My colleague Valerie Strauss reports that French president François Hollande has called for abolishing homework. The idea is part of a broader set of proposed education reforms, which include longer school weeks (they're currently only four days) and hiring more teachers. The appeal to students is obvious, but is this such a good idea?
On the one hand, the evidence suggests that homework helps student to learn. Duke University's Harris Cooper, Jorgianne Civey Robinson and Erika A Patall summarized the research on homework conducted between 1987 and 2003 and found that students who were required to do homework outperformed those who didn't by 0.6 standard deviations, with a correlation of 0.24. The correlation means that homework only explains about 5.8 percent of the difference between students' achievement levels. But it's still a significant, positive effect.
There are other advantages to homework, as well. Carnegie Mellon's Steven Schlossman and Mathematica Policy Research's Brian Gill have found that homework encourages parents to get more involved in their children's education.
But Hollande is also proposing more instructional time. Cooper et al., note that older studies found that homework is superior to in-school individual study sessions, but no recent research has been done, and even the old studies show a very small effect. What's more, the effects of increased instructional time are unambiguously positive. One example: Using randomized results from charter school lotteries, Harvard's Will Dobbie and Roland Fryer found that schools with 25 percent more instructional time have results that are 0.059 standard deviations above those with less instruction. Hollande's approach, then, could end up as a net positive.
That still leaves the homework abolition as a downside. Hollande, however, defends it not in terms of student learning but as a way to equal the playing field for all students. Poor children, he argues, are less likely to get parental aid for homework, and so requiring homework can widen the achievement gap.
The evidence on this is more mixed than one might expect. Min Zhan at the University of Illinois found that parental involvement in homework improved student achievement in reading, but not math, and that there was no statistically significant correlation between how rich parents are and how much they help out their kids. Married parents are more likely to get involved with homework, but, otherwise, disadvantaged parents supervise homework as much as other parents. Another study found that while the assignments encourage parents to help with math homework, that does not in turn improve student achievement in that subject.
This is not to say that students from a wealthy or otherwise socioeconomically advantaged household don't get more benefit from doing homework. But it does suggest that the effect is indirect, similar to the effect on student achievement in general. Hence, it doesn't say much about what the balance should be between homework and other kinds of instruction.
It's hard to evaluate Hollande's plan without considering how much the extra instruction would replace homework and whether the benefits would be the same. But all else being equal, homework is an effective educational tool.
If Hollande wants to tackle a bigger culprit behind the achievement gap, he should take a look at summer vacation. The data suggests that it mainly functions to allow rich parents to instruct their kids more while poor students fall further behind. Johns Hopkins researchers found that summer is behind two-thirds of the achievement gap between high- and low-income students in Baltimore. A study from the RAND Corporation found that summer educational programs can help close the gap.
More homework and shorter summer breaks are not going to win you the third-grader vote, but it could do a lot for educational equality.