Evidence on Single-Sex Schooling Is Mixed
Now that the U.S. Department of Education has signaled its intent to ease long- standing federal rules that made it tough for public schools to teach boys and girls separately, some educators may be looking to research for advice on how best to do it.
Chances are they'll be disappointed.
Despite the political and media hype that has surrounded single-gender schooling, studies on the idea are mixed, according to experts.
"The issue is overpoliticized and underresearched," said Cornelius Riordan, a sociology professor at Rhode Island's Providence College. "That's unfortunate."
The Education Department on May 8 began seeking comments on plans to scrap a strict interpretation of Title IX, the federal law that prohibits schools from discriminating on the basis of sex. ("Department Aims to Promote Single- Sex Schools," May 15, 2002.)
Part of the Education Amendments Act of 1972, the law—and subsequent court rulings—made it difficult for any school receiving federal aid to organize single-gender classes and schools.
In the resulting controversy over the department's plan, some commentators have noted that the lack of research undergirding single-gender education is ironic, given the Bush administration's emphasis on "evidence-based" educational practices.
But Mr. Riordan and others who have studied the issue say the paucity of conclusive research is not reason enough to limit experiments in single-sex public schools and classes.
"If you're opposing them, you're still constraining research on them," Mr. Riordan said. "What we really need is more research."
The problem is not that there are too few studies, the experts say. It's that there are too few high-quality ones—studies, for example, that use comparison groups or credible national databases. Also, many studies, researchers say, tend to focus on women's colleges, elite private schools, Roman Catholic schools, or single-sex schools in faraway nations, rather than the American public schools that are the focus of the proposed policy change.
Reviews of the literature, in fact, reach differing conclusions about what all of those studies have to say, at bottom, on whether single-sex schooling improves student achievement.
Most prominent among them was a 1998 report by the American Association of University Women, the group that helped launch a surge of interest in all-girls educational programs in the early 1990s. In its report four years ago, however, the Washington-based group concluded there was insufficient evidence that single-sex education "works or is better'' than coeducation.
But Mr. Riordan, who has spent decades reviewing the literature, contends that single-sex schools do produce small academic gains for boys and girls. He says the gains are stronger, though, for minority students and those from low-income families than for students overall.
In another 1998 review of studies on single-sex programs, Fred A. Mael, a researcher for the American Institutes of Research, a Washington think tank, also says the preponderance of evidence points to some academic and social benefits for students in those settings. It's just that most of the studies focus on girls. Evidence is more anecdotal, Mr. Mael said, for boys in all-male classes and schools.
Some newer research suggests, for example, that large concentrations of boys in classrooms can detrimentally affect learning.
In a forthcoming study in the Quarterly Journal of Economics, Caroline M. Hoxby, a Harvard University economist, takes a look at what happens in classes where imbalances in the share of boys or girls in the room occur naturally.
For example, a family may have one son whose kindergarten class is overwhelmingly populated with girls and another who comes to school two years later to find that most of his classmates are boys.
Examining those variations through statewide data on Texas 3rd through 6th graders, Ms. Hoxby found that having higher numbers of boys in a classroom was associated with lower achievement for both girls and boys.
"It's a substantial enough difference that, if I were a parent, I'd care about it," she said. "But I certainly wouldn't, on the basis of my study, rule out the idea that all-boy classes aren't good."
As Mr. Riordan points out—and Ms. Hoxby concurs—studying what happens at the classroom level is not the same as looking at whole schools populated by a single gender. That's a distinction, they say, educators have to keep in mind as well when they look at the evidence.
One of the newest and largest studies on the subject looked at six years of test-score data for 270,000 Australian students. The authors adjusted the data to account for differences at the outset in students' abilities or in the type of schools they attended—in other words, whether the schools were religious, private, or public. The researchers then found that boys and girls in single-sex schools scored, on average, 15 to 22 percentile points higher than peers in coeducational settings.
Ken J. Rowe, who led the project for the Australian Council for Educational Research, said those gains "pale into insignificance," however, compared with the gains that students make from having skilled teachers—regardless of their classmates' gender.
Getting a good education was all 10th grader Ileana Garcia had in mind four years ago when she switched from her regular coeducational public school to an all-girls charter school in New York City's East Harlem. Even though it had just opened its doors, the Women's Leadership Academy had a better reputation.
Now, she says, it's difficult to sort out what it is about her school that seems to motivate her to work harder. Without boys in class, she said, "it's more relaxed, and there's less distractions during class and less worry about how you look."
At the same time, she points out, the small size of her school means that "teachers can really get to know you, and you can get to know teachers."
That, too, is something that researchers have yet to sort out.
Coverage of research is underwritten in part by a grant from the Spencer Foundation.
Interesting Ideas? Send suggestions for possible Research section stories to Debra Viadero at Education Week, 6935 Arlington Road, Suite 100, Bethesda, MD 20814; e-mail: email@example.com
Vol. 21, Issue 40, Page 8