In addition to my other comments, I’d like to ask what your overall goal is for grade-school education. (It seems to me that most of your arguments are about grade school and not college. This may be wrong.)
The current (public school) system, inasfar as it has a single consistent goal, tries to optimize the minimal level of education that almost all students attain. Therefore it invests more in students who are doing worse: a lesson advances at the pace of the slowest students. There are sometimes multiple classes for differently performing students, but these are limited to a few subjects in a few years and are still very coarse-grained. (Of course the system also has non-education-related goals, which makes things harder to analyze.)
If I wanted to change the system to optimize average academic achievement instead, I would first of all group students by ability, not by age cohort. And this grouping would be separate for separate subjects. In today’s system, if a student fails one class but does well in others, she must either repeat that year for the sake of one class, or advance a year and continue to fail to keep up in that one subject.
But it may be that neither approach is strictly better than the other; it depends on what you are optimizing for. The school system optimizes for socializing kids (even if their models of how to do so are wrong and harmful) and for pleasing their parents. Academic achievement isn’t a top priority, and even if it were, there is no objectively “correct” choice between maximizing average achievement, top 10% achievement, or top 90% achievement.
In addition to my other comments, I’d like to ask what your overall goal is for grade-school education. (It seems to me that most of your arguments are about grade school and not college.
I’d like to amplify that point. Educating people who mostly want to be educated is very different to educating people who mostly don’t want to be there.
This is one of the most important things in education. Sometimes a completely opposite strategy is required. For example, if students want to learn, you can give them necessary resources and let them study at their own speed. If students don’t want to learn, using this strategy they would learn nothing. Similarly, if students want to learn, it is good to have a debate about the topic. If they don’t want to learn, they will only use the debate to waste time.
We could separate the students who want to learn from those who don’t, and use different strategies for these groups. Problem is, how to separate them. If you ask students, some students will mistakenly believe they want to learn (for example, they would say they want to learn about computer science, when in reality they only enjoy playing computer games). Some of them will lie to make a good impression; or their parents will make them lie.
A possible approach would be to use some costly signalling. For example, to offer “advanced lessons” for volunteers in the afternoon, and later separate the students who participated in these advanced lessons. But this will only work once; when your method becomes known, again, even people without interest in the topic will “volunteer” for the advanced lessons.
It seems your first paragraph outlines a solution for the second paragraph. First spend some time teaching students using a method that works well for motivated learners and poorly for unmotivated learners. Then give a placement test measuring how much they learned. Put the people who did well in the placement test into a group that teaches motivated students, and the people who did poorly into a group that teaches unmotivated students.
I think the arguments apply to grade school as well as college. I’m not sure what my overall goal is in terms of balancing the trade-offs between weak/average/strong students. However, I don’t think it matters much. I think that if you pooled resources, you’d be able to cater to all types of students.
However, I don’t think it matters much. I think that if you pooled resources, you’d be able to cater to all types of students.
I think you’re sidestepping the problem about which goal to optimize for.
In private schools, the investment in a student is partly proportional to the money spent by the student’s parents. If a student is worse than average, they can pay for better tuition, in a more intense school program or from a private tutor. Or they can switch to a better, more expensive school. So a given school or program can (in the simplest case) divide resources between students equally, demand equal tuition pay and approximately equal aptitude on tests, and let market forces sort out the rest.
In a public school, however, resources are apportioned to students based on political factors and goals are set on a population, not an individual level. Academically poor students can’t just be failed out of public legally-mandated school, and they also can’t be forced to repeat classes forever. So the system has a natural tendency to invest more resources in poor students, because the incentives are much greater than any reward there may be for supporting already-good students. And as long as some students aren’t doing well enough, any additional marginal resources will be spent to help them rather than someone else. The whole system is satisficing more than it is optimizing, because it’s incentivized to care more about graduation/diploma rates than about average scores among those who do graduate.
(Disclaimer: I have not researched the subject extensively, and I am not very certain of the above.)
My main claim is that we should pool resources and create great lessons/tools that everyone can use. Ie. I don’t see how it’d be the case that these lessons/tools would benefit a certain group of students disproportionately. And thus I don’t see where the question of catering to certain students comes into play here.
As for my opinions though. First off, I’d like to say that I’m a big believer in the idea that education should be self-paced and personalized. But to answer your question, if we’re going to fix time instead of mastery, I think that we should cater to the strong students. I believe in the idea of giving people sufficient (not necessarily equal) opportunity, and rewarding those who work hard. I think that in most cases, the weaker students are weak largely because they don’t care and don’t work hard. I don’t have strong opinions about this stuff at all though.
Personally, I agree with your goals and like most of your ideas. But:
I believe in the idea of giving people sufficient (not necessarily equal) opportunity, and rewarding those who work hard.
The public school system disagrees, which makes your proposal impractical no matter how technically good it might be. Your ideas would be much more suited to private, higher education—which is already moving in that direction on some counts. And this was my point: when you observe public and/or legally mandatory schools are very badly designed, you should keep in mind their goals are different from yours.
Ah, I see your point now—practicality. I haven’t been addressing that. So far I’ve just argued for what I think it should be, not really taking practicality into account.
That’s not to say that I haven’t thought about and don’t care about practicality though. Just that this is a huge topic and I thought it’d be more productive to break it up.
Anyway, I really don’t know too much about politics and “how the system works”. It’s something I definitely plan on learning about in order to someday get my ideas implemented though.
I doubt that governments can be convinced via sensible argument to do something like this. But I envision something like the following being plausible: invest some 10s/100s of millions of dollars to create a prototype for one topic that is sufficiently great. That is clearly The Superstar, and thus everyone learning this topic will want to use the prototype. Hopefully this would lead to people wanting to invest/donate more money, and the project could expand. And finally, if the project grows large enough and is good enough, there’ll be pressure for governments to adopt it or to adapt.
I think that the key is getting a prototype that is sufficiently dominant. That is clearly The Superstar. I’m not really sure what it’d take to achieve that, but I sense that it’s doable with 10s/100s of millions of dollars. As for how to get that money… personally I’m going to try and start a startup.
In addition to my other comments, I’d like to ask what your overall goal is for grade-school education. (It seems to me that most of your arguments are about grade school and not college. This may be wrong.)
The current (public school) system, inasfar as it has a single consistent goal, tries to optimize the minimal level of education that almost all students attain. Therefore it invests more in students who are doing worse: a lesson advances at the pace of the slowest students. There are sometimes multiple classes for differently performing students, but these are limited to a few subjects in a few years and are still very coarse-grained. (Of course the system also has non-education-related goals, which makes things harder to analyze.)
If I wanted to change the system to optimize average academic achievement instead, I would first of all group students by ability, not by age cohort. And this grouping would be separate for separate subjects. In today’s system, if a student fails one class but does well in others, she must either repeat that year for the sake of one class, or advance a year and continue to fail to keep up in that one subject.
But it may be that neither approach is strictly better than the other; it depends on what you are optimizing for. The school system optimizes for socializing kids (even if their models of how to do so are wrong and harmful) and for pleasing their parents. Academic achievement isn’t a top priority, and even if it were, there is no objectively “correct” choice between maximizing average achievement, top 10% achievement, or top 90% achievement.
I’d like to amplify that point. Educating people who mostly want to be educated is very different to educating people who mostly don’t want to be there.
This is one of the most important things in education. Sometimes a completely opposite strategy is required. For example, if students want to learn, you can give them necessary resources and let them study at their own speed. If students don’t want to learn, using this strategy they would learn nothing. Similarly, if students want to learn, it is good to have a debate about the topic. If they don’t want to learn, they will only use the debate to waste time.
We could separate the students who want to learn from those who don’t, and use different strategies for these groups. Problem is, how to separate them. If you ask students, some students will mistakenly believe they want to learn (for example, they would say they want to learn about computer science, when in reality they only enjoy playing computer games). Some of them will lie to make a good impression; or their parents will make them lie.
A possible approach would be to use some costly signalling. For example, to offer “advanced lessons” for volunteers in the afternoon, and later separate the students who participated in these advanced lessons. But this will only work once; when your method becomes known, again, even people without interest in the topic will “volunteer” for the advanced lessons.
It seems your first paragraph outlines a solution for the second paragraph. First spend some time teaching students using a method that works well for motivated learners and poorly for unmotivated learners. Then give a placement test measuring how much they learned. Put the people who did well in the placement test into a group that teaches motivated students, and the people who did poorly into a group that teaches unmotivated students.
I think the arguments apply to grade school as well as college. I’m not sure what my overall goal is in terms of balancing the trade-offs between weak/average/strong students. However, I don’t think it matters much. I think that if you pooled resources, you’d be able to cater to all types of students.
I think you’re sidestepping the problem about which goal to optimize for.
In private schools, the investment in a student is partly proportional to the money spent by the student’s parents. If a student is worse than average, they can pay for better tuition, in a more intense school program or from a private tutor. Or they can switch to a better, more expensive school. So a given school or program can (in the simplest case) divide resources between students equally, demand equal tuition pay and approximately equal aptitude on tests, and let market forces sort out the rest.
In a public school, however, resources are apportioned to students based on political factors and goals are set on a population, not an individual level. Academically poor students can’t just be failed out of public legally-mandated school, and they also can’t be forced to repeat classes forever. So the system has a natural tendency to invest more resources in poor students, because the incentives are much greater than any reward there may be for supporting already-good students. And as long as some students aren’t doing well enough, any additional marginal resources will be spent to help them rather than someone else. The whole system is satisficing more than it is optimizing, because it’s incentivized to care more about graduation/diploma rates than about average scores among those who do graduate.
(Disclaimer: I have not researched the subject extensively, and I am not very certain of the above.)
My main claim is that we should pool resources and create great lessons/tools that everyone can use. Ie. I don’t see how it’d be the case that these lessons/tools would benefit a certain group of students disproportionately. And thus I don’t see where the question of catering to certain students comes into play here.
As for my opinions though. First off, I’d like to say that I’m a big believer in the idea that education should be self-paced and personalized. But to answer your question, if we’re going to fix time instead of mastery, I think that we should cater to the strong students. I believe in the idea of giving people sufficient (not necessarily equal) opportunity, and rewarding those who work hard. I think that in most cases, the weaker students are weak largely because they don’t care and don’t work hard. I don’t have strong opinions about this stuff at all though.
Personally, I agree with your goals and like most of your ideas. But:
The public school system disagrees, which makes your proposal impractical no matter how technically good it might be. Your ideas would be much more suited to private, higher education—which is already moving in that direction on some counts. And this was my point: when you observe public and/or legally mandatory schools are very badly designed, you should keep in mind their goals are different from yours.
Ah, I see your point now—practicality. I haven’t been addressing that. So far I’ve just argued for what I think it should be, not really taking practicality into account.
That’s not to say that I haven’t thought about and don’t care about practicality though. Just that this is a huge topic and I thought it’d be more productive to break it up.
Anyway, I really don’t know too much about politics and “how the system works”. It’s something I definitely plan on learning about in order to someday get my ideas implemented though.
I doubt that governments can be convinced via sensible argument to do something like this. But I envision something like the following being plausible: invest some 10s/100s of millions of dollars to create a prototype for one topic that is sufficiently great. That is clearly The Superstar, and thus everyone learning this topic will want to use the prototype. Hopefully this would lead to people wanting to invest/donate more money, and the project could expand. And finally, if the project grows large enough and is good enough, there’ll be pressure for governments to adopt it or to adapt.
I think that the key is getting a prototype that is sufficiently dominant. That is clearly The Superstar. I’m not really sure what it’d take to achieve that, but I sense that it’s doable with 10s/100s of millions of dollars. As for how to get that money… personally I’m going to try and start a startup.