We all walk around with theories in our heads that might just be wrong. One of my own theories, which I always believed was underpinned by rigorous research, is that girls perform better in all-girl schools and that boys perform better in co-educational settings – and worse in all-boy settings.

This is an issue for some parents and for society, but the surprising thing is that there are few controlled studies that compare single-sex and co-educational schools.

I recently discovered that my theory may not be supported by enough credible studies, including a gold-standard 2014 meta-study by Pahlke, Hyde and Allison in the prestigious American Psychological Bulletin. Titled “The Effects of Single-Sex Compared With Coeducational Schooling on Students’ Performance and Attitudes”, it involved 1.6 million children in 21 countries.

These researchers set out to use meta-analysis to synthesise the results of research comparing single-sex with co-educational schooling regarding multiple student outcomes, including mathematics and science performance and academic performance in other areas, as well as motivation, interest and attitudes.

In the US, like in many other countries, experts decry the poor performance of children in standardised tests of mathematics and science knowledge, by comparison with their peers from other nations. One of the solutions being explored is single-sex classrooms or schools. Some public schools in the US have already adopted this potential solution.

Many proponents of single-sex education believe that separating boys and girls increases students’ achievement and academic interest. Other proponents take the view that regardless of the effects of single-sex schooling, single-sex schooling should be available as an option for interested families.

Another worry is that boys’ sexist attitudes and behaviours decrease girls’ interest in traditionally masculine fields like the sciences.

In the American context, the question of whether single-sex schooling improves student outcomes is important, particularly because it is expensive and cumbersome to implement in public schools.

Proponents who believe single-sex schooling increases students’ achievement and interests draw on several perspectives to support their claims, the most prevalent being views that gender differences in psychological characteristics relevant to learning are substantial and either biological or social-psychological in nature.

From the biological difference perspective, some supporters of single-sex education argue that boys and girls do better when they receive instruction that is targeted towards the substantial, biological differences that they believe exist between boys and girls.  

The other supporters, including myself, of single-sex schooling hold the “girl power” view and relate the problem of domineering boys in co-educational classrooms as a reason for separating boys and girls. The argument is that in co-educational classrooms, boys tend to seek out and receive most of the teachers’ attention.

Read more in Daily Maverick: A longer school day or shorter holidays? There are ways to improve pupil outcomes

Another worry is that boys’ sexist attitudes and behaviours decrease girls’ interest in traditionally masculine fields like the sciences. The reasoning is that in single-sex classrooms girls develop self-confidence in mathematics and science, i.e. single-sex classrooms are empowering to girls.

In the gold-standard meta-analysis, the researchers synthesised research on the effects of single-sex compared with co-educational schooling on a wide array of variables, including mathematics performance and attitudes, science and verbal performance, gender stereotyping, self-concept and interpersonal relations. The quality of the studies was addressed by coding studies as controlled (random assignment or selection effects were controlled in some fashion) or uncontrolled (no random assignment and no controls were included for selection effects).

The effects of variables such as dosage (class vs school) and age were examined.

The evidence

The findings show that single-sex schools make little or no difference. When one looks at the results for the controlled studies (those that used the best research methods), single-sex schools produced only trivial advantages over co-educational schools.

There was little evidence of an advantage of single-sex schools for girls or boys for any of the outcomes.

Why do advocates for single-sex schooling believe that it has such positive effects when the data suggest otherwise? Many reasons are possible.

Read more in Daily Maverick: AI can help us find some answers in the elusive quest for ‘quality’ in education

Based on mixed-effects analyses, uncontrolled studies showed some modest advantages for single-sex schooling for both girls and boys, for outcomes such as mathematics but not science performance. Controlled studies, however, showed only trivial differences between students in single-sex schools vs co-educational for mathematics and science performance, and in some cases showed small differences favouring co-educational schools (for girls’ educational aspirations, for example).

Separate analyses of American studies yielded similar findings. Results from the highest-quality studies, then, do not support the view that single-sex schooling provides benefits compared with co-educational schools.

Theory sticks

Even with this magnificent study, it’s difficult to dispel the theory stuck in my head. I remember a comprehensive study by Lee and Bryk in the 1980s in which they showed that “single-sex schools deliver specific advantages to their students, especially female students”.

They argued that separation in schools may “actually facilitate adolescent academic development by providing an environment where social and academic concerns are separated”.

Australian researcher Professor Gary Falloon points to a more recent Queensland study that looked at the confidence levels of boys and girls attending single-sex high schools and found no difference between the two. “The study provides the first large-scale study that demonstrates that women are no less confident than men under conditions where gendered structures are mitigated by their environment,” Falloon says.

“What was really interesting, as a general rule of thumb, was that as kids got older and went through schooling systems, all the markers that show differences between the genders narrowed to almost nothing by the time they got to year 11.

“However, in primary school, boys had marginally better self-efficacy than girls did – that is, the sense of belief in one’s ability to do something, which we know is a strong determinant for achievement. But when they got to the middle-school in years 7 and 8, the girls had started to catch up.”

When it comes to private single-sex schools, there are perceptions that children will have greater opportunities. Falloon argues “that may be the case, but not necessarily because they achieve better, but because within those environments children may have connections and networks that can facilitate access to certain things that may not be available to others”.

Falloon says it comes down to personal choice, what works best for individual children and which social outcomes are valued. “My personal opinion on this, if your son or daughter is thriving in a single-sex environment, then don’t feel you have to change to co-ed and vice versa. When it comes to individuals, both school types have their place.”

Despite this research, I still have a strong belief (perhaps not that securely positioned) that in a country with such high levels of gender-based violence like ours, it must be better for girls to be educated separately.

I still stubbornly cling to my “girl power” view that girls perform better without boys present – particularly from my anecdotal experience in Catholic all-girl schools, in these safer spaces. DM168

Dr Mark Potterton is principal of the primary school at Sacred Heart College and director of the Three2Six Refugee Children’s Education Project.

This story first appeared in our weekly Daily Maverick 168 newspaper, which is available countrywide for R25.