Posts

Comments

Get Posts by E-mail

Benjamin Riley offers two reasons related to cognition and learning why we shouldn’t attempt to personalize student learning. Here’s his second:

This is also why I think it’s a mistake to place children in charge of the speed of their learning, particularly during the early years of their education. If left to decide for themselves, many kids — and particularly those from at-risk backgrounds — will choose a relatively slow velocity of learning (again, because thinking is hard). The slow pace will lead to large knowledge deficits compared to their peers, which will cause them to slow down further, until eventually they “switch off” from school. The only way to prevent this slow downward spiral for these students is to push them harder and faster. But they need to be pushed, which means we should not cede to them control of the pace of their learning.

My own argument against personalized learning is that – in Audrey Watters’ fine formulation – it “circumscribes pedagogical possibilities.” Which is to say, a lot of fun learning in math class – argument, discussion, and debate chief among them – is impossible very difficult when you aren’t learning it synchronously with a group. Riley’s argument adds new dimensions to those concerns.

BTW. I left my own version of Riley’s second argument on Will Richardson’s blog, a forum where the value of student-personalized curriculum is, IMO, too often assumed to be utterly obvious and questioned only by cowards and cranks. Rather than spending his time tangling with anonymous Internet commenters, I’d like to know how a thoughtful technologist like Richardson would engage a critic like Riley.

2014 Jun 24. Mike Caulfield:

I often warn about overgeneralizing across disciplines but let me overgeneralize across disciplines here: if there is one thing that almost all disciplines benefit from, it’s structured discussion. It gets us out of our own head, pushes us to understand ideas better. It teaches us to talk like geologists, or mathematicians, or philosophers; over time that leads to us thinking like geologists, mathematicians, and philosophers. Structured discussion is how we externalize thought so that we can tinker with it and refactor it.

2014 Jun 25. Alex Hernandez writes a thoughtful rebuttal.

27 Responses to “Don’t Personalize Learning”

  1. on 23 Jun 2014 at 7:36 amTed Dintersmith

    I couldn’t agree more with Dan’s comment about the drawbacks of personalized learning. In schools and classrooms I’ve visited, and in the online offerings I’ve observed, the most powerful learning experiences come from peer interaction. The reason so many MOOC’s are producing dismal results is that very little real learning takes place by watching a lecture and then, within minutes, answering multiple choice questions validating “learning.” The online resources can be quite powerful if they’re treated as exactly that — resources for engaged learners. Instead, far too many people have jumped on the big data/personalize learning/”school should be all about sitting in front of a screen” bandwagon, to the detriment of real learning.

    Ted

  2. on 23 Jun 2014 at 8:03 amKenneth Tilton

    If one’s individualization effort looks like “go as slow as you want and learn as little as you want”, it needs fixing.

    The point of individualization is to foster success by not forcing the learner to exceed their hull speed. If one is not seeing that success, a fix is needed.

    Riley need not worry about “at risk” kids not wanting to think too hard. When I taught them they were geniuses at talking themselves out of trouble. :) The worst student in the worst school takes pride in learning if we can arrange for that learning (more below). Tapping that pride is the Grail.

    If students are not tearing into personalized instruction it just means personalization was not enough to save the lousy instruction. Forcing them through that same curriculum at a uniform pace — well, that’s where we are today. So Riley is fixing the wrong problem. Keep the self-pacing, fix the lessons.

    Individualization must also be accompanied by continuing unbending performance standards (“take your time, but you still have to learn it”) and an ongoing search for those other pedagogical fixes needed to arrange for the first spark of learning.

    ie, Individualization alone is not the answer.

    As for Watters’s concern, I am thinking of Algebra, she seems to be thinking about history or literature. I would love to see the better Algebra students not just zoom through the subject but also take time to help others. That will actually make their mastery tighter, because teaching someone else exposes the gaps in our mastery.

  3. […] Meyer has just published a provocative post called “Don’t Personalize Learning,” inspired by an even more provocative post with the same title by Benjamin Riley (as well […]

  4. on 23 Jun 2014 at 10:26 amJessie Arora

    Thanks for short, sweet post. Completely agree with this approach and it is driving a lot of my thinking as I build Embark Labs. Our goal is to create student-centered learning communities where kids can learn computer programming/computational thinking together, in a peer-driven environment. I don’t think it’s very revolutionary since it’s based on significant education research on how people learn and our desire to be a part of connected communities IRL.

  5. on 23 Jun 2014 at 10:34 amAlex

    A brief note – personalised learning doesn’t have to come from personalised lessons.

    I’m sure if you think back to your best lessons, there’ll have been moments of group work where you spot a struggler and take them aside, or find someone sparking and ask them a few thought-provoking questions. They might receive the same ‘primer’ while you stand at the front and speak, but – because people are different! – they’ll have different levels of engagement, and different cogs will whirr in their minds.

    I always took the cry for personalised learning to mean that every student should be encouraged in their thoughts, and addressed according to their moods, whatever they may be.

    Sure, that sometimes means extension work – especially if you’re just working from a textbook. But normally it just means giving them a chance to state their own position and follow it through.

  6. […] a great discussion going on about the myth of personalized learning, both at Dan Meyer’s blog and at Benjiman Riley’s. Michael Feldstein has also stepped into the conversation, pointing […]

  7. on 23 Jun 2014 at 12:40 pmKevin Hall

    Do people really think teachers should relinquish control of the content to be learned in their courses? That’s crazy. Obviously, some kids would prefer to just chug away at easy objectives even if that means they miss out on the deeper stuff. I didn’t even realize there were respected people out there who want to let kids choose everything.

    @Dan, but I’m curious how far you’d push “Don’t personalize learning.” Sounds like you’re saying that we shouldn’t personalize the introduction to a concept, because the intro phase is best when it involves social exploration that can’t be done solo. I agree there, mostly, and I’m beginning to see that the intro phase of teaching a concept is where you put most of your effort as an educator. It’s something I should have noticed before, but it only hit home in our conversations on twitter and my blog.

    How far would you take it, “Don’t personalize learning,” though? Would you also say, “Don’t personalize practice?” That’s what I thought you meant in the Neil Diamond iPod post. If so, here’s the reply I mean to write for that post:

    Non-adaptive practice is like an iPod that lets you buy absolutely any song you want, but it only downloads a random 70% of the songs you pay for…and it can’t tell you which of the songs you bought were actually downloaded. Each week, it randomly replaces 3% of the total minutes of playtime with the sound of static or with minutes from another song…and it can’t tell you which songs are most or least corrupted. My point is that non-adaptive learning tends to proceed from the assumption of “well, I taught it, so they got it”, because without adaptive technology, it’s simply too difficult for a mortal teacher to know which students need to work on which concepts/skills and to deliver the right practice opportunities to each student based on those assessments.

    I know you’re not opposed to using tech for practice. I keep coming back to it because I personally know 0 teachers who think impersonal videos or computer exercises are the best way to introduce a new topic, and yet that belief seems to be what you consistently argue against.

  8. on 23 Jun 2014 at 3:30 pmDan Meyer

    Kevin Hall:

    I know you’re not opposed to using tech for practice. I keep coming back to it because I personally know 0 teachers who think impersonal videos or computer exercises are the best way to introduce a new topic, and yet that belief seems to be what you consistently argue against.

    There are some very ambitious teachers at charter networks here in the Silicon Valley who believe exactly that. Both the teachers and the networks. Start at the bottom of page five in the Khan Academy SRI Report and read about “playlists.”

    But for the most part, I’m not arguing with teachers in these harangues, I’m arguing with technologists.

  9. on 23 Jun 2014 at 5:18 pmDean Shareski

    I wrote about this a few years back. http://ideasandthoughts.org/2011/09/27/the-dangers-of-personalization/ but this makes the added point around community and common experience. I’m pretty careful to discuss personal and self directed learning with caution.

  10. on 24 Jun 2014 at 4:57 amAndrew

    What is the role of educational experts? Would it be possible for the right group of the appropriately trained and experienced people to sit down together and say “At a minimum, all American math students should be learning THIS”.

    Would it be possible for the masses to trust that group? Or the recommendations? I am very much in favor of communities, parents, and students having input into the inner workings of their educational systems, but are there places where it makes more sense for educational experts to make the decisions? What if the focus group is lacking the appropriate background to make those decisions? At that point, shouldn’t we ask that personalization fall by the wayside in favor of trusting the reasonable recommendation?

  11. […] Dan Meyer: don’t personalize learning […]

  12. on 24 Jun 2014 at 9:24 amihor Charischak

    In order for personalization to work you need an inspired curriculum, not the kind that we currently have (based on CCSS) where we add bells and whistles to try to make math more engaging. Most kids are not interested in pursuing a math learning path on their own. My vision and I think it is Will’s as well is that we develop effective student centered curriculums that students buy into. We may lose some traditional math in the process but we will gain more empowered students that really want to learn what they are studying.

  13. on 24 Jun 2014 at 10:40 amIra Socol

    Kind of shocked by the casual colonialist attitude here, not to mention the disregard for brain research.

    Two things: First, though “community cognition” and “social memory” are powerful facts, all learning is individually paced and constructed. Ignoring that simple reality is why 180 years of factory schools have produced colossal failure.

    Second, ‘at risk’ children are no less capable of understanding themselves than the rich, no matter how the above author would like to demean them. In fact my experience suggests that they often possess better critical judgment abilities because that has driven their survival. But it is very handy for those in the power structure to pretend the deficits are the thing to focus on and thus deny the poor real educational opportunity.

  14. on 24 Jun 2014 at 3:25 pmIra Socol

    For Ihor Charischak.

    Kids love math. Just not school math. They work in math all the time. Its just the teaching that’s the problem:

    http://speedchange.blogspot.com/2009/04/real-world-math.html

  15. on 25 Jun 2014 at 6:50 amihor Charischak

    For Ira.

    Most younger kids love non-school math. The problem is not the teachers so much but the kind of curriculum that they are forced to teach. (I call it the Royal Road to Calculus.) Alternate curriculums should be available for students who aren’t keen on math. If they decide later that they really want to be an engineer they can make up for “lost” time with options like Khan Academy.

  16. on 26 Jun 2014 at 12:28 pmKevin Hall

    Dan, I’d like to push back on your post in 2 ways.

    1). I take your main point to be this: “a lot of fun learning in math class – argument, discussion, and debate chief among them – is impossible very difficult when you aren’t learning it synchronously with a group.” I mostly agree with this point, but where you state it as a fact, I think it’s really a value judgment. The reason it’s not a fact is that you actually CAN get kids really excited by badges, energy points, etc. Kids like to make progress and see their progress, just like mastering levels in Angry Birds or something. Although I’ve learned much from your reading of Harel, I also know from experience that that kids can be motivated without feeling epistemic need: you can excite them with gimmicks, too. I don’t think KA blog posts like this are made up. Instead, what I believe (and I think you do, too) is that this kind of motivation is in some way shallower or less edifying. What’s your response? At the very least, even if you don’t think kids really enjoy the gimmicks, do you at least think your claim — that typical “personalized learning” takes the fun out of class — is falsifiable in any way? Or is this an axiom for you?

    2). Secondly, I think Michael Feldstein makes some good points that support using ed tech for personalization. His post (which is a reply to your post) basically says that KA in it’s current form customizes learning based on prior knowledge, but doesn’t really /personalize/ at all. In his usage, a personalized curriculum would relate content more closely to what I’m interested in (to my personality). This is pretty close to the point I made in your “what’s real-world math” discussion. My main point was that our goal shouldn’t be to make math feel like it’s part of the real world…we should try to make it feel like part of /their world/. I remember you making similar points in your concluding comments on the post. Don’t you think ed tech is the most likely way to do this? We could each study the same topic in different contexts, varying by our interests (sports, science, video games, etc).

  17. on 26 Jun 2014 at 1:52 pmBrian Walach

    I think there are merits to doing both.

    Personalized learning at the high school level can be very powerful for the right type of learners. Collaboration and talking over common topics are definitely necessary, but at some point, you don’t need the group for everything. And not everything needs to or should be covered by everyone. Personalization can help with that. I agree that it’s not the be all end all. But setting high expectations and structuring it in the right way could produce an environment where students wouldn’t just lag behind because learning is hard – they might just do better than we think.

  18. on 27 Jun 2014 at 6:19 amJulie

    A sidebar here – I think it’s a mistake to aim toward teaching kids to think and talk like biologists physicists or whatevers. Lets just try to get them to think in ways that they can apply no matter what they are doing. I drink Daniel Willingham’s coolaid at will. It took me a graduate degree and postdoc to even begin thinking like an educational researcher and trying to get kids to do this is when they are 16 or even 22 is misdirected, IMO. Let’s simply get them to think, direct their thinking to what is most important, help them have sophisticated lexicon for talking about subject matter and most importantly help them learn to transfer their knowing and understanding to unfamiliar contexts.

  19. on 27 Jun 2014 at 10:25 amDan Meyer

    Julie:

    A sidebar here – I think it’s a mistake to aim toward teaching kids to think and talk like biologists physicists or whatevers. Lets just try to get them to think in ways that they can apply no matter what they are doing.

    Interesting sidebar. While argument and critique (for example) are skills that transcend context, I have a hard time seeing how teachers are supposed to decontextualize them. Who teaches these generalized skills? Also, there are lots of ways to contextualize those skills without asking students to perform them at a graduate or post-graduate level. Why do two even numbers always add up to an even number? Argue about it! That’s the epistemology of a mathematician, accessible at the fifth grade level.

    @Kevin, a lot of fun learning is possible with adaptive learning software. That doesn’t contradict my point that the software circumscribes a lot of other fun learning possibilities. Which is, as you rightly call out, a value judgment on my part. I’ve never claimed otherwise.

    Kevin:

    Don’t you think ed tech is the most likely way to do this? We could each study the same topic in different contexts, varying by our interests (sports, science, video games, etc).

    I’m actually enormously cynical about the success of this approach, a few research citations notwithstanding.

    The number of [ ipads / points / party guests ] increases with time according to the function f(t) = 2t + 4. How many of [ ipads / points / party guests ] were there after four hours?

    It’s easy personalization but, for me, pretty depressing.

    If I ask students to select their content preference in advance, sure, maybe over a few problems students will be more interested in the problems and that interest may even translate into higher achievement.

    But how sustainable is that strategy? What returns will it pay us over a year? We’re applying different shades of lipstick but it’s the same pig. The problem is the problem. Not the context around it.

  20. on 28 Jun 2014 at 9:32 amKevin Hall

    @Dan, oh, I honestly thought you considered badges & energy points to be absurd. My misunderstanding.

    Regarding my second point, I’m not suggesting just changing the cover stories for drill problems, though that is better than doing nothing. I’m suggesting having kids do different activities as their intro to a new concept. E.g., introducing standard deviation with an invention task asking kids to come up with a measure of dispersion for given data sets. Different groups could do the task with the same or isomorphic data sets in different contexts: sports, movies, etc. In reality, they’d probably invent Mean Absolute Deviation, but that would be a good launch for lecture/notes on standard deviation. My guess is ed tech will have us to this point relatively soon, don’t you think? Even Penny Circle on Desmos could be personalized, I would say. Just put the circle and the filling-in of it in some personalized context (decorating a circular cake with m&m’s, etc). I know you dislike just changing the cover story and saying you’ve improved a task, but I think if you can personalize the context to feel to students like it’s responsive to their interests, that would be a qualitative improvement.

  21. on 29 Jun 2014 at 6:48 pmDan Meyer

    Kevin Hall:

    Different groups could do the task with the same or isomorphic data sets in different contexts: sports, movies, etc. In reality, they’d probably invent Mean Absolute Deviation, but that would be a good launch for lecture/notes on standard deviation. My guess is ed tech will have us to this point relatively soon, don’t you think?

    I do, but I think this is really low-hanging fruit.

    Further, the difference between Penny Circle Desmos and Penny Circle Textbook is so much larger than the difference between Penny Circle Desmos and Personalized Circle Desmos. Sure, we can do both. But our finest minds aren’t anywhere close to delivering Personalized Circle Desmos. They’re delivering Personalized Circle Textbook.

    I see it as the personalized ad experiences on, for instance, Hulu. Maybe some people love choosing a flavor for their advertising. My guess is that most people would prefer a less lousy viewing experience overall.

  22. […] last week’s post knocking around “personalized learning”, Michael Feldstein argued that the term is too […]

  23. […] Dan Meyer, “Don’t Personalize Learning” […]

  24. […] Riley wrote a post criticizing personalized learning. Dan Meyers agreed.I would like to offer a response. First Benjamin criticizes two “definitions.” I will […]

  25. […] Dan Meyer, “Don’t Personalize Learning” […]

  26. […] Dan Meyer – Don’t Personalize Learning  […]

  27. on 18 Nov 2014 at 8:40 pmGreg

    Personalized Learning, like anything, can be done poorly.

    There are dangers.

    There are also dangers associated with other crap and irrelevant kinds of “education” shoved in so many kids’ faces.

    As kids grow older, they need to practice increasing amounts of self-direction. Of course there will be variation between individuals.

    Of course most H.S. students will need the collaboration of an experienced and educated adult to create a personalized curriculum. Otherwise, just send the kid to “unschool”.

    Thanks for the discussion.

    all best,
    Greg

Leave a Reply