Found this study via Dylan William and well, it makes you think. Basic insight? A student’s share of opposite gender school friends negatively affects high school GPA, more girls surrounding a boy and vice versa will reduce the academic results. Before you start shouting single sex schooling, do note that the effect is more the case for students older than 16.  If younger, the negative effect is only present for maths & science. But it’s not really the school composition who has an effect, rather the friendships:

This paper finds that it is whether school friends are the same or opposite gender rather than male or female that matters for school performance, which distinguishes it from the grade composition literature that generally finds that the share of female schoolmates is positively related to achievement. Despite the relative imprecision of the gender-specific estimates, the null hypothesis that the effect of girls on boys has the same magnitude and opposite sign as the effect of boys on girls is tested and rejected ( p-value: 0.06), indicating that peer gender composition effects for school friends are different to peer gender composition effects for same-grade schoolmates.

But still:

…”these results suggest that an increase in the share of opposite gender school friends negatively affects classroom behavior. High school students appear to be distracted by opposite gender friends in the classroom. This may negatively affect teacher-student relationships, possibly as teachers try to deal with distracted students, and may also reduce the quality of interactions with other students.”

So this study seems indeed to be a new element in the discussion and gets other results than the ones also mentioned in our book because of the novel way of approaching the subject. The researcher, Andrew Hill, concludes:

This paper provides new evidence on the impact of the gender composition of school friends: an increase in the share of opposite gender school friends causes a reduction in high school academic achievement. Part of this effect seems to operate through changes in classroom behavior such as increased troubles getting along with the teacher and paying attention in class. Opposite gender school friends also increase the probability of being in a romantic relationship, which may be another mechanism for the negative effect on GPA given students in relationships may both substitute time away from studying and be distracted in the classroom. These results speak to the continuing debate around single-sex and mixed gender education (Halpern et al. 2011; Pahlke, Hyde, and Allison 2014). Reorganizing classroom gender composition is a relatively low-cost policy as it need not require more teachers or resources. The difficulties getting along with the teacher and paying attention in class may be eliminated in single-sex classrooms that exclude opposite gender friends. The negative effects of opposite gender friends for younger students are also found in mathematics and science and not in English and history, providing a dimension of empirical support for educators using single-sex mathematics and science classrooms in mixed gender schools.

 

Abstract of the paper:

This paper finds that a student’s share of opposite gender school friends negatively affects high school GPA. It uses the gender composition of schoolmates in an individual’s neighborhood as an instrument for the gender composition of an individual’s self-reported friendship network. The effect occurs across all subjects for students older than 16, but only in mathematics and science for younger students. Additional results indicate effects may operate inside the classroom through difficulties getting along with the teacher and paying attention, and outside the classroom through romantic relationships