Hacker News Reader

Comment by mchusma | Hacker News Reader

This is an odd/strawman take. Personalized education in the form of tutoring is consistently the biggest intervention in education, don't have the definitive studies handy but there are many (here is one https://edworkingpapers.org/sites/default/files/ai20-267.pdf).

I would suspect almost nobody (Gates, Kahn, etc) suggest having an entirely asocial education experience.

As we can figure out how to replicate the effects of tutoring with technology, we should see massive improvements in education outcomes.

View on HN

Tutoring is not personalised learning. It's personalised teaching.

The success of tutoring is the ability of the tutor to completely understand where the child is at, to figure out what the blockage is and to work at unblocking that specific thing.

For example, if the student "knows something", but it happens to be false, then spotting that, and removing it, is the key to unlocking the future learning. Or identifying a weakness in times tables may unlock the ability to do more advanced algebra.

A good tutor doesnt really teach new things. They pave the road already taught, critically filling in the potholes and removing the speed-bumps.

> A good tutor doesnt really teach new things. They pave the road already taught, critically filling in the potholes and removing the speed-bumps.

You've confused tutoring with remedial tutoring.

FWIW, when I was doing tutoring, which was not “remedial,” I found that my expectations of what was needed of me were kind of completely backwards, and what students actually needed could be better likened to athletic coaching. They needed

- a cheerleader who could celebrate success,

- a sounding board for “how could I do this better” when they brought it up,

- for me to back the fuck off when they don't, because however it makes sense in their head is valid even if it doesn't make sense in mine,

- someone who recognizes “come game-time (exams), I won't be in the room, so you'll have to be able to do this without me,” and thus focuses them on adequate results over and over again, “is this the right answer, can we check it quickly, can we see anything about it that makes us understand the problem in a simple way now” etc.

- and honestly just that “pairing pressure” to not fall behind, to keep skills in practice.

I found that “teaching new things” had to be confined to only like 10% of a given session, if that. The job really was “paving the road already taught,” except that's kind of not the right metaphor because roads are centrally planned e.g. by the teacher, it's more the road currently used, which corresponds often but not always to the road the teacher preferred to teach.

It sounds like the tutoring you were doing was supplemental to your students' primary source of education. That's probably mostly what the parent was referring to as remedial tutoring, as opposed to an education primarily delivered through tutoring.

good stuff, sounds smart, thanks for sharing. i'm going to be raising my boy in a pretty competitive school district and have been trying to come up with a plan for tutoring, definitely going to steal these ideas.

Yes, I picked up on that too. It's one of those phrases that sounds clever and insightful on first read, but you realize it's just incorrect if you think critically.

Actually, a tutor can be someone who plans and guides your learning process, as you described. But there is nothing to stop you as the student from inverting it, designing and planning your own leaning goals, and then going to a tutor for specific advice.

I do this for foreign languages. I plan out everything and my own syllabus and goals, and hire a tutor for speaking practice. I give instructions and context to the tutor so they can best help me.

Well, it is a fact that Zuckerberg (and Gates') personalized education projects didn't work. Replicating the effects of tutoring with technology was exactly what they tried to do. If there were such massive gains to be had, why didn't they at least manage to capture some gains?

You don't have a theory to answer that. The article writer does. In their view, learning is a social activity, and kids don't like to sit in a room with dozens of others each with their own personalized algorithmic "teacher".

That's not really all that surprising. If it was only about personalization, it would be easy enough to do that with far simpler technology than modern computers.

The article writer's theory also neatly and immediately suggests an explanation for why personal tutoring still works, when automated custom tutoring manifestly hasn't so far. That's because it is social. Even if it's just you and your overworked homeschooling parent, you care because they care.

Personalization isn't key. Personalization from a tutor works because it's hard and costly, and convinces the student that this matters.

That's just a theory. It's not necessarily right, but it does present a convincing explanation for what we observe, and in particular the observation worth repeating, that no one can seem to get algorithmic personalized education to work, no matter how rich and clever they are.

And from my personal experience with teachers (including my parents) I have often heard from them that the challenge isn't explaining, it's to get the students to care.

Well said, and it rhymes with many of the mistakes we have made in the past. We discount the social, the emotional, the storytelling... Often they're what matters.

It also rhymes with common parenting folk wisdom that having "the wrong friends" is detrimental to grades.

>the challenge isn't explaining, it's to get the students to care.

This.

The reason not one student learned French in my four years of mandatory French classes is that we didn't want to know french. We didn't know any french people. We didn't care about french songs or movies.

OTOH, when telenovelas became a local fad, every girl had a clique of Spanish speaking 13 year olds.

The actual problem in that case is: How do you teach french/philosophy/whatever to a class full of kids who don't want to learn that stuff? Maybe they want to pass exams and get grades. Maybe you can rejigger discipline mechanisms to motivate them... but that is never going to get you to good.

I'd reword your French example a bit. I had the proverbial four years of high school French which I cared about and performed at probably about a median level (which is to say fairly well) of my high school subjects in general.

Especially in a pre-Web world though, absent pestering my parents to arrange some sort of exchange student situation--which would have gone well beyond my level of caring for any other subject and may or may not have been practical--I was probably never going to get to a high level of French proficiency without some sort of immersion.

After about age 10, children are more influenced by their peers than by parents or teachers. If friends aren't interested in learning then they won't care either.

https://www.psychologytoday.com/intl/blog/a-matter-of-person...

Not yet, anyway. It might be possible with an AI that modelled human psychology. I don't think that's an impossible goal. In fact I wouldn't be surprised to see it happen within 10 years (climate change and war permitting.)

The problem is that people like Gates and Zuckerberg are fully paid up members of the "You can't let people use technology without exploiting them" group.

So AI-assisted personalised teaching would come with a cost - a corporate learning and skills profile of the pupils.

Which would inevitably be sold to other corporates.

From a market point of view that kind of information would be immensely valuable for hiring and HR.

But it would be it would be apocalyptically disastrous from the point of view of personal freedom and political accountability.

And no doubt the service would be tiered, so a minimal unsophisticated model would be available to almost everyone, while more sophisticated and effective models would require a paid subscription.

Teaching is one of those professions where you can see what cultures really value. In the US it's clearly not open access to high quality education.

Automated tutoring won't change that unless cultural values change too. Without that, it will just recreate the current situation with added social costs.

> Personalization isn't key. Personalization from a tutor works because it's hard and costly, and convinces the student that this matters.

> That's just a theory. It's not necessarily right, but it does present a convincing explanation for what we observe

Far from it. We observe that people who already believe that they need to learn something do better with individual tutoring.

> and in particular the observation worth repeating, that no one can seem to get algorithmic personalized education to work, no matter how rich and clever they are.

People got algorithmic personalized education to work quite a long time ago. It's called a book. The algorithm is "when the student asks about something, tell them". And it works very well, just like it always has.

What people aren't getting to work is forcing students who don't want to know things to know those things anyway.

Nothing beats good old fashioned one on one tutoring.

You don't need fancy foreign math syllabuses or newfangled gadgets

https://en.m.wikipedia.org/wiki/Bloom%27s_2_sigma_problem

You do need one tutor for each student, though. While students can tutor each other in some cases and while being a tutor can itself be a great learning experience, in many educational contexts one-on-one tutoring is hard to scale. Hence the persistence of large classes, and hence the continued search for technological solutions.

Not quite, in my limited experience, just one hour tutoring per week makes a huge difference. If this generalizes well we could have more of a 1 to 30 ratio.

> Personalized education in the form of tutoring is consistently the biggest intervention in education,

> As [soon as] we can figure out how to replicate the effects of tutoring with technology, we should see massive improvements in education outcomes.

As soon as we can figure out how to replicate social context and emotional support with technology, ...

Even homeschooling falls into the personalized education bucket, and the data I've seen seems to indicate that parents with no educational background are actually able to outperform public schools.

Personal anecdote here: my homeschooling experience, with a curriculum that accelerated learning in certain areas, and had plenty of data showing that their students outperformed public school students, did not properly prepare me for college or the “real world.” This is ignoring the religious indoctrination that was behind the whole of it. Even though my wife is a teacher, we chose not to home school our children because of my experience with it.

Though it sounds like you two could do a great job at it and improve on the shortcomings of your experience easily

> ignoring the religious indoctrination

Although this is correlated with homeschooling, the movement is much larger in 2023. Also remember the alternative is state or corporate indoctrination. When I think about all the ridiculous ideas and fads I was taught in public school, maybe it wouldn't be so bad to leave childhood with some stories about Moses and Abraham instead.

I am with John Stuart Mill when it comes to education: Free choice for everyone, with the state setting the minimum threshold when it comes to curriculum and quality. Modern US homeschooling does not comply with that idea, as there are no standards today.

> as there are no standards today.

This isn't true. Home schoolers in many states still have to pass state standards test - the only check of quality on the public school.

And empirically the standards are higher if you measure performance on ACT/SAT.

You're definitely not alone. I feel like I can spot someone who has been homeschooled from 100 yards away.

I know what you’re talking about, but I don’t think the confounding factor of who and why home schooling parents choose to home school can be easily dismissed.

The simple fact that social activity needs to be fostered in a different, but not necessarily worse way, alone means a minority approach is going to cause problems, not to mention that fringe parents are likely to pass their fringe qualities on, especially when doubling down with their child’s education.

They are the ones who haven't been tricked into spending all their money on social status symbols.

Yeah totally. Instead their tricked into spending all of their money on tithing, or so socially awkward that they can't get a job to make money in the first place.

Interesting, because I think I can too, but for a positive reason. They stand out to me as the ones more immune to groupthink, and when I encounter them in public, there's something about their style or level of interest and engagement that separates them from the government-trained masses.

Yeah I more mean their literal inability to hold a conversation or dress themselves.

I am completely in sync with your point. Bit I feel there are multiple (different) goals in play here, and the outcomes vary greatly based on the goal chosen.

For example, covid showed that a primary goal for many is simply day care. Child has to be looked after while parents work. If they actually learn things that's a bonus.

Prep for college, and college admission, is another goal. Learn how to effectively take tests, how to memorise somewhat-random material, how to write an essay. Learn to analyse reading material and discuss Hamlet's indecision.

Or Teach the child how to learn. Let them follow what interests them. Explore the world. Visit museums, art galleries, build a bicycle. All great, some useful, but none of the rigidity and discipline required when entering the work-force. (Ironically, probably the most learning, while at the same time the least empowering.)

Or what about trades? Wouldn't it be useful if the school taught plumbing, or how to wire a plug, or how to fix home appliances, or how to paint a wall?

So don't feel bad about being home-schooled. You will have gained some head-starts, and missed out on others. (There are plenty of kids who went to school, and that also did little to prep them for college or the work-place).

The best prep for today work-place, is, well work. A summer job, or weekends at MacD will help you with that skill set. (Or your first couple jobs after college).

To prep for college you need a college-aimed curriculum, coupled with strong analytical, summary, and essay writing skills. You need to learn "what's being taught" regardless of what it is. (Which is why the best college advice is to pick a major you enjoy, because you'll be depending a Lot of time with it.)

Yes, we all had different school experiences. Some good some bad. To some degree we all succeed or fail at life in spite of, not because of, school.

Could be that parents who are able to homeschool have the financial means to not have to work. Kids from affluent places tend to do better relative to those with lesser means. Solving this is very challenging, but there are clear correlations between socio-economic classes when it comes to education.

I've never thought of tutoring as an "intervention", rather as complementary, but considering how ineffectively some subjects were taught to me in the classroom, I wouldn't disagree that it'd be a useful complete intervention.

If you don't have access to a tutor, it's basically up to blind luck whether you're able to learn from the resources available.

Could we reproduce the effects of tutoring by getting rid of all the government contracts for school paraphernalia like smart boards and hiring lots of teachers while giving them reliable access to basic necessities?

Sometimes teachers actually want non-basic necessities like smart boards. We just need to give them what they want, not prescribe to them yet a different approach.

I mentor a robotics team that’s attached to a set of high school courses. One of the best things I’ve seen when it comes to bureaucracy is that the three teachers who basically run those courses were handed a budget and bought exactly what they needed. A mitre saw, 3D printers, a plasma cutter, etc.

No?

A smart board is $3000. Another teacher is $66,397.

You can buy 20 boards per year for the _salary_ (ignoring benefits and other non-payroll costs) of a single teacher.

https://usafacts.org/data/topics/people-society/education/k-...

And? One has a negative effect, and the other a positive one.

Also interesting that $100m would have bought 1,506 teacher years. It makes me think that philanthropy doesn't work as well as some would make out. This is another example of rich people wasting money on things they know nothing about. If it had been paid in tax it could have been spent on more teachers by people who do know what they are doing. This would have had a real benefit for kids even with some degree of inefficiency and waste.

The closest large city to me is Oakland, which spent $750m on schools last year:

https://oaklandside.org/2022/01/12/oakland-schools-are-facin...

So this cost 13% of one years spending in a mid-size district.

Presumably there are a number of fixed costs in the $750m. That means the extra money might have disproportionate benefit.

> This is another example of rich people wasting money on things they know nothing about. If it had been paid in tax it could have been spent on more teachers by people who do know what they are doing.

On the other hand the rich person has a strong incentive to succeed - avoiding ridicule and getting more recognition. The public employee on the other hand gets no recognition, or ridicule, and is spending someone else's money.

If we could combine the incentives of the rich person with the knowledge of the public workers that understand the programs that work, we could have something on our hands.

> On the other hand the rich person has a strong incentive to succeed - avoiding ridicule and getting more recognition.

I would argue that Zuckerberg couldn't care less what the world thinks of him. Otherwise he would have been rather embarrassed by the Meta project. I would put Elon Musk and X up as another example of a very rich person who seems unperturbed by ridicule.

>The public employee on the other hand gets no recognition, or ridicule,

So politicians don't have a need for recognition and to avoid ridicule?

> people who do know what they're doing.

What they're doing isn't working. Even they would tell you there's a results crisis. Children aren't learning. They certainly aren't becoming well-prepared for adult life in an efficient manner. They're barely even being supervised and kept out of trouble. The school-to-prison pipeline is getting worse, not better.

These people may "know what they're doing," but they also know what they're doing isn't accomplishing the goals they have for it.

Knowing how to do something that isn't working isn't a benefit, and is certainly not a reason to give them money.

This thread says multiple times that experts agree that education would be better with more teachers.

That suggests 'they' do know what they are doing, they are just denied the resources to do it

They've been adding teachers for ages. It's still not working.

Agreeing on a solution when the results don't bear out that that solution actually solves the problem doesn't make them right.

Where are we going to manifest more teachers from? In the current labor market, there are a lot more openings than job-seekers. You can always get more people into a specific job by raising wages, but that mostly that’s just a bidding war that redistributes the labor shortage to whoever can afford the least. It can raise labor force participation, but there is not so much juice to squeeze there. Kind of like how higher home prices don’t fix a housing shortage.

Historically the main way that labor productivity has been improved was through technology. Without that you get Baumol’s cost disease.

You could use the people that are currently wasting their time on Zuckerberg's project, since he is laying them off anyway.

In fact you could tax the money off Zuckerberg and then he would have to give up on other ridiculous personal projects like the multi- billion dollar metaverse shambles. That would free up loads of people who have wasted years of their lives as a billionaires play thing

Lower gatekeeping requirements would solve it. Higher wages would solve the problem locally. Keeping gatekeeping requirements and increasing supply by paying for training and creating more training spaces. A % of education budget could be allotted to this yearly.

Many school districts with teacher shortages have already eliminated gatekeeping requirements. They will hire new teachers with just a relevant Bachelor's degree, then pay them while they obtain a proper teaching credential.

> Could we reproduce the effects of tutoring by getting rid of all the government contracts for school paraphernalia like smart boards and hiring lots of teachers while giving them reliable access to basic necessities?

Very unlikely.

We are short of good teachers as it is. Adding more teachers basically means you’re pulling from the bottom of the barrel, so overall instruction quality will decrease.

Plus, lots of people don’t actually want to be educated, so no sense in broadly spending money on those folks (large percentage of the population).

> Adding more teachers basically means you’re pulling from the bottom of the barrel,

Ask the good teachers what would make the job more attractive to them.

I can make wild guesses...

* Better pay. (Seems 100% obvious to a techbro forum.)

* Better administrations.

* Better career flexibility.

* No colleagues of the crappy variety.

* Less curricula mandated by politicians.

* Students taught well in prerequisite classes.

* Families helped with difficulties that translate to behavior problems.

* Covid sensitivity (e.g., no forbidding mask use in a classroom of germ factories during a pandemic).

...but you really have to ask the good teachers, not guess.

I would quit my job in an instant to become a high school teacher if the pay was better. But more than I want to teach, I want to be able to support my family. I save most of what I make in my tech job, and I'll look at becoming a teacher 20 years from now after the kids are on their own.

> I'll look at becoming a teacher 20 years from now after the kids are on their own.

If you're leaning towards being a teacher, wouldn't it benefit your kids more to teach them now, while they're still young?

I’m not the GP, but them becoming a teacher isn’t likely to have any particular effect on their own kids, right?

It’s best for their own kids if mom/dad makes good money and can pay for things like private school or tutoring.

Same. Unfortunately, I don't have four years to get a teaching degree, let alone a master's degree, let alone $50,000 to pay for one.

> Unfortunately, I don't have four years to get a teaching degree, let alone a master's degree, let alone $50,000 to pay for one.

Depending on what your undergrad is (preferably STEM), there are plenty of districts that will hire you, let you teach on a provisional license with a crash course, and pay for you to get your teaching qualification.

See this article for some examples:

https://edsource.org/2022/what-you-need-to-know-to-become-a-...

In CA, the term of art to look for is “intern” or “internship” (even though the pay is either regular teacher pay or close to it).

    if the pay was better

What is it in your school district? What level do you need?

I was a teacher for a while and the main fustration we had was class size. Large classes (maybe more than 20 pupils but 15 was really nice size) were really hard to control and engage whereas smaller classes were always a pleasure.

If you're looking reduce class sizes by hiring more teachers, you'll get further by scrutinizing the admin headcount and salaries.

I would like to see a citation, because that sounds like right wing propaganda to me.

All I can cite is my wife's boss, a teacher and very left wing, who nearly got fired for talking about it.

I've been talking with him about maybe gathering the necessary data (it's available upon request but the district records people aren't exactly a sane API, you get PDF's with hundreds of pages). If I do that analysis to confirm his claim, I'll send you a link to it.

But even without the budget information, we can just look at headcount. What ratio of teachers to admin would you consider unhealthy? How many assistants to the vice principal do there need to be before it starts to look like a retirement program for people who can't be bothered to teach anymore?

Can anyone explain how smartboards work, who creates the software, etc. My kid is in pre-k and he likes smart boards but i suspect that’s because of very limited screen time at home. Are smartbords detrimental in any way having that they’re not excessively used? I wish I could see them myself but am not allowed in the school.

I think good tutoring should/does demonstrate the standard. It shows us what education can do, when the limiting factor is (ideally) the students' ability to learn instead of resources available to teach.

Idk if any technology, institutional framework or other fix can ever achieve that standard. They don't necessarily have to. Its still useful to have a comparative standard.

There will always be a difference between learning to (say) pole jump & hammer throw by self coaching, vs training under an olympic coach. You can, however, use that coaches' methods and standards to teach a cheap, recreational class.

The mechanism is complex, but recreational field athletics is better (farther/higher) by existing in a world where olympic training exists. They'd be even better in a world where those trainers coach middle aged normies, to remove the "youth and talent" factors.

I suspect, for example, that small groups or 1-v1 can comfortably pace a primary school curriculum with a small fraction of the desk hours. That in itself is a "performance gap" with potentially actionable crossover.

In learning, like in every activity, motivation plays a big role. Competition and approval among peers is a huge motivation for kids and young adults, far more than remote career earnings plans their parents have in mind. That component lacks in online or at home education.

When I see that total amateurs (ie, parents) can home school their kids more effectively than school in a fraction of the time, I just don't understand why we can't improve traditional schooling more easily.

> total amateurs (ie, parents) can home school their kids more effectively than school

This is a loaded statement. I'm sure many parents are great at home schooling their kids but there are many who are also terrible at it and home schooling shouldn't be the solution. Parental involvement is always better for educational outcomes and often times the solution is for parents to just be more in the loop with their kids' education. Unfortunately, oftentimes, parents decide to completely change things and send their kids to a private school or homeschool instead of minor course corrections with how they engage with their kids' education.

Reply on news.ycombinator.com