Khan Academy: Criticism as an Email to my Colleagues

(Dan Meyer’s take on Khan Academy)

Last night 60 minutes did a story on Khan Academy (video here, text here). A member on my staff emailed the links to all staff, recommending the videos as a good resource for students.  Admittedly I have not watched said video yet, but judging from what came across my twitter feed, it’s the same basic Khan story. As a result, I jumped on my [email] soapbox with the following reply;

[Start of email]

I totally agree that the videos below are a good resource for students.

However, Khan Academy as a larger concept scares me. Seriously scares me.

When I first learned about KhanAcademy about a year ago, I was very excited that this excellent resource was available to my (and any) students. The problem is that Khan [and others] want to take something that is a good external resource and turn it into the main method for student learning. In a Khan school, students would walk in, sit down at a computer, and watch lectures over and over, then regurgitate the information in computer-generated problem sets. There is SO MUCH research out there that lecture in general is NOT the best way for students to learn. Over the last 7 years of teaching physics I have shifted from 90% lecture 10% learning activities to about 80% learning activities 20% lecture. Lecture is a great way to summarize what students have learned and pull concepts together. But, to be honest, I in no way believe that any class should be taught with 100% lecture. It just really isn’t how students learn, per the research and my own experience. Specifically in physics, I have found that through lecture they can memorize Newton’s 3rd law, but they cannot correctly apply it to new scenarios unless they have experienced how it works. This is consistent with research; lecture works if all you are assessing is memorization and basic skills, but not if you want students to use higher-order reasoning skills. They need to be able to think for themselves.

The concept of Khan schools is to reverse the last 20 years of solid research on constructivism and make force kids to learn traditionally, through lecture. His ‘assessments’ in math test basic algorithms and don’t get into complex problem solving or analytical thinking. Do students do better on that kind of assessment when they do Khan Academy? Sure. But do they actually think better, become better problem solvers, better analytical thinkers, better citizens? I think not.

Much of my thoughts about this stem from a wonderful blog written by a seriously awesome physics educator, Frank Noschese. He has become a leading voice nationally on criticism for Khan, along with Dan Meyer, an inspirational math educator. Dan gave a nice, sarcastic, analogy here. Frank has a number of nice posts; You Khan ignore how students learn, Khan school of the future, and here is a post comparing constructivist learning with Khan’s methods, even using video.

I’d be happy to dig up some of my research on constructivism, but I must warn you that most of what I know specifically relates to physics education. I know there is a body of info out there supporting constructivism in other subjects, I just haven’t delved into it as deeply.

Sorry for the treatise. This is currently one of my hot-button issues (I know, I know, I have many of those…)


[End of email]

As a side note, I am not at all against the flipped classroom. I have actually done some flipping myself, when I thought the topics at hand (conversions, etc, in physics) warranted the flip. We used the extra class time to practice conversions in the context of some cool nano-science activities.  I do, however, think Sal Khan has taken what could be a good thing (the videos themselves as a resource for students) and turned it into something awful (a complete instructional program dominated by kids sitting passively at computers).

I welcome your thoughts.

***Update***  Dan Meyer, as always, has great things to say about the 60 minutes piece.


13 responses to “Khan Academy: Criticism as an Email to my Colleagues

  1. As a seminary student, having had a mixture of online lectures and in-class lectures I can say that my engagement is maximized when I become a participant rather than an observer. Students have a stronger appetite and desire to learn when they invest themselves in the subject matter. What ever we can do to facilitate student participation, the better.

    I will say that lectures are helpful but only as a cursory to actual integrative, long-term learning. In my opinion, lectures are easily forgotten. If forgotten, has any actually learning occurred?

    It is tragic that Khan homogenized their program to be purely computer-based. When it comes to learning potential, context is just as important as content. Relationships with teachers and peers is part of being a life-long learner.

    Thanks for the post Casey. Definitely have more to discuss on this topic.

  2. Your paragraph on lecture reminded me of this post by Stephanie Chasteen A time for telling via @fnoschese.

  3. Yah, why bother watching a video when you can just ask for the hints to show you how to do one problem, and then do ten more just like it – makes perfect sense. I think the videos are reasonably good resources, but the problem sets are just awful.

    That white paper that Dan referenced is disturbing, both in what is described, and what seems to be a relatively positive attitude towards the experiment. In all seriousness, most of the problem solving by the students seems to involve finding a way to get 10 problems in a row correct without actually understanding the content.

    Keith Devlin has an interesting post on a very well known study from the 1970s where a student progressed through a series of worksheets, apparently showing mastery, but on closer inspection showing almost no understanding at all.

  4. You bring up some great points regarding methods of teaching and as a college student myself I would have to completely agree about learning by doing. One area I would possibly contend your criticism is with his innovative visually stimulating lectures. This is not merely text book and chalk board. His methods remind me of the innovation of the smart board in classrooms across America. I would say, for what it is, Kahn academy has well connected new with old.

  5. Khan is not actually kids sitting passively in front of a computer. This is what media like 60 Minutes show on KA segments to emphasize that the website is a part of the program. But really, a lot of classroom time is spent on peer tutoring and project based learning. See some projects here…

    The bit about KA as a complete instructional program is not entirely accurate. KA (video and exercises) is done at home and partly in school. The rest of the school time (major amount) is spent on projects like those described in the link above. These projects come from diverse sources like PBS and other mathematics instructors uploading content to the web. Other projects are devised by the KA team.

    • Thank you for saying this. Too many people are missing the point. Representing KA as a complete education program, or referring to a “Khan school” as a place where students will just sit in front of computers all day shows that some people have not taken to the 5 minutes to read the “about” section or watch the videos where he very clearly describes his vision for the future of education, with examples of the programs he has already been doing with kids.
      If I were a math teacher, I would be excited that so many of the things I hate about the job can be automated and I can concentrate on the art of guiding learning rather than the gruntwork of grading tests.

      • Why, when you go to practice, the exercises are something straight from a bad worksheet I could have taken directly from an old-school textbook? Why, when I see any footage whatsoever of students working in Khan Academy, do I see them sitting at computers working on said exercises? Part of my point is that Math Education in general seems to be broken…students can ‘cross multiply and divide’ but they have no idea why it works (The other day I asked my physics students why it worked and they couldn’t tell me). Students don’t have number sense; they don’t understand what dividing means, and they can’t apply math to anything other than what strict algorithm we have shown them. ‘Exercises’ that have them regurgitate information without actually understanding why it works short-changes students. My problem with Khan Academy is that I don’t see it fixing this larger problem; in fact, I see Khan increasing this problem due to repetitive, algorithmic, lecture-based methods. If we want students to be able to use algorithms and use math as magic, then I’m sure KA will be great. If we want them to think mathematically, we need to change a whole lot more than just the means for which lecture and practice takes place.

  6. I think the real issue is that Salmon Khan did not come from the “establishment” educational system, which is mafia-like in its desire for absolute control. Your commentary really illustrates how the present US education system is broken to its core.

    I was a scientist/engineer/entrepreneur and in my retirement teach math at a public high school. I have been extremely impressed with the Khan academy, and use it as part of my classroom. 100% of my math students are in 90 percentile on ACT in math. Khan works . . . get over it.

    • “Khan works…” Sure, for improving test scores. I’m simply not convinced that is what is best for students. Can you do synthetic division? Great, here’s a badge. Now solve a ‘real’ problem that can’t be assessed via multiple choice.

      I am very open to change…I have done a little bit of flipping the classroom myself. However, I am very selective of what topics can be flipped. If students must be able to solve a particular type of problem a particular way, then flipping is great (I did it for conversions in Physics). Larger problem solving skills? Conceptual development? I don’t see this (Khan, Flipping) as a solution. I could be convinced, but I need to see some evidence.

    • Boris Korsunsky

      I think it’s kind of said to hear a “scientist” say “100% of my math students are in 90 percentile on ACT in math. [ which is a proof that] Khan works . . .” It’s like saying that you’ve never been in a car accident because you eat tomatoes every week.

      There have been NO studies that show that using KA is better than using many other resources… and I seriously doubt that there will be any, ever. KA is like Newt Gingrich’s campaign – if it weren’t for a couple of wealthy backers, you wouldn’t have heard of it.

      Anything that is “good” about KA, does NOT require KA: peer tutoring, extra work, reading ahead before discussing in class. KA is simply a giant with the feet of, well, Gates and Schmidt…

      We should all remember that forty years ago, TV was supposed to “save education” – at least, that’s what the commercials promised. There is no reason whatsoever why a bunch of poorly made videos (or even *well-made* videos, like MIT ICW and many others) will save education.

      Mr. Jackson is clearly excited and concerned about the “ed. mafia”. It is this disdain for the teachers, the idea that “everyone can do it”, that is the biggest hindrance to education in the US, not the lack of poor videos… I wonder if Mr. Jackson would want some “former entrepreneur” without a medical degree perform a heart surgery on his child so that the “med. mafia” will not have their say. Probably not – I am sure he thinks that medicine, unlike teaching, does require skills, training and knowledge.

      Well, keep watching those videos, Mr. Jackson… but do go to the doctor when you are sick and go to a teacher, not an ex-financial analyst, to teach your own children.

      Does KA has a video on “correlation vs. causation”, I wonder?

  7. The problem you describe is a vexing problem for math teachers, and not just the Khan Academy. A student masters something like factoring and quadratics, but when they get to science class, and see something like V^2 + mpV=4, they simply can not solve it. They do not recognize it as a quadratic, and are clueless as to how to solve it. If it were written as aX^2 + bX + c = 0, they know how to do it. If you tell them it is a quadratic, they can solve it, but they are not able to recognize that they already have the math tools needed to do it. As an example, Ohm’s law is V=IR. Have students build a circuit, apply varying currents, measure voltage and then plot results. Ask them to determine the slope of the line, and they will give you the correct number, but ask them what it represents, and they can not see that the slope represents R. They do not see the relationship between y=mx +b, which they know and V=IR. Also, they learn algebraic manipulation in math, but when they get to science they can not isolate a variable. Since I teach math and engineering, and I get my math students in my engineering classes, I have no one to blame. One of the things that appears to help is that when teaching math not keep using (X,Y), but change up the variables so they learn that the technique is not associated with specific letters. Also, introduce more of the science into math class to start building that skill of recognizing to apply the tools they have.

    That being said, what I love about the Khan Academy is that it forces students to master a skill before moving on. I have all my classes on Khan Academy, and give them various incentives to complete Khan lessons. I student at the Junior level can start at lesson one in Khan and with a little work get through the entire program in one school year, working on it in the background, in addition to normal math class work.

  8. Hi- We met summer 2012 at the Hamline Phys4All training. I teach Physics First and AP Chem at Chaska HS. I really appreciate this blog- so many people I know are “flipping” their classes but it has been hard for me to articulate why it bothers me so much, so thank you.

    I am also wondering if you know if any modeling training in the area that I could tap into, or if it is only done in the summer? I really want to take my physics classes in this direction.

    Thanks again,

    Liz Houtz

    • Hi Liz!

      I don’t know of any modeling training during the year. Typical courses are 3 weeks long, though Winona State has hosted a 1 week course the past couple of summers, Nathan Moore is the contact at Winona. I went the last two summers (one week for Mechanics and one for selected topics beyond Mechanics), and it was a great workshop. Better than great, actually!

      Thanks for checking this out, and take care.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s