Only Do Math Homework

By Robin Hanson · June 23, 2011 11:00 pm · Comments (29) · ·

Following an identification strategy that allows us to largely eliminate unobserved student and teacher traits, we examine the effect of homework on math, science, English and history test scores for eighth grade students in the United States. Noting that failure to control for these effects yields selection biases on the estimated effect of homework, we find that math homework has a large and statistically meaningful effect on math test scores throughout our sample. However, additional homework in science, English and history are shown to have little to no impact on their respective test scores. (more)

So does anyone think that once this result becomes better know, they’ll stop assigning all but math homework? Me neither. Yes, maybe homework helps kids to learn things that their tests do not test. But more likely, homework functions to get kids used to doing a lot of work, in preparation for their future industry era jobs. Learning seems secondary.

GD Star Rating
loading...
Tagged as:
Trackback URL:

2 Trackbacks

  1. [...] You Should Practice Robin Hanson points us to a study that shows kids should only do math homework. He takes the opportunity to bang on his drum of school as indoctrination, but I was reminded of a [...]

    By What You Should Practice « feed on my links June 24, 2011 at 1:54 pm

  2. [...] posted before that ever though most homework doesn’t seem to help learning, I predict we’ll keep [...]

    By Overcoming Bias : Sleepy Kids Learn Less July 31, 2011 at 10:01 am

27 Comments

  1. salacious
    June 23, 2011 at 11:12 pm | Reply

    Or maybe math skills are more reliably tested.

    • gwern
      June 24, 2011 at 1:13 pm | Reply

      One would hope that history, science, and writing at least would be testable at all… Other areas of the educational establishment don’t seem to have trouble doing so.

      If the tests for those subjects were simply noisier, then you would still expect to see an average improvement, no? Just with bigger standard deviations.

    • Silas Barta
      June 24, 2011 at 2:25 pm | Reply

      Yes, because there’s something objective to test, in contrast to the other areas in which it’s more like, “Can you read the fluff in the teacher’s mind?”

  2. Robert Wiblin
    June 24, 2011 at 3:16 am | Reply

    Couldn’t they try to change the homework they assign to make it more educational instead?

    Is there really a tradeoff between teaching people to be disciplined and work many hours, and teaching them useful content?

    • dWj
      June 24, 2011 at 10:03 am | Reply

      I was thinking this; perhaps just assign more math homework in place of everything else. On the other hand, a willingness to do useless work might have signalling content that employers value.

      • IVV
        June 24, 2011 at 11:44 am |

        Certainly not! I only assign useful work to my underlings. Only people who don’t see the big picture like I do think that the work is useless.

    • Jess Riedel
      June 24, 2011 at 3:18 pm | Reply

      Couldn’t they try to change the homework they assign to make it more educational instead?

      What dWj said. Also, developing homework assignments in non-math subjects which actually are educational is clearly hard. (Otherwise, such homework would already exist…unless you think they suppress it intentionally.) So there’s no easy way to implement your suggestion. However, there is an easy way to implement Robin’s suggestion: tell teachers to stop assigning the useless non-math homework. (Or at least decrease it.) Robin’s claim, of course, is that no one will implement this easy suggestion because the point of the non-math homework isn’t to teach the students the material.

    • Lex Spoon
      June 25, 2011 at 10:49 am | Reply

      Sure, but you are reinforcing Robin’s point. First we assume that homework will be given out, and only then do we try to maximize the benefit of it.

      My favorite example is that school always lasts 12 years no matter how fast or slow the student progresses. It’s conceivable to put people in different classes, but it’s inconceivable to let them GED out at grade 6.

  3. Floccina
    June 24, 2011 at 9:41 am | Reply

    The whole structure schools and the way most people talk about schools seems to indicates that leaning is secondary.

  4. Colin
    June 24, 2011 at 10:26 am | Reply

    This is also interesting in the face of many ed reformers who wish to soften math instruction and not “drill” students, whereas anyone from a math background knows math takes a lot of practice.

  5. hamilton
    June 24, 2011 at 10:30 am | Reply

    Why should we base what we assess off of one study of one population in one journal? Why do social scientists continue to pretend that humans are like physical particals, so that knowing what happens with one group of humans is sufficient to generalize to all of them?

    • Dustin
      June 24, 2011 at 7:26 pm | Reply

      Why shouldn’t we? Of course, there’s the usual problems of the not having a large enough sample size and whatnot, but are you implying that it’s not possible to generalize results of studies on smaller groups to larger groups?

  6. Curt Adams
    June 24, 2011 at 10:44 am | Reply

    Showing *additional* homework has little value doesn’t indicate there should be *no* homework. Strictly speaking, it doesn’t even show moderate reductions would be helpful, although it would certainly be worth investigating.

    The (elementary) school my son attends assigns relatively little homework, and only math, reading, and writing. A nearby magnet school assigns tons – multiple hours even for first graders. FWIW, The parents of the magnet school kids complain about the homework being excessive, although they do still send their kids there, by definition.

    • Dan Weber
      June 27, 2011 at 3:13 pm | Reply

      1. Smart kids do a lot of homework.
      2. I want my kids to be smart.
      3. Therefore, they should do a lot of homework.

  7. Mercy
    June 24, 2011 at 11:00 am | Reply

    At least in the UK, there are two types of homework in English and History, the early science style makework stuff to check that you are actually paying attention (what did Huck say to Jim in chapter 3) and the actual essays which the teacher will actually go through beforehand and after, giving you tips on how to structure essays/research and improve your grammar.

    The former is more prevalent, especially at the early years, and it makes this result unsurprising to me- teachers never even tried to pretend that shit was helpful. But I’d be deeply surprised if the second style is useless, as it’s the only way to teach essay-writing unless you want to burn lesson time when the teacher could be talking.

    I suppose that’s the test of where homework’s important- where kids can’t be relied on to do it, to teachers take time out of lessons to do it instead. That’s what happened with essay-writing homework in the lower sets, but not with the questionsheet makework.

    • Buck Farmer
      June 24, 2011 at 12:48 pm | Reply

      I too am surprised.

      My writing quality has declined precipitously from when I had to churn out an essay or two every week.

      Similarly, I’d imagine that science homework would also have an effect. At least as I recall, my physics homework was much more about developing the correct physical intuitions (i.e. when to use the right symmetries, always check limiting cases) than it was chugging through the math. These disciplines I still apply, and I only picked up via homework.

      It might be that math homework is better aligned to testing, but I suspect instead it is just that we ought to be assigning fewer hours of harder homework to our 8th graders.

      This requires teachers that can design and grade it though.

  9. Twan van Laarhoven
    June 24, 2011 at 1:21 pm | Reply

    Math homework is a good way of forcing students to actually do something. As a student it is easy to listen to a teacher, and think that you understand what he is saying. But when the time comes for you to actually apply what you have learned, it is suddenly a lot harder. Actually sitting down and doing math problems is the only good way to learn. These problems don’t have to be homework problems, a good student can find problems anywhere, but homework does give an easy source of them.

  10. Anonymous
    June 24, 2011 at 2:41 pm | Reply

    Homework should be voluntary. In fact, school should be voluntary. Learning should be presented like a game, challenge-driven and curiosity-driven. Children don’t get paid, and they don’t usually sign voluntary contracts that would allow teachers to force them to do things.

    I didn’t get this a as a child, but in hindsight, the idea that a non-consenting child is somehow obligated to do anything a teacher orders them to do is absurd.

  11. Michael Bishop
    June 24, 2011 at 5:20 pm | Reply

    I predict that reading this post most readers will get the impression that a) Robin has a high degree of confidence in the quoted research, and that b) Robin thinks that people who prioritize student learning and accept this research result must abandon homework. I understand the argument against disclaimers but I don’t think this post accomplishes what Robin wants. It would have been better to paste the abstract without commentary.

  12. daedalus2u
    June 25, 2011 at 1:31 pm | Reply

    Robin is exhibiting the definition of economist as “someone who knows the price of everything but the value of nothing.”

    The point of the article (which everyone has missed), is not that non-math homework has no value, but that the testing of the value of non-math homework doesn’t measure the value that it has. Simply because you can’t measure something, doesn’t mean you get to assign an arbitrary value, negative, zero or positive.

    In this example Robin is assigning a value of zero to non-math homework because the value can’t be measured. What measure is there of the value of tax cuts? Either Regan’s or Bush’s?

    http://krugman.blogs.nytimes.com/2011/06/22/reagan-and-revenues/

    By any truthful measure those tax cuts reduced federal revenues. But if those who want lies pay better than those who want truth, the invisible hand will provide those with money what they want. Teach your children to be good servants because that is something the rich will always want, until they get used up.

    Einstein said:

    Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted.

    If you view other humans simply as machines made out of meat to use (and use up) to do things that you want enough to pay them to do them, then only things that can be monetized have any value.

    There is another Einstein quote:

    All that is valuable in human society depends upon the opportunity for development accorded the individual.

    But of course this mindset is anathema to Robin and other libertarian leaning conservatives. Their mindset is that the only thing of value is what profits me personally. If education doesn’t further my agenda, then it is a complete waste and should be eliminated.

    • Radford Neal
      June 25, 2011 at 8:02 pm | Reply

      How did you get from Robin’s post to this rant? Where does he imply that the only thing of value is what profits him personally? Since he’s past the age of being assigned homework, he would seem to have only an altruistic interest in reducing other people’s homework.

      Furthermore, while it might well be true that homework has effects that are not easily measured, whey do you assume that these effects are beneficial? Drudgery usually isn’t.

  13. Kendall
    June 27, 2011 at 7:54 am | Reply

    I assume nobody is claiming students don’t need to do any work at all to learn. Homework is only the work which students didn’t have time to finish in class. It seems unlikely to me that we have the perfect class lengths which allow us to do all useful work during classtime.

    • Radford Neal
      June 27, 2011 at 2:29 pm | Reply

      It does seem likely that school classes are not timed to be exactly the perfect length needed for learning. But why would you assume that class time is too short, rather than too long?

  14. Kendall
    June 29, 2011 at 11:26 pm | Reply

    It seems reasonable to me that at least part of the time the classes are too short for at least some of the students to complete the work necessary to learn the required material. Some students learn faster than others and some students will waste class time and will have to finish the work at home.

  15. lemmy caution
    July 12, 2011 at 7:52 pm | Reply

    The expansion of the amount of homework that kids are given nowadays drives me nuts. It is all well and good if your kid happens to do well in school, but it is close to a childhood crusher if your kid happens to do poorly in school.

Post a Comment

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>