Always consider the Lesson Behind the Lesson
Johannesburg
Joined July 2011
- Tweets2K
- Following434
- Followers360
- Likes2K
Pinned Tweet
I currently believe that the greatest opportunities of AI in education are in enabling the best of online learning (adaptability, mastery, pace, etc) in an offline environment with teachers and peers, allowing students (and teachers) to get the best of both worlds.
I haven't seen many places developing this, in part because EdTech leaders are constitutionally and economically pushed toward the scale of online learning software.
The quoted tweet explains the main elements. We are building something that incorporates these things at ASE. Always looking for friends/partners (and we need them!)
Good question. Several things:
1) I do think there is a role for adaptive procedural software, certainly in maths and possibly in science. (Eg math academy)
2) Analysing transcripts and filmed lessons to provide guidance for teachers and teacher-coaches, allowing teachers to get more coaching and more targeted coaching more quickly. Right now good coaching is so time consuming and expensive many teachers in most systems don't get much.
3) Marking and analysing of student work to provide feedback faster and allow more deliberate practice in an offline environment
4) Analysing and sorting student data to give consolidated feedback on student progress to students, parents, and teachers to inform instruction and other interventions. This type of coordination has long been considered the ideal but it is really difficult to manage across teachers and subjects today.
I want to clarify why I think this is particularly bad pedagogy.
It is not because it is technology or LLM generated, or because it puurports to be (though isn't, really) "DI". And I think this example is instructive.
Normal force has a simple definition but is a challenging and counterintuitive concept. It is very easy for a child to remember the definition (though they are likely to eventually forget it) without understanding at all what normal force is.
We can debate whether this understanding is best built through an explicit program or a problem based program or an inquiry based program. But what does appear clear to me is that the type of instruction found here with dovery little to build a student's understanding of normal force.
I have loved Who's on First since my dad played me the audio when I was like 8.
Am I a comedian (I don't think so, but that would be a nice compliment) or a member of "the very old" 😵⏳🪦?
Dewey agreed with @MartinCothran about this one, at least:
“Natural science is recommended on the ground of its practical utility, but is taught as a special accomplishment in removal from application. On the other hand, music and literature are theoretically justified on the ground of their culture value and are then taught with chief emphasis upon forming technical modes of skill."
Still...I'm not so sure. I find that nothing has made me better able to appreciate and derive value from literature than trying and failing to write worthwhile literature. Just as nothing has improved my understanding of and appreciation for basketball than trying and failing to become great at basketball. Same for soccer. Same for chess. Same for mathematics.
(I am mediocre at a lot of things it turns out!)
When I was one-and-twenty
I heard a wise man say,
“Give crowns and pounds and guineas
But not your heart away;
Give pearls away and rubies
But keep your fancy free.”
But I was one-and-twenty,
No use to talk to me.
When I was one-and-twenty
I heard him say again,
“The heart out of the bosom
Was never given in vain;
’Tis paid with sighs a plenty
And sold for endless rue.”
And I am two-and-twenty,
And oh, ’tis true, ’tis true.
I am comfortable stating that I was adamently (and unpopularly) against every single one of these.
And I still think there is a role for technology in education.
This is good advice not just in math but across the board
8 failure modes to watch out for in math learning resources, that will tank your learning efficiency:
Failure Mode #1: Not enough scaffolding.
You periodically run into situations where you bang your head on a wall thinking "how the heck did they get from here to there?" and it takes a long time to figure out what kind of logical leap is happening (if you figure it out at all).
Failure Mode #2: Doesn't track your knowledge and make sure you've mastered the prerequisites for anything new you're supposed to learn.
You often feel a large gap between your level of knowledge and the new material, which leads to more banging your head on a wall trying to figure out what prerequisite knowledge you're missing and how to learn it.
Failure Mode #3: No spaced review.
You quickly get rusty on a lot of what you learn, which not only means you come out of the course having forgotten a lot of content, but even during the course, you're constantly forgetting prerequisites.
Failure Mode #4: Doesn't adapt to your level of performance.
You waste a lot of your time doing the wrong amount of work. Sometimes you grasp a topic quickly and end up doing way more practice problems than you need; other times you struggle with a topic and don't do enough practice problems to reach mastery.
Failure Mode #5: Leaves the definition of "mastery" open to interpretation by the learner.
As a learner, it's hard to know when you've mastered something well enough to continue moving forward.
You often think you've learned something well enough, when you actually haven't -- but you won't know unless there's an expert who is evaluating your knowledge.
On the flipside, you can also take things too far being a perfectionist, spinning your wheels on the same topic for a week over some minor point that doesn't make perfect intuitive sense to you, when it would be more productive to just keep moving forward and solidify your understanding by building on top of it.
Failure Mode #6: Not enough successful problem-solving experiences.
In a typical college course, you might solve 50-100 homework problems all semester. There's so much educational friction that it takes you 20+ minutes to struggle through each problem, often not even getting it right until you give up and look at a solution or get help at a TA session.
But if the problems were broken down and presented in a more finely scaffolded sequence, and each problem only presented once you've mastered the prerequisites, then you could get through problems much faster, say, just a couple minutes per problem on average.
That's what a good tutor would do: scaffold your learning experience so that you're solving a problem every couple minutes. They would build up your learning in bite-size increments. You could get through 10x as many problems that way, with a much higher success rate.
Failure Mode #7: Not enough knowledge audits.
Most college courses have only a handful of exams throughout the entire semester (and even grade school classes seldom have more than one quiz per several weeks).
But quick, frequent timed quizzes -- say, 15 minutes every couple days -- are a powerful way to engage in retrieval practice and build automaticity while simultaneously identifying weaker areas in need of additional practice.
Failure Mode #8: Not enough targeted remediation.
It's rare to find a resource that gives you additional practice after you miss a question on an assessment. At best, you might take it upon yourself to review the questions you missed.
But it would be far better to complete a battery of additional problems like each one you missed, until you're able to successfully and consistently solve those problems -- and then you'd want to evidence that knowledge on a quiz retake (with different problems of the same types).
I would contrast this with the @Dale_Chu post doing the rounds saying "actors use scripts," etc.
Some lessons, like this one, are not conducive to a script because we are asking non-rhetorical questions and we don't know what the students will answer.
So what do we do with that? One answer is to say, "we avoid that type of lesson, because it is unpredictable and therefore can't be tested."
But I think that is an impoverished and scared way of thinking about learning. These unpredictable learning moments have been at the core of the most admired education programs across cultures for millenia, and for good reason.
Throwing them away because they can't be scripted and because their impact is harder to standardized and measure is misguided thinking.
If your friend had been driving drunk, would you report him?
In Plato's Gorgias, Socrates argues that we should bring our friends to justice if they engage in wrongdoing. When we read it with students, we work to understand his argument and then apply the question to their own lives. They already know something about friendship, trust and betrayal. They have reasons to care about the answer.
We certainly don't tell them what moral conclusion they should reach. But all positions must be defended with reason and evidence. The guide asks questions to tease out inconsistencies, and students begin questioning one another when they perceive weaknesses in an argument.
This also gives them something worth writing about.
Before they sit down to write an essay, they have taken a position, heard objections and tried to respond. We still provide direct instruction on essay structure, but now the student has a morally loaded topic on which they have strong opinions and have rehearsed some of the argument in dialogue.
This is also why we get so little AI-written work. By the time students begin an essay, they care about the question and have something of their own to say.
This was my grandfather's favorite comic. He was a mine sweeper.
I don't disagree that I should read more novels in general, including war novels.
Some of my favorite war novels include:
Adichie, Half of a Yellow Sun
Vonnegut, Slaughterhouse Five
Heller, Catch 22
O'Brien, The Things They Carried
Hemingway, Farewell to Arms
Shaara, Killer Angels
I'd absolutely put Adichie up with any of those by any measure, thematic or philosophical depth or character development or realism or the beauty of the prose or entertainment value or whatever other criteria might be included in defining a great novel.
I'll say that my favorite war book is not a novel at all, it is Up Front by Mauldin. But that's certainly because it was my grandfather's favorite war book (and favorite book in general; he wasn't much of a reader), and when we read it together he would laugh and tell me stories about Anzio.
Why I take the "surprisingly" as a compliment:
"If I know your sect, I anticipate your argument. I hear a preacher announce for his text and topic the expediency of one of the institutions of his church. Do I not know beforehand that not possibly can he say a new and spontaneous word? Do I not know that, with all this ostentation of examining the grounds of the institution, he will do no such thing? Do I not know that he is pledged to himself not to look but at one side,--the permitted side, not as a man, but as a parish minister? He is a retained attorney, and these airs of the bench are the emptiest affectation."
-Emerson.
(You could replace preacher and minister in the above quote with with traditionalist or progressive or technologists or whatever term and it would describe twitter quite well)
Jay Kloppenberg retweeted
Schemas operate on many levels. They were both behind the switch to connectionism (Rumelhart), cognitive linguistics (Lakoff/Fauconnier), and expertise studies (Ericsson). I think we have a good idea about the 'shape' of what schemas are but only a very tenous grasp of how they are formed. The basic lesson from both connectionism and deliberate practice is repetition to strengthen neural connections but exactly how it happens is something we can guess and marvel at.
This makes sense on the surface, but my issue with it is that it hand waves the schema formation process.
Forming conceptual understanding is not merely an automatic function of gaining enough facts.
Saying "understanding relies on memory" is not the same as saying "learning is memory." The first is well backed up by the research, the second is a logical leap. The word "well connected" in the second sentence of this tweet is doing a lot of work.
(Unless, as I have said elsewhere, we mean something different by memory. If memory means any change in the brain, then yes, of course, saying learning is memory is a tautology. But if we mean the colloquial definition, that memory consists of facts you remember, then this statement goes far beyond the research.)
The processes by which we form complex schema and develop our own ideas are often different from the processes by which we learn isolated facts.
We need both.
At the end of the day, learning is memory.
Understanding amounts to memory that is well-connected and deeply ingrained.
If someone is "just memorizing" as opposed to "truly understanding," it really means they haven't stored enough information in memory.
This is more than just semantics.
If you don't realize that learning is memory, then you won't realize that the most effective way to learn is to use memory-supporting training techniques.
It's easy to get confused, thinking:
"Truly understanding something is different from just memorizing it, so learning doesn't require memory-focused techniques like active retrieval practice and spaced review. Those are about memorization, not true understanding."
And if that's what you think, then you'll likely shirk the hard work required to build memory, use fun/easy but ineffective training techniques instead, and end up not actually learning much.
I live in D. It's amazing but I don't need 2 weeks all expense paid.
C is the best region on earth for animal lovers and I am an animal lover, so my head says that.
But my heart says E.
In my two weeks: I'm returning to Casamance, one of my favorite places I was not supposed to visit. I'm partying in Abidjan, which I've always wanted to do. I'm seeing the national parks of Gabon, the lowland gorillas and the forest elephants and the primates. And I'm hitting up Lagos for a night or two because, I mean, come on.
Sao Tome probably doesn't make the cut. Accra definitely doesn't make it (sorry, Ghana). Cameroon is a tough cut as well. Luanda, sorry.
And if I'm feeling really adventurous, I fly to Kinshasa and take a cruise up the Congo River. Ohhhh, baby.
This, I think, is another important issue with RCTs that I failed to mention.
Every class is a subpopulation, and every student an even smaller subpopulation. What works for a general population is helpful to know, but what I really care about as a teacher is what works with *this particular student* who is in front of me.
Sometimes I can do better for this student than what is generally best for a standard population. If I can, I should do so.
"Novices learn best from explicit instruction" is a robust (though far from universal) finding.
But what is a novice? In American educational discourse, students are often considered perpetual novices, because each math class, for instance, introduces a new fact or procedure.
That student=novice appears so obvious it is rarely questioned. But this is less a fact than a perspective, a choice.
Japanese and Singaporean math classes do not treat children at perpetual novices. Neither do Socratic seminars. Neither do basketball practices for 10 year olds.
In fact, rarely is a child a true novice at whatever you're doing with them. My daughter is 9 years old and was asking me this weekend about exchange rates, and we got into a conversation about international monetary policy. Even there, she had background knowledge to build on!
And that's before you start talking about something like literary analysis, a field in which she already has maybe 100 hours of experience (what is a discussion of a Disney movie from the perspective of the villain, if not a form of literary analysis?) which itself builds on maybe 1000 hours of reading. Again, not a novice.
Knowledge matters. But students rarely have no knowledge of a subject, just as teachers rarely have perfect knowledge of it. That fact provides opportunities for unscripted discussion and mutual learning that are the foundation of meaningful human relationships.
RCTs have value but they are overrated in education. Education is a highly dynamic system. Isolating a single variable feels right but it can be misleading.
Consider a similarly dynamic and interconnected field, cooking. You can isolate a single ingredient and study if it tastes good or bad, but that is mostly irrelevant when cooking a dish. Corn flour tastes terrible on its own but that doesn't mean it's not useful. Salt is perhaps the most important ingredient in all of cooking, but too much of it will make your dish inedible (and eventually kill you)
Even if Explicit Instruction is salt (which it might be! I think it's really important!), that doesn't mean it's all you need.
Sometimes I feel like I'm reading people on twitter who went from "season your food!" (which is good advice) to "let's all eat nothing but salt."
larrycuban.wordpress.com/202…
"As an historian of teaching practices, I have written about how teachers decade after decade have combined both teacher- and student-centered instruction in both in elementary and secondary school classrooms."
Perhaps it is because I am not a product of teacher preservice programs in America, I'll grant that.
But I have heard more about PFT over the course of the last 20 years I've been involved in education than any other single study ever conducted.
To call it "buried" is very confusing to me.
(I will say that most people get the main takeaway wrong, still, and the current discussion is no exception)
Compare, for example, Hattie's meta-meta analysis of education research (do give us as broad as view of the field as possible) with PFT. That DI can be effective is not a disagreement between the two.
But Hattie says "basically everything works," it's just a question of how much. PFT says that basically nothing works. The most effective single system found works about half the time; the rest were worse.
To me, that implies two things:
1) it calls into question the validity of much of the research that formed the meta analyses
2) it hints that perhaps multiple methods used intelligently for the right things (as great teachers have done for centuries) is more effective than any one method
I can't do much with the first one, except approach research I read with some skepticism. I find the second insight quite useful in learning design, though.
As a former resident of NYC, it is wonderful to see an NY mayor who so deeply understands what makes NY special. And to see any American politician who understands what makes America special.
New York City, wetin dey sup? It was a joy to commemorate 34 years of the Nigerian Day Parade alongside the many Nigerian New Yorkers who fill our city with life.
From the entrepreneurs whose businesses strengthen our neighborhoods, to the sounds of Afrobeats pouring out of graduation parties in Southeast Queens, and the aroma of Agege bread calling out to passersby in Brooklyn — the impact of Nigerian New Yorkers is felt across the five boroughs.
Our city is richer in culture, stronger in community, and more vibrant because of your contributions.
