Homework

I've been wondering whether the amount of homework given in schools is increasing, which seems, from my personal observation, to be the case.

According to a March 2000 Detroit News article titled "Students and Parents Are Buried in Work":

Today, U.S. fourth- and eighth-graders receive more math and science homework than most other children worldwide, according to an international education assessment of a half-million children, including 30,000 in the United States.

"We tend to give a lot of homework," says William H. Schmidt, a Michigan State University statistics professor who is national research coordinator of the Third International Mathematics and Science Study. "Both in the fourth and eighth grades, we tend to give among the highest amount."

For example, almost half of U.S. eighth-grade science teachers assign homework three to five times a week, according to the 1996 study. That's true of only 12 percent of science teachers in Germany and 4 percent in Japan.

However, according to an October 2003 article in the Chicago Sun-Times titled "Studies Agree: Homework Load Light":

The 1995 Third International Mathematics and Science Study showed American high school seniors tackle "an extraordinarily light load'' compared to their counterparts in other countries, the Brookings report contended. U.S. 12th-graders tied for second-to-last among 20 countries in the amount of time spent studying.

Notice that the two conclusions are based on the same study.

Back to the Detroit News article:

The nation's public school students are spending more time than ever on homework -- more than most students in the world, and the nightly grind is starting in earlier grades.

A 1997 University of Michigan study found that 6- to 8-year-old children spent on average more than two hours -- 129 minutes to be exact -- doing homework each week, nearly triple the 44 minutes they spent in 1981. For 9- to 12-year-olds, the amount of time climbed to three hours and 37 minutes weekly, up from two hours and 49 minutes in 1981.

But, according to the Chicago Sun-Times piece:

The typical U.S. student, even in high school, does not spend more than an hour a day on homework -- an amount that hasn't changed much since the 1980s, according a report to be released today by the Brookings Institution that summarizes four U.S. studies on homework trends.

The report's author, Tom Loveless of Brookings' Brown Center on Education Policy, said complaints of kids weighed down by heavy backpacks and homework overload are the exception rather than the rule. The most interesting study he reviewed, he said, was done by UCLA, in which college freshmen nationwide were asked how much time they spent on homework or studying as high school seniors.

In 1987, the first year the question was asked, 47 percent of kids said they spent more than five hours a week on homework as high school seniors. But the figure has shrunk every year since, and hit a record low of 34 percent in 2002....

Brookings' results jibe with a study called "A Nation at Rest: The American Way of Homework," which contends that for the last 30 years, only about 10 percent of high school students spent more than two hours a day on homework.

There's no real contradiction here, it's just that homework is increasing for younger students, but not, apparently, for high school students. Each of the above articles has chosen to focus on a different age group and therefore comes to a different conclusion.

Certainly, it looks to me like homework is increasing, but we live near a college town with an "excellent" school system, and our own district is also affluent. I still remember the first time I saw an elementary-school-aged child walking home, pulling behind him a backpack on wheels, clearly chock full of books. I was literally nauseated. Since then I've seen this so many times I've realized it's just the way things are in these parts. Your kid hits second grade and you buy them a small suitcase on wheels in which to cart their schoolbooks home each night.

The Detroit News article is full of anecdotes which support this notion of greatly increased homework in elementary schools. Notably, the article provides this partial explanation:

Homework increases may be more typical in affluent school districts, according to Harris Cooper, a University of Missouri psychology professor and homework historian.

"My sense is that the amount of homework is increasing in particular kinds of school districts -- generally affluent with parents who are professionals and want their children to get into competitive colleges," he says. "That's a big part of where the pressure for more homework is coming from."

Fred Riley, a counselor at Southfield's Levey Middle School, also believes that "demanding" parents are responsible for more homework. "The homework load has increased tremendously -- kids are being inundated," says the 15-year veteran of Southfield and Detroit schools. "Parents, especially those at some of the prestigious schools, feel they're not getting their money's worth unless the kid is just bogged down with homework."

It's not just parents who send their kids to private schools who may be concerned about getting their money's worth. You will pay about $50,000 extra for an average house in one of the two "good" school districts in our area, compared to neighboring districts. Two different parents I know, whom I've told about our homeschooling plans, said they wouldn't homeschool because the whole reason they had moved here was for the schools-- the implication being, if they homeschooled that would turn out to have been a waste of money.

If more homework is given in more affluent districts, then I wonder if a lot of homework studies which appear to show benefits aren't simply picking up on socioeconomic status. Sure, most studies attempt to control for SES, but you can only control for a variable insofar as you can measure it, and it's quite difficult to encapsulate SES in a single number. I worked as a statistical consultant at a major university for 3 years, and you can take it from me that SES is usually measured by annual income alone. That doesn't tell you about the wealth of the school district, nor the parents' education level. And as we know, in the public schools, parents' education level has a large effect on academic achievement as measured by test scores. If the only thing you've removed from the equation is the direct effect of household income, there's a lot still left over that could be causing higher achievement in those affluent districts which love to pile on the homework. It has long been my opinion, as a statistician, that you cannot fully control for or remove the effects of socioeconomic status, regardless of the subject under study.

As long as the myth persists that educating children is highly difficult and time-consuming, parents may continue to believe that more homework equals better education. And if the parents who are determined that their children be well educated are the ones who accept or demand more homework, the potential for a significant but artificial correlation between homework and achievement will persist as well.