Study Finds Single-Sex Schools Benefit Some—But Not All
Caribbean study may offer insights for United States
A study on publicly run schools in the Republic of Trinidad and Tobago has found that, while single-sex schools may benefit female students who prefer a single-sex environment, they are not inherently beneficial for boys or most girls.
While the findings are based on data from one Caribbean nation, experts say they may carry implications for public schools in the United States. Even as single-sex options in U.S. public schools have multiplied in the past decade, both their proponents and detractors point to a dearth of reliable research on the topic.
The Trinidad and Tobago study, which set out to fill that gap, runs counter to anecdotal evidence of single-sex programs' successes, but may bolster the cases both of opponents and of advocates who say single-sex schools should be a choice for those they do help.
The findings come at a time when single-sex public schooling remains a point of contention in the United States, with the American Civil Liberties Union, for one, contesting individual programs.
There were 510 public schools offering either single-gender classes within coeducational schools or fully single-sex schools at the beginning of the 2011-12 school year, according to one informal estimate.
Such programs began sprouting after the 2002 federal No Child Left Behind Act that explicitly allows states to fund "programs to provide same-gender schools and classrooms." A 2006 update to U.S. Department of Education regulations allows single-sex options as long as "substantially equal" opportunities are provided to both genders.
Flaws in Research
Research on the effectiveness of single-sex schooling is often considered flawed because, in the United States at least, most students at single-sex schools opt into their programs. Also, since curricula in single-gender schools often differ from curricula in coed schools, any academic and social benefits directly caused by the single-sex environment can be difficult to extract.
For the new study, lead researcher C. Kirabo Jackson, a labor economist at Northwestern University, in Evanston, Ill., tried to address the troublesome issue of self-selection by situating the investigation in Trinidad and Tobago, where a national curriculum and school application process allowed him to control for students' preference and academic performance. He analyzed data on 219,849 students from 123 schools to determine whether attending a single-sex school between 6th and 10th grades improved students' performance on 10th grade exams. The results will be published next month in the Journal of Public Economics.
"There's a common notion out there that single-sex ed. is beneficial for girls and not for boys," Mr. Jackson said. "These findings are consistent with that notion. But an important thing from this study is that not all girls benefited from single-sex schools." He also noted that students in Trinidad's all-girl schools were slightly less likely to take math or science courses.
Mr. Jackson said that while cultural differences between the countries were worth considering, he believes the results were relevant to many U.S. schools, especially in certain immigrant or cultural communities where traditional gender roles are particularly strong.
"Trinidad's not a place that's extraordinarily misogynistic, even though things are very traditional; ... it's a country where females tend to outperform males," Mr. Jackson said, and one that is strongly influenced by American culture.
Mr. Jackson said it was difficult to draw conclusions from most current studies on single-sex education.
"Even in the best studies, we don't know why one kid is in a single-sex and one in a coed [program]," he said, "so even if we find patterns, at best they suggest correlations." His study also calls for further longitudinal research.
Similar concerns about the state of research on the subject have surfaced in a number of recent reports, including one last fall in the journal Science on the "pseudoscience of single-sex schooling" that drew national attention. In that study, the authors write that "although excellent public [single-sex] schools clearly exist, there is no empirical evidence that their success stems from their [single-sex] organization."
Dr. Leonard Sax, a retired physician and the president of Alliance for Choice in Education, an advocacy group for single-sex education, in Exton, Pa., agreed that the scholarship on single-sex schooling is "very disappointing."
In response to the Trinidad and Tobago study's finding that the schools may not benefit all subgroups, Dr. Sax said he agreed that simply separating the sexes does not cause academic benefits. "It's an opportunity," he said, for educators to use different teaching styles that he contends are more suited to boys or to girls.
When asked if those methods were backed by research, he said: "I do not have a scholarly paper that says this. Experience gives me this information."
But Lise Eliot, a Chicago Medical School researcher and a co-author of the Science article, said that gender-based teaching methodologies were unproven and might exacerbate differences. Mr. Jackson's findings on girls' coursework in single-sex schools, she said, "raise some concerns about stereotyping of girls."
"There aren't many randomized studies, certainly not in the United States, and basically this demonstrated that there wasn't an advantage of single-sex schools," Ms. Eliot said.
Schools need to have an "exceedingly persuasive" case for separating students by sex, said Galen Sherwin, a staff lawyer for the ACLU's Women's Rights Project in New York.
But "the schools that are separating kids entirely on the basis of sex generally have very little in the way of justification to show academic benefits," she said. "That suggests that that kind of program is unconstitutional."
In 2011, the ACLU reached agreements with school boards in Pittsburgh; Tallapoosa, Ala.; and Adrian, Mo. and won a lawsuit against Vermillion Parish in Louisiana, resulting in the end of single-sex programs in those districts. An ACLU suit against Kentucky's Breckinridge County school board was dismissed in 2008.
No new cases are currently in court, but the organization "has open-records requests pending in several states and is evaluating the best approach" to challenging such programs, Ms. Sherwin said. "We have had some successes in eliminating the most problematic programs and will be continuing to do so."
In comparison, the Education Department's office for civil rights has not staked out a firm stance on single-sex programs, according to Russlynn Ali, the assistant secretary for that office.
"Like all policies, we are reviewing past policies to ensure their applicability, their relevance, their compliance with the law," she said. "We are looking to see court determinations [on the ACLU suits] and... also trying to get a handle on the world of research."
The Education Department does not keep an official tally of single-gender schools. For the first time, the 2011 Civil Rights Data Collection report will include information on which districts have single-sex schools, but, Ms. Ali said, "it's still just a sample." A universal collection is planned for the 2011-12 school year.
Dr. Sax's Alliance for Choice in Education—formerly the National Association for Single Sex Public Schools—listed schools with single-sex programs on its website until recently. It tallied 510 public schools—391 coeducational schools with single-sex programs and 119 single-sex schools. But Dr. Sax noted that the list was self-reported. It was taken down last year out of concern that opponents were using it to target certain schools.
Ms. Sherwin of the ACLU said her organization has found that "many programs don't meet existing legal requirements, which had led to a string of legal successes [for the ACLU]." And programs were often started without public discussion or scrutiny, she added.
Dr. Sax acknowledged that some programs did not comply with regulations. "The cases [the ACLU is] proud of are cases where the administrators need to do their homework," he said.
Providing both single-sex and coeducational options can be expensive for schools, said David Chadwell, who coordinated South Carolina's single-sex programs until October. The number of single-sex schools in South Carolina has declined after a peak two years ago, he said, dropping from 232 to 129 last fall.
Ms. Sherwin said that even when programs are in compliance with the law, "we question the regulations of the Education Department, which we believe go beyond Title IX," the main federal law against sex discrimination in education. "There's a reason courts look skeptically at separation due to sex—it's often been used to enforce sex stereotypes," she said.
Meanwhile, advocates for single-sex schools have begun using the language of school choice.
"Our argument has shifted," Dr. Sax said. "At first, we were here to spread the news that single-sex schools are legal. Now, ... we have to work on helping parents understand—is this a good choice for your daughter, good choice for your son?"
U.S. schools serving the population that Northwestern University's Mr. Jackson singles out as most likely to benefit from single-sex schools—girls who prefer that environment—have devoted champions. Beverly Hibbler, the principal of the Detroit International Academy, lobbied to open an all-girls public school after working at a school for pregnant girls.
"We'd have girls pretend they were pregnant so they could come to our school," she said.
Her school now enrolls girls from outside its drawing area and offers mentorships and a successful robotics department. Ms. Hibbler said girls who had suffered sexual abuse found refuge in the all-girls environment of Detroit International.
"It's an important choice," she added.
Northwestern's Mr. Jackson is now working on a project to gauge the impact of single-sex schools on lower-income students, especially boys. A federal review from 2005 on single-sex schools noted that research is lacking on the impact of such programs on boys and disadvantaged students.
Vol. 31, Issue 17, Pages 1,12-13