You are right that eight year olds can’t give each other the “correct” passphases to academic work in all cases. On the other hand that doesn’t mean that the feedback isn’t useful.
As far as spelling errors go, I see no reason why eight year old should be unable to find them. As far as style issues go, a eight year old has a much easier time to understand feedback from another eight year old than to understand feedback from an adult.
I think you may be overestimating the writing abilities of the average eight year old. When it comes to style issues, more third graders are in need of learning basic coherence than stylistic elegance.
I think that most of the Less Wrong population was probably at least somewhat precocious, which may make it hard to relate to what the average third grader needs to learn, as opposed to exceptional ones.
Having third graders attempt to correct each others’ work sounds to me like trying to get white belts to try to correct each others’ form in a martial arts class. Low level students may practice together, but you don’t try to put them in a position of giving each other instruction until they achieve some measure of basic competence.
Having third graders attempt to correct each others’ work sounds to me like trying to get white belts to try to correct each others’ form in a martial arts class.
I haven’t used the word ‘correct’. A third grades should be able to give you an answer to: Does this sentence look coherent? Why do you think so?
I see no reasons why he shouldn’t be able to share the answer to those questions with his classmates.
If the teacher wants he can offer the correct solution afterwards.
I think it would probably be more effective to have the students work on samples provided for them by the teacher, perhaps ones written by students from other classes, so as to avoid bickering and posturing between students.
Students are used to receiving corrections and critiques from their teachers. Having a peer critique one’s work and offer corrections, and not take it as a status affront, is and act that requires some maturity, and many people never learn to do it at all.
Learning to receive correction from peers gracefully is an important skill which is probably worth taking the time to teach to students, but I don’t think that paired peer critiques would be the best way of doing this. Some students will naturally produce better work than others, and thus have more criticism to offer and less to receive compared to their peers. The exercise would provide an element of “smart people don’t have to receive criticism from their peers, ordinary/stupid people do,” which the teacher would have to actively work against.
To teach students to accept correction from peers without taking it as a status challenge or an affront, I think I would try an exercise along these lines. Split the class into two groups. In each group, students work solo on a task provided by the teacher. However, the teacher provides different information to the students in each group. Neither group has all the information needed to perform the task correctly. After the students have tried and failed for a while, integrate the groups and pair the students up together, and tell them to pool their information.
Another way to get students to accept critizism would be to let each student rewrite his text after the feedback and then grade the revised version.
Having a peer critique one’s work and offer corrections, and not take it as a status affront, is and act that requires some maturity, and many people never learn to do it at all.
Saying many people never learn the skill at all is just another way of saying that today’s schools are very bad at teaching the skill.
To me the skill seems on of the most important that children should learn in school.
I think it makes sense to try to teach the skill as early as possible. Of course, when trying something new it’s importance to see how it works in practice and calibrate on the reactions of actual students.
Another way to get students to accept critizism would be to let each student rewrite his text after the feedback and then grade the revised version.
But this runs into the problem I mentioned before, that some students will have plenty of correction to offer others’ work, but little correction to receive from their peers. So instead of “everyone needs to learn to accept correction gracefully,” you risk teaching them “regular/stupid people need to learn to accept correction from smart people.”
If I had been put through such activities at that age, I suspect it would have amplified the arrogance that I was already developing at that time. In third grade, I tested as having a twelfth grade reading level; it’s highly probable that no student in my class at the time could have offered any useful critique on my work, whereas I could have offered critique on all of theirs.
It was already hard enough to unlearn the habits of thought that sort of disparity cultivated in me without participating in class activities which reinforced it.
So instead of “everyone needs to learn to accept correction gracefully,” you risk teaching them “regular/stupid people need to learn to accept correction from smart people.”
Which is a good and useful thing for stupid people to learn.
Better than their thinking they don’t need to accept correction at all, but then, if you don’t teach them measures for determining whether “corrections” are actually correct, they’re liable to get their heads filled up with garbage.
And teaching the stupid students to learn to accept correction gracefully from smart students is a lot less good and useful than teaching all the students to accept correction gracefully from their peers whenever they’re mistaken and receive it.
Well of course, if after hearing an explanation they realize they were wrong and why, then they should accept that correction no matter where it came from. And they should be open to listening to such proposed corrections from all sources.
But often even after the error is pointed out and explained, they still don’t understand why they were wrong. The one correcting them needs to know more than the minimum to be able to teach others, and needs to invest the time to explain it. There’s a long way from “remember correctly the right answer” to “be able to explain why this answer is the right one”. Most children are not able to understand why what they are taught is right, and even if they are, they are rarely taught these reasons.
So if children know some other, smarter children are pretty reliably right, then they should definitely use a heuristic of deferring to them whenever they disagree about the study subject matter. This is easy to calibrate empirically based on previous experience, even for children—people are good at tracking who tends to be right.
I don’t think students tend to be particularly shortchanged with respect to that sort of lesson by the current system though. Most students are aware of who the top students are, and that the top students are more reliable in their areas of expertise than the lesser ones.
Should they defer to the top students in matters outside of what they know to be those students’ areas of expertise? Not necessarily; a lot of smart people are not particularly good at getting right answers outside their areas of expertise. For a not-so-smart person, simply trusting that a smarter person knows what they’re talking about on any given subject, even when they don’t know that person to be an expert on the subject, is not a very trustworthy heuristic.
Trusting people who’re more expert than you in a particular field over your own judgment in that field is something that most students already learn. Being unable to assess who’s more expert in domains in which they’re not well trained is natural and probably unavoidable, and students can’t be trained to be expert in everything.
So instead of “everyone needs to learn to accept correction gracefully,” you risk teaching them “regular/stupid people need to learn to accept correction from smart people.”
There’s a corollary: “Smart people need to learn to give feedback in a way that doesn’t annoy stupid people but let stupid people accept their feedback.”
You probably would have profit from learning that skill in third grade.
I did learn that skill much earlier than I learned to accept correction gracefully from others. I had much more occasion to practice tailoring my feedback to others so that it would be accepted than I had to practice receiving and incorporating it.
I don’t think smart students are as shortchanged when it comes to learning that skill as the reverse, with the current status quo.
I think you may be overestimating the writing abilities of the average eight year old. When it comes to style issues, more third graders are in need of learning basic coherence than stylistic elegance.
I think that most of the Less Wrong population was probably at least somewhat precocious, which may make it hard to relate to what the average third grader needs to learn, as opposed to exceptional ones.
Having third graders attempt to correct each others’ work sounds to me like trying to get white belts to try to correct each others’ form in a martial arts class. Low level students may practice together, but you don’t try to put them in a position of giving each other instruction until they achieve some measure of basic competence.
I haven’t used the word ‘correct’. A third grades should be able to give you an answer to: Does this sentence look coherent? Why do you think so? I see no reasons why he shouldn’t be able to share the answer to those questions with his classmates.
If the teacher wants he can offer the correct solution afterwards.
I think it would probably be more effective to have the students work on samples provided for them by the teacher, perhaps ones written by students from other classes, so as to avoid bickering and posturing between students.
Students are used to receiving corrections and critiques from their teachers. Having a peer critique one’s work and offer corrections, and not take it as a status affront, is and act that requires some maturity, and many people never learn to do it at all.
Learning to receive correction from peers gracefully is an important skill which is probably worth taking the time to teach to students, but I don’t think that paired peer critiques would be the best way of doing this. Some students will naturally produce better work than others, and thus have more criticism to offer and less to receive compared to their peers. The exercise would provide an element of “smart people don’t have to receive criticism from their peers, ordinary/stupid people do,” which the teacher would have to actively work against.
To teach students to accept correction from peers without taking it as a status challenge or an affront, I think I would try an exercise along these lines. Split the class into two groups. In each group, students work solo on a task provided by the teacher. However, the teacher provides different information to the students in each group. Neither group has all the information needed to perform the task correctly. After the students have tried and failed for a while, integrate the groups and pair the students up together, and tell them to pool their information.
Another way to get students to accept critizism would be to let each student rewrite his text after the feedback and then grade the revised version.
Saying many people never learn the skill at all is just another way of saying that today’s schools are very bad at teaching the skill.
To me the skill seems on of the most important that children should learn in school.
I think it makes sense to try to teach the skill as early as possible. Of course, when trying something new it’s importance to see how it works in practice and calibrate on the reactions of actual students.
But this runs into the problem I mentioned before, that some students will have plenty of correction to offer others’ work, but little correction to receive from their peers. So instead of “everyone needs to learn to accept correction gracefully,” you risk teaching them “regular/stupid people need to learn to accept correction from smart people.”
If I had been put through such activities at that age, I suspect it would have amplified the arrogance that I was already developing at that time. In third grade, I tested as having a twelfth grade reading level; it’s highly probable that no student in my class at the time could have offered any useful critique on my work, whereas I could have offered critique on all of theirs.
It was already hard enough to unlearn the habits of thought that sort of disparity cultivated in me without participating in class activities which reinforced it.
Which is a good and useful thing for stupid people to learn.
Better than their thinking they don’t need to accept correction at all, but then, if you don’t teach them measures for determining whether “corrections” are actually correct, they’re liable to get their heads filled up with garbage.
And teaching the stupid students to learn to accept correction gracefully from smart students is a lot less good and useful than teaching all the students to accept correction gracefully from their peers whenever they’re mistaken and receive it.
Well of course, if after hearing an explanation they realize they were wrong and why, then they should accept that correction no matter where it came from. And they should be open to listening to such proposed corrections from all sources.
But often even after the error is pointed out and explained, they still don’t understand why they were wrong. The one correcting them needs to know more than the minimum to be able to teach others, and needs to invest the time to explain it. There’s a long way from “remember correctly the right answer” to “be able to explain why this answer is the right one”. Most children are not able to understand why what they are taught is right, and even if they are, they are rarely taught these reasons.
So if children know some other, smarter children are pretty reliably right, then they should definitely use a heuristic of deferring to them whenever they disagree about the study subject matter. This is easy to calibrate empirically based on previous experience, even for children—people are good at tracking who tends to be right.
I don’t think students tend to be particularly shortchanged with respect to that sort of lesson by the current system though. Most students are aware of who the top students are, and that the top students are more reliable in their areas of expertise than the lesser ones.
Should they defer to the top students in matters outside of what they know to be those students’ areas of expertise? Not necessarily; a lot of smart people are not particularly good at getting right answers outside their areas of expertise. For a not-so-smart person, simply trusting that a smarter person knows what they’re talking about on any given subject, even when they don’t know that person to be an expert on the subject, is not a very trustworthy heuristic.
Trusting people who’re more expert than you in a particular field over your own judgment in that field is something that most students already learn. Being unable to assess who’s more expert in domains in which they’re not well trained is natural and probably unavoidable, and students can’t be trained to be expert in everything.
There’s a corollary: “Smart people need to learn to give feedback in a way that doesn’t annoy stupid people but let stupid people accept their feedback.” You probably would have profit from learning that skill in third grade.
I did learn that skill much earlier than I learned to accept correction gracefully from others. I had much more occasion to practice tailoring my feedback to others so that it would be accepted than I had to practice receiving and incorporating it.
I don’t think smart students are as shortchanged when it comes to learning that skill as the reverse, with the current status quo.