Moving the Goalposts on Personalized Learning

Mike Caulfield:

But the biggest advantage of a tutor is not that they personalize the task, it’s that they personalize the explanation. They look into the eyes of the other person and try to understand what material the student has locked in their head that could be leveraged into new understandings. When they see a spark of insight, they head further down that path. When they don’t, they try new routes.

EdSurge misreads Mike pretty drastically, I think:

What if technology can offer explanations based on a student’s experience or interest, such as indie rock music?

Mike is summarizing what great face-to-face tutors do. They figure out what the student already knows, then throw hooks into that knowledge using metaphors and analogies and questions. That’s a personalized tutor.

But in 2016 computers are completely inept at that kind of personalization. Worse than your average high school junior tutoring on the side for gas money. Way worse than your average high school teacher. I don’t think this is a controversial observation. In a follow-up post, Michael Feldstein writes, “For now and the foreseeable future, no robot tutor in the sky is going to be able to take Mike’s place in those conversations.”

So it’s interesting to see how quickly EdSurge pivots to a different definition of personalization, one that’s much more accommodating of the limits of computers. EdSurge’s version of personalization asks the student to choose her favorite noun (eg. “indie rock music”) and watch as the computer incorporates that noun into the same explanation every other student receives. Find and replace. In 2016 computers are great at find and replace.

This is just a PSA to say: technofriendlies, I see you moving the goalposts! At the very least, let’s keep them at “high school junior-level tutor.”

BTW. I don’t think find-and-replacing “indie rock music” will improve what a student knows, but maybe it will affect her interest in knowing it. I’ve hassled edtech over that premise before. In my head, I always call that find-and-replacing approach the “poochification” of education, but I never know if that reference will land for anybody who isn’t inside my head.

I'm Dan and this is my blog. I'm a former high school math teacher and current head of teaching at Desmos. He / him. More here.


  1. These points of view look at personalisation from the tutor’s/teacher’s perspective. From the student’s perspective, if a number of representations are available, a varying amount of time is available and, if different learning modalities are available then a student can select, a la cart, how much time and in which way she can assimilate those various representations and make connections within her own sphere of experience. This is student directed personalisation. The teacher is one (possibly multiple) modalities available to the student but, unless (and even if) it’s a personal tutor, the time, representations and modalities are limited. Computers can provide different (possibly multiple) representations and modalities that teachers cannot provide (desmos?) and, of course, computer tutors never get tired.

  2. This post hits home with me nicely as I am always optimistic that technology can help students learn mathematics much more effectively than it currently does. I agree that the tech cannot do what a good, human tutor could do. I am still optimistic that a well-designed system could assist in helping students make some knowledge and understanding gains to help fill gaps along the way. My hope is that we can find a way to make boring, knowledge based math practice more engaging and less painful in order to free up more class time to explore depth of knowledge style tasks that model the beauty of mathematics.

    Pipe dream? Maybe.

    I hope not, though.

  3. As a Tier 2 and Tier 3 math interventionist at the high school level, I need to run 3 or 4 individualized programs simultaneously. I do not believe that technology can effectively personalize explanations for struggling students, but it can be a piece in the puzzle to filling in gaps when a teacher or tutor are not available. Mike stated that the biggest advantage of a tutor is that they personalize the explanation rather than the task. I’ve found that many teachers struggle with this concept in regards to differentiation and personalization. Specifically, my experience with my colleagues has proved that a good number of teachers still believe that differentiated instruction is changing the task, when it should focus on changing the method of instruction. With the opportunity to sit one-on-one with my students, that ability to take the calculated steps to provide personalized explanations is why I do what I do–because it works and is effective.

  4. @Sadler, this all sounds great in theory. But where is the evidence that computers in 2016 are anywhere close to realizing your vision?

    Kyle Pearce:

    My hope is that we can find a way to make boring, knowledge based math practice more engaging and less painful in order to free up more class time to explore depth of knowledge style tasks that model the beauty of mathematics.

    I’d only question the premise that knowledge-based math practice has to be boring. If that’s the case and computers can do as good a job as humans at supporting students in that skill development, then sure. Let them have it. A good place to look for a counter point is Boaler’s Fluency Without Fear piece.

  5. “Mike is summarizing what great face-to-face tutors do. ”

    And how many great tutors are there?

    Bloom/84 famously found a two-sigma win for quality, full-time private tutoring, but VanLehn/11 found only a little under one sigma because of serious deficits in tutoring, such as tutors who simply lectured in the one-on-one settings.

    I think this echos @alise’s concern with the variability of teachers’ approaches to differentiation.

    The good news is that VanLehn found roughly the same efficacy for computer tutors, a bit under one sigma, who cost a lot less and are available 24×7.

    I agree with @sadler that the ideal is a tireless automated platform that let’s the student choose from multiple activities how to sort out their confusion. Shucks, the simple act of choosing engages and empowers the student, whereas activities chosen by a teacher or adaptive system leave them passive and uncommitted.

    In any situation with struggling learners, getting their buy-in is the first hurdle to pass.

  6. Brendan Appold

    May 20, 2016 - 6:41 am -

    This all assumes that the student is motivated to learn. Let’s not forget that one of the biggest challenges in K-12 education -especially for at risk youth – is motivation. A consistent tutor/mentor who is able to build a relationship with a student and his/her parents can go a long way on this front.

  7. Admittedly, Andreas Schleicher’s OECD report on the use of technology doesn’t show it as a glowing success.


    However, slide 54 gives some indications as to why technology isn’t as effective as it might be. Lack of skills, lack of instructional design, resistance to technology and low quality of software and courseware.

    On the other hand EEF have commissioned a study of flipped learning in 24 schools using Khan Academy.


    While the full results have still to be released (later this year) I understand that the model has shown a range of advantages not least whole levels of improvement in attainment. I hope they release the results soon.

  8. But how does the high school junior know what skills the student needs to practice? You’re right that you need a human to do the tutoring, but you need a computer to keep track of what topics need more practice. Many students can’t articulate where their skill gaps lie, so the tutor ends up just helping them with tonight’s homework even if there are other skill gaps that are vastly more important. With current technology, the best configuration of resources is something like: kid uses adaptive software to identify skill gaps, while human tutor helps kid progress through those modules.

    It’s especially important to remember that only a computer or a VERY dedicated teacher can really track what students have forgotten from Ch. 3 now that we’re on Ch. 4. And in my experience, the lower the ability of the student, the more retention issues outweigh learning issues.

  9. Kevin Hall:

    With current technology, the best configuration of resources is something like: kid uses adaptive software to identify skill gaps, while human tutor helps kid progress through those modules.

    Maybe. Another system has the students writing down their standards-based grading scores on a log in their notebook, then pulling that paper out when they need help. I’m not making any claims about relative efficacy or easy, just that the problem isn’t as intractable as you make it out to be.