The Pyramid of Myth

There are many myths in education.  I believe they all result from well-intentioned educators and/or researchers.  However, with these myths, evidence simply does not exist to back up their claims.  A few of the most prominent myths are: learning styles, left brain/right brain theory, and brain gyms.  I have also previously written on the popular statement of edumyth, “Students don’t learn from people they don’t like.”  These myths may appear harmless, but I don’t see it that way.  For example, if students believe they learn via any one of the popular VAK learning styles, that will lead to biases and misinformation in how they learn.  Now imagine if a teacher believes this to be true.  A classroom of students will be affected by the popular myth.  Now imagine if a school’s administration supports this myth…you get the point.  All of these myths are not harmless.  They misinform and incorrectly steer our classrooms.  

In this article, I want to focus on a quite popular edumyth that I frequently see while participating in edchats and even on university websites.  This is the myth of the learning pyramid, the cone of learning, or as a misrepresentation of Dale’s Cone of Experience.  

Here’s a common representation found by searching google images for “the learning pyramid”:

Screen Shot 2017-11-29 at 8.59.31 AM.png

The claims made are quite clear.  Using different methods/strategies of presenting material to students, or students presenting material to each other, results in differing average rates of retention.  On the surface, this seems to fit in with modern education philosophy.  A lot of edchats I participate in and professional development lauds classroom strategies that advocate for a more student-centered or student-led classroom.  This pyramid appears to support that movement.  According to the pyramid, a student will only retain, on average, 5% via lecture but will retain 90% of what they teach others.  That sounds correct.  It follows what many teachers hear about how our classroom should be run.  

But…it is so incorrect.

Think about the statistics involved.  Here are but two examples of how the numbers figuratively and literally don’t add up:

-Let’s begin with retaining, on average, 90% of information by teaching others…how did the student learn the information to be able to teach others?  If he or she learned the information via lecture, they only retained 5% of the information.  So now they’re teaching others the 5% they retained from lecture to now only retain 90% of that teaching to others?  And what of the retention of the student being taught (new learner)?  If the student is telling (audio-visual) the new learner the material, the new learner should only, on average, retain 20% of the 90% of what the original learner retained.  Hopefully you can understand the silliness of the pyramid.  I sincerely don’t understand the logic.

-What about combining methods?  Let’s say a student is teaching another student (average 90% retention) via a demonstration (average 30% retention)…is the student now going to remember, on average, around 120% of the material?  I hear you, I hear you, “No.  They would only retain 100% of the material.  You cannot remember over 100% of anything.”  Ok.  Good point.  I’ll concede.  But…if there’s a method for retaining, on average, 100% of material, don’t you think we’d all be using it in the classroom?  This also seems to imply there’s one, best way to teach in all circumstances.  In my experience, this goes against the philosophies of differentiation and personalization of learning in the classroom.

I hope this proves my point.  But, even if it doesn’t, even if you recognize that my examples are a bit anecdotal, that’s ok.  I agree.  Don’t take my word for it.  Kudos for not accepting information without thoughtful consideration.  Here is a fantastic look at the learning pyramid myth by Will Thalheimer.  You want statistics?  He’s got statistics.  You want research?  He’s got research.  The article, Mythical Retention Data & The Corrupted Cone, goes into much greater detail and provides a history of the cone/pyramid.  

Also, another great resource for mythbusting in education is Urban Myths about Learning and Education by Dr. Pedro De Bruyckere, Dr. Paul Kirschner, and Dr. Casper Hulshof.  Chapter by chapter, the book discusses different myths in education and presents research to dispel the myth.  It also discusses why these myths appear to have a hold on education around the world.  I highly recommend this book.  

Still not convinced?  That’s ok.  You’ll only retain, on average, 10% of this information.  🙂

25 comments

  1. I think your logic gets flawed at “So now they’re teaching others the 5% they retained from lecture.” I would suggest instead that there are multiple ways for students to “construct” knowledge well beyond lecture; in fact, one of those ways is doing something we don’t do when we are listening to or observing a lecture: thinking about and then developing a plan to “teach” the concept based on what we know. Treated as a formative process, this can be a poweful leanring opportunity, especially if it’s done with other learners and not in isolation. My best math teachers are practicing elements of flipped and blended learning, with more of the direct instruction (lecture) shortened into videos that students can access on their own. This is better than a live lecture in one key sense – students can push pause and review if necessary. The real change happens in class however because the real “work” happens togther as students are developing understanding by solving math together instead of alone through homework after daily lectures.

  2. Ken,

    Thank you for the comment. I used the “so now they’re teaching others the 5%…” quote to illustrate the frailty of the pyramid’s logic. I do not, for one minute, believe that statistic is correct. You make a good point…unless explicitly told to do so, students are not thinking about how to teach a concept while listening to a lecture. Students who are taking notes are thinking about and synthesizing information for their notes. Something that has been shown to increase retention of materials.

    I would ask you, how do you know the math teachers who are blending and flipping their classrooms are your best? I don’t mean that to be negative or combative, I would just like to know how you justify teachers being the best.

    Lastly, my classroom is full of retrieval practice and spaced practice…I want my students working through and met with the material as much as possible…after it is presented to them. My class isn’t simply a ‘let me talk for an hour and then give you something to do for homework’ class. We work through and interact with the material as much as possible together, when appropriate.

    Again, I really appreciate the above comment. Thanks.

    Blake

  3. I am completely on your side for busting myths that are not supported by evidence, and this pyramid seems one of them. However, your argument about “so now they’re teaching others the 5%” can be objected to with the following reasoning: we know there are two types of memory, short-term and long-term memory. “Retention” in such contexts probably refers to long-term memory, not short-term. So to play the devil’s advocate, 5% retention from lecture means 5% of lecture becomes long-term retrievable memory.
    But perhaps shortly after the lecture, you can retrieve 80-90% of the material. So if shortly after the lecture, you teach that information to others, you will turn 90% of that short-term memory into a long-term memory.
    Again, I’m agreeing that the myth is myth, but I think this particular argument was not really spot on.

    1. According to the theory, if I teach by demonstration, it’s me, not the person I am teaching, who’s supposed to retain 90%. The person I am teaching should retain his 30%. So it makes no sense to add the 90 and 30 to get 120% retained. Apples and oranges.

  4. This article is irresponsible and is damaging to the conversation around education. I think it should be deleted.

    The fundamental tenets of the pyramid, I believe, are correct – going through the exercise of “teaching” a concept (explaining it to thin air even, a la the Feynman technique, for example), certainly helps ME retain it for the long term. The logic of this pyramid is of course bogus, for the reasons you so emphatically stated, but the message it delivers, is dead on! It’s very short sighted to refute the concepts it espouses because the delivery is poor. The message it delivers is critical for our current education system to understand and integrate.

    Lecture is ineffective, but along with homework, makes up the bulk of how we teach. This gets the info in, which is an essential step as you noted, but doesn’t solidify the info. We must integrate other active forms of retrieval and retention, of which TEACHING is a really great one. Phrasing your understanding of something into a coherent explanation is hard and reveals the gaps in your knowledge. A valuable exercise. Your article muddies the waters on this topic, which may make headline-skimming-teachers disbelieve the value of something that I think is beyond refute.

    Do you disagree that teaching something leads to retention?

    1. “This article is irresponsible and is damaging to the conversation around education. I think it should be deleted.”
      What if your ‘discussion group’ is led by a novice and the 50% retention claim only reinforces errors and misconceptions? We cannot get past the teacher as an expert in a domain paradigm. Just putting a pleasing shape into an infographic does not confer it with veracity.

  5. You put a lot of faith in the article by Will Thalheimer which is, I admit very interesting and makes several good points. However, his thesis is that education can be victim to mischaracterizations of research and the sloppy combination of dubiously-related information (the corrupted cone). He doesn’t actually address the central idea, that students retain more when engaged in tasks beyond simple direct instruction. This article does nothing to convince readers that the fundamentals of the cone are errant.

  6. Effortful Ed,

    The 90% retention has to do with the “generation effect” [1], [2] for which there is plenty of evidence.

    The 90% retention develops DURING the teaching. Nobody says that the teaching involved is smooth, brilliant or linear. The claim made is simply that if you STRUGGLE (aim, work, actually make an effort) to explain it to someone (others, yourself, an imaginary friend etc.) once you figured out how to teach it you have 90% of it in you. You have to include in that the Q&A (so, from that perspective an imaginary friend, or the self, won’t challenge as much as an actual audience). We know very well from experience that preparing a lecture isn’t easy (I am not talking about giving the same lecture you have given time and again for many years, I am talking preparing a good comprehensive and clear lecture on a topic that you have to teach for the first time).

    Think about learning to bike, skate, swim: a lecture gives you 5% or less. Doing it is key to learning it and right up there in the pyramid (75%) with trying to teach it.

    There’s a wonderful book [5] by Saundra McGuire: “Teaching Students How to Learn”. If we think about it literally the title is a true absurdity and the author says so somewhere in the book. The key is to understand that learning is a process. It’s already existing to some extent in the student (the ability to learn, I mean) we just need to hone/teach it. The most successful ways for a learner to learn are: experimenting (that is, learning is a contact sport, so actually doing it is a key step, in order to internalize it and eventually own it) and trying to teach it to others. These are the last two levels in that pyramid.

    McGuire [3] [4] asks at some point: what is harder, to ace the test or teach a review session for the test to your classmates? Trying to ace the test is not as daunting as trying to answer all the questions in a review session. Students need to assume they will be in some shape involved with teaching the material (next year as a teaching assistant would be the easiest think to imagine) in order to be effective learners. It’s only when you try to teach someone that you reach the highest levels in Bloom and Fink taxonomies. Without some form of metacognition learning is inevitably shallow (the relation is inversely proportional, that is, the more we tap into metacognitive practices the deeper our learning is).

    Hope this makes sense. All the best,

    Sincerely,
    Adrian German

    [1] https://www.alleydog.com/glossary/definition.php?term=Generation+Effect
    [2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3556209/
    [3] https://www.youtube.com/watch?v=XdW49Ir0c6o
    [4] https://www.youtube.com/watch?v=Z40sDnbn5ms
    [5] https://www.aacu.org/resources/stem-higher-education/mcguire

  7. Not sure how this blog works. I suppose you need to approve my comments. In any event with respect to the comment that I just sent, I see you are a big believer in and proponent of retrieval practices. That’s what happens when you try to teach. It is not productive to assume someone should try to teach a subject after just having attended the lecture (5%) and that would magically result in a sudden and astonishing rate of 90% retention. My understanding is that the pyramid’s levels can’t be skipped and each level is accessible only from the previous level (at least that’s how I see it). In this regard teaching alone gives you a net 15% (from 75% to 90%) but that 15% is not accessible by itself, outside the 75% accumulated by going up through earlier levels: lecture, read, demo, visual, recitation/lab etc.

    Take it easy, all the best.

Leave a Reply