The problem is that many of those who choose to homeschool or 'unschool' are not looking to do anything near this; instead, they want to buck the system in a way which does not create a usefully educated human--usually in a very narrow siloed way (e.g., religion-focused).
"Bloom's 2 sigma problem refers to the educational phenomenon that the average student tutored one-to-one using mastery learning techniques performed two standard deviations better than students educated in a classroom environment."
https://en.m.wikipedia.org/wiki/Bloom%27s_2_sigma_problem
We just keep doing it this way because now it's a cargo cult.
Policy inertia is always a factor in adopting techniques that appear better in limited testing, but we also do it this way because 1:1 instruction with mastery learning techniques is more than 30 times as expensive as 30:1 instruction with classroom techniques, and because education sits as #2 behind healthcare (ahead of things like debt service defense, and non-healthcare entitlement spending) in overall public spending in the US, and if it was multiplied 30 fold it would be over 150% of GDP, which is...problematic.
It's never been controversial that being educated by an expert personal dedicated tutor is better than classroom learning, but it just doesn't work at scale. Unfortunately, the success at the challenge laid out in the article you reference of finding group learning techniques that achieve the same benefits, in practice, has been less than successful.
(And, no, unschooling/homeschooling isn't a general solution here.)
https://news.ycombinator.com/item?id=41500932
Considering how much more effective tutoring is, we do it this way not because it's the best way to provide free child care (it isn't and doesn't even cover a whole work day). It's not the most effective way to educate students. Nor is it even the most cost effective way to teach students.
I'm left with the belief that it's either a cargo cult or a program designed to make a servile, uncreative population.
No, we keep doing it this way because the public refuses to actually engage the real challenges of actually educating a nation. Comments like this one being excellent cases in point.
"It only takes about 50 contact hours to transmit basic literacy and math skills well enough that kids can be self-teachers from then on. The cry for "basic skills" practice is a smokescreen behind which schools pre-empt the time of children for twelve years and teach them the six lessons I've just taught you." [1]
Unless you have some strong evidence to argue why "sit down and lecture" is the most cost effective way to educate a population, I will retain my belief that it is a cargo cult at best.
Can you also substantiate the claim that 50 contact-hours is sufficient for students to become "self-teachers" and that they actually do that?
I have volunteered at a school that allows students to become self-teachers and it is very effective. They follow the Sudbury model, and students entirely teach themselves everything from reading to mathematics. They graduate from college at higher rates than traditional school students, make more, and have very high levels of self reported happiness. It's a really cool model, I wish it was more common. The school I helped at had no teachers, only 3 adults for 120 young people. They could all come and go throughout the day as long as they signed in and out. It's been up and running for decades, so it's not just a flash in the pan.
See:
A long term side project of mine is to build a neighborhood that can sustain a model like what you’re describing so I’m always looking for inroads into the practice of it. I’ll dive into the Sudbury Model but keen for real world examples as well!