Science vs. single-sex classes

Science Doesn’t Support Single-Sex Classes, argue Rosalind Barnett and Caryl Rivers in Education Week.

The loud, hissing sound you hear may be the air coming out of the tires of a much-hyped vehicle for improving American public education: the single-sex classroom.

. . . A consensus is emerging among scientists that single-sex classrooms are not the answer to kids’ achievement issues. This fact appears to be true even for students of color, who are often seen as those most likely to be helped by sex-segregated classrooms.

In The Pseudoscience of Single-Sex Schooling, published in Science, eight psychologists and neuroscientists “found the rationale for setting up separate classrooms for boys and girls ‘deeply misguided’ and ‘often justified by weak, cherry-picked, or misconstrued scientific claims rather than by valid scientific evidence’,” Barnett and Rivers write.

About Joanne


  1. “All authors are founders and uncompensated board members of the nonprofit American Council for CoEducational Schooling”

    While the obvious possibility of bias doesn’t automatically rule out their conclusions, it does raise a red flag.

  2. ‘often justified by weak, cherry-picked, or misconstrued scientific claims rather than by valid scientific evidence’

    So, just like every other educational fad, then?

  3. Science can’t quantify personal experience. As a teacher in a Catholic school which had occasional separation, the atmosphere was definitely positive. But it doesn’t and wouldn’t show up in testable data.

  4. Yes, because clearly all kids are identical and interchangeable. There is no possibility that some kids might benefit from single-sex education and some might not. The one thing we must never permit in primary education is any diversity of approach.

  5. I agree that single-sex classrooms may not significantly affect academic achievement.

    What they do affect, however, is classroom management. Especially in middle school, the interactions between the sexes, and because of the sexes, becomes less of a problem.

    Boys act dumber in the presence of girls. Most of the really stupid things they do are to impress the girls.

    Ditto, these days, for the girls. Their histrionics are because of the presence of boys. Keep them separate, and the Diva act disappears.

    • But if mixed-sex classrooms have greater problems with classroom management, shouldn’t this indirectly negatively affect academic achievement? Or do you believe that issues with classroom management have no affect on academic achievement?

  6. If parents want single-sex classrooms or schools, then the burden of proof should be on opponents to show that outcomes are significantly worse in them than co-ed classrooms or schools. The default should be towards greater parental choice.

  7. This is part of an ongoing attempt by this group to set up a bullying “scientific consensus” on the question of single-sex classes. Not much scientific about it, but it’s very political.

    The study of Trinidad and Tobago students, mentioned in the edweek article, is interesting to look at. I can only access the summary, but these are the highlights:

    ? This is the first study to credibly identify the causal effect of single-sex schools on student outcomes. ? Single-sex schools have no effect on over 85% of students. ? Single-sex schools have large positive effects only on students with strong preferences for single-sex schools. ? Some girls benefit from single-sex schools but boys do not. ? Girls at single-sex schools take fewer math and science courses.

    This is Barnett and Rivers’ report of the same study: The data show that while sex-segregated classrooms may benefit a subset of girls, they don’t automatically benefit all girls and boys. (The girls who did well were those who wanted to have an all-girls class.)
    The research team was headed by C. Kirabo Jackson, a labor economist at Northwestern University, who analyzed data on 219,849 students from 123 schools to find out whether single-sex schools improved student performance between 6th and 10th grades. The study found that students in all-girls schools were slightly less likely to take math or science courses. Given the fact that such courses open the door to lucrative careers in science, technology, engineering, and math, or STEM, fields, this fact is disturbing.

    “Large positive effects” for the 15% of the population who want to attend single-sex classes seem worthwhile to me. I suspect much improvement in education does arise from finding environments for learning for particular students.