can constitute a complete educational solution.

As for the videos themselves, I’ve written about this elsewhere. In fact, oddly enough, just this past Sunday afternoon (in a completely independent Khan-related event) I gave this answer to a Quora question on how KA can be improved that spells out my criticism — namely that his videos are buggy, unfocussed, mistake-prone, and quite clearly put together with minimal planning. (The Time article has details.) My point at this post was that Khan Academy videos would improve greatly if they just brought in some people to help out who are fluent in what the research says about teaching and learning, can plan tight and efficient lessons, and can edit video.

Until Khan Academy can address more than just the bottom 1/3 of the Bloom’s Taxonomy pyramid, it is not a complete educational resource. Does Khan realize this? It’s hard to say because we get mixed signals. On the one hand we have the Time magazine article that lays out pretty steep ambitions, but on the other we get Khan and his supporters saying otherwise. I would like to believe the latter, but the former keeps getting in the way.

Strangely, what I find so disappointing about this position paper is that I sympathize with the less-regulated approach to schooling that the authors support, but their single-minded insistence on Khan Academy as the only instance of the flipped classroom and their uncritical defense of his product ruins any chances that their position will be believed by anybody other than themselves. Critics such as myself and the others in the paper are lumped into the single category of “liberal/progressive” (top of page 22) ivory tower dwellers who “caricature” Khan’s work.

I got a good laugh out of being called a liberal, but I didn’t laugh long. Demagoguery is not going to advance the cause of education reform, and I hope the Pacific Research Institute can rethink what they’ve said and try to find some common ground towards the best approach to the flipped classroom, and not just try to defend Khan Academy against all comers.


Updates for 2023

  • The reason this article grabbed my attention when looking for something to repost, is that I'm thinking a lot these days about where extremely affluent people and organizations are putting their money towards improving education. I'm grateful that the Bill Gateses of the world are giving their wealth to this cause. But I'm frustrated that the money is being spent on ideas that are essentially tech-infused versions of the same old things that have been tried for 100 years or more that got us to the state of needing improvement in the first place. This was my main beef with Khan Academy ten years ago, namely that when you boil it all down, it's just a guy giving a lecture, and lectures – while useful tools in some situations – are by all metrics miserable failures in all the places where it counts. So why is all this money being dumped into this stuff, when instead it could be used for funding research and development of all kinds of new ideas? I will be writing more about that in the coming weeks.
  • So it seems like the more things change, the more they stay the same. In 2013 it was all about Khan Academy; in 2017, MOOCs; today the new hotness is AI. What will it be next?
  • To its credit, Khan Academy has done a lot over the last 10 years to become more like a fully-fledged curriculum that actually engages learners meaningfully across different cognitive levels, and not just a glorified YouTube channel. And the videos are better (although Sal Khan still rambles a lot).
  • I was a lot spicier in my blogging ten years ago than I am now! Call it getting older. And as for the politics, I still lean libertarian and I still support school choice, but in my typical go-against-the-grain fashion I have definitely gotten more liberal as I've grown older.
Robert Talbert

Robert Talbert

Mathematics professor who writes and speaks about math, research and practice on teaching and learning, technology, productivity, and higher education.
Michigan