I know there's been plenty of discussion on the efficacy of homework here. Just wanted to point out a post on Dy/Dan, for those that don't already subscribe:
I Do Not get Homework At All Sometimes
The comments are interesting. Most lean to the Alfie Kohn view: homework=worthless. I know I'm not alone in believing that math takes practice.
For those of you in the know, where's the research?
There is a definite correlation in the top graph. The correlation isn't anywhere close to 1.0, but it *is* there. I'll make two observations about it:
ReplyDelete1) Every student that did 85% or more of the homework got an 'A' (better than 90%).
2) No student that did less than about 50% of the homework got an 'A'.
Standard 'correlation isn't causation' arguments apply.
But ... pretty much *every* other skill activity I can think of requires practice to get good. I think of good homework as practice. For me, this argument tends to turn into an argument against practice.
Since these kids will be competing against my kid to get into college, my attitude is, "Go for it! Don't practice, please!"
Now ... there *can* be worthless homework. I think this discussion would be more useful if we could bin the homework and argue about the value of the indivual bins, but I don't see this happening...
-Mark Roulo
Practice doesn't help in general or for this specific case? It would be difficult to argue the former, and we don't have enough information to evaluate the latter.
ReplyDeleteWhat's most scary are the posters who seem to think this is a deep enough evaluation to tell them anything.
We know how many completed the homework, but we don't know how many got the homework problems correct. What percent did the easy homework problem. Duh! We need to see some examples of homework and compare them to test questions.
One commenter suggested eliminating homework to see how the results change. Students become guinea pigs for very badly done experimental science. It could be that the teacher assigns stinking, lousy homework, or it could be that the standards are low. If the standards are low, then some students will achieve high grades with little effort. That makes it much more important to use practice to bring the rest up to the top level. Is it OK to have a bell-shaped results distribution for ensuring mastery of adds and subtracts to 20 in third grade? No, you want 100% achievement.
The other question re. homework is who is doing it. I know of situations where the parents were either doing it outright or were providing so much input/correction that the end product was not the child's work. I know of one situation where an Honors Algebra II student always had perfect homework, received only Ds and Fs (with the exception of one C)on quizzes/tests and received and A in the course because the homework was counted so heavily. If the educational establishment acknowledged such a phenomenon, it would have been educational malpractice,punished by loss of job/license for the appropriate teacher/administrator cohort.
ReplyDeleteOne more factor is the question of supplementation outside of school. Our school uses Everyday Math, and in the early grades, the teachers tend not to give a great deal of math homework.
ReplyDeleteI know families who have arranged for their children to have tutors for math, who assign extra math work. And of course, there's always Kumon.
So, any experiment which tries to make guinea pigs out of children would need to control for family circumstances. A child who gets no homework from school may very well still be doing math homework, rendering the results worthless.
Steve, not only do we not know how many problems were done correctly, we don't know IF IT WAS CORRECTED AT ALL.
ReplyDeleteIs there or is there not a feedback loop present for the students? Is the loop closed--that is, they do the homework, they get it corrected, they are at least once told where they went wrong?
Assigning homework that isn't corrected, or is corrected a random fraction of the time isn't going to produce the same effect as homework that is corrected 100% of the time. It's not going to produce much effect because students aren't going to bother to do it well, and aren't going to learn from it anyway.
but 2 problems a night isn't enough practice to learn anything anyway. No wonder there's no correlation in those that are doing "well"--perhaps those good students are learning it somewhere else entirely.
My district assessment policies specify that homework is not to be used for grades (homework completion is reported on separately as a "learning skill"), and quantity and quality of homework assigned tend to be hugely variable. Thus, I doubt large-scale data on "homework" tells us much about its value, because its value depends on its being appropriate for the child, linked to ongoing assessment, etc.
ReplyDeleteMuch homework is nothing but busywork (coloring, cutting, pasting, copying), while some other assignments may be overkill -- 50 long division problems where 10 would suffice to demonstrate mastery. Or, for a number of students, the homework may require skills the students don't have at sufficient mastery to permit their ndependent application. If kids don't have parents who can help them, they can't do the work and just get frustrated and quit.
If the homework is targeted, valued by the parent, monitored (if not corrected in its entirety -- not always a reasonable expectation), allows the child to see progress in his learning, etc. it can be very valuable. Too often however I see children trying to do homework that is too difficult, too easy, irrelevant or time-wasting.
At the secondary level it may be entirely another story -- I don't know.
doesn't mean much. completion could be a consequence rather than a cause of skills(good grades).
ReplyDeleteare the kids getting better grades because they are completing homework(aka practice)?
or are they completing the grade because they have the instructional/economic/social support?
*or are they completing the homework because they have the instructional/economic/social support?
ReplyDelete