Headmasters’ Blog – Co-Ed Debate

Co-ed versus single-sex education: a debate without substance.

Given that our Senior School has recently taken the decision to admit boys with effect from September 2019, it seems timely to put down my thoughts on the long-standing educational and social debate of whether co-ed or single sex schooling is preferable. When considering any question such as this, the rational thing to do is to look at as wide as possible a range of empirical research. So, what does the research tell us?

The weaknesses of the research

It is commonly understood that much educational research has pretended to a scientific status it does not really enjoy, this is due to two factors:

1. The fact that much ‘research’ sets out to present existing data in order to support conclusions already reached on instinctive/ideological or business grounds
2. The fact that (as with much social scientific research) it is harder to achieve the necessary level of control than with natural scientific research. In the case of comparisons between single-sex and co-educational schooling, it is notoriously difficult to effectively filter out all other significant factors (such as culture, class, ethnicity, academic ability and so on) across a wide enough set of data

What is clear about the single-sex versus coeducational debate is that both sides rely heavily on ‘evidence’ and arguments that do not bear close analysis.

The weaknesses of single-sex arguments

1. Many single-sex arguments rest on very old sources. The classic being Myra and David Sadker’s Failing in Fairness: How America’s Schools Cheat Girls (1992). Drawing on sources such as these ignores nearly three decades of profound changes in social attitudes and behaviours. It also ignores the fact that much of middle America is actually more likely to conform to ‘traditional’ gender expectations than many European societies.
2. Many arguments point to problems that exist anyway, and are down to the quality of teaching. For example, many will raise the issue of inclusion (that is, the risk that boys may dominate in the classroom), ignoring the obvious fact that someone (an individual or a group) will always dominate every classroom if allowed to do so. The answer to this problem is for the teacher to control who answers questions and how they answer them, to ensure that they set up processes that ensure inclusion, and to ensure that they respond to all answers (good or bad) in a fashion that encourages further learning.
3. Many other arguments rest on clumsy misunderstandings of human behaviours and what causes them. For example, the tendency of girls to spend a lot of time worrying about their appearance is assumed to disappear under single-sex conditions. It doesn’t. The problem is the underlying assumption that girls ‘preen’ for boys when it is actually more probable that they ‘preen’ for (or, rather, against) other girls.
4. Some arguments have a tendency to reach for spurious generalisations allegedly ‘based’ on neurological science or (worse) evolutionary biology. All the literature currently expresses the consensus that (as far as researchers can currently tell) differences in neurological development between the sexes are much less pronounced and much more nuanced than had previously been assumed. The only clear difference seems to be that girls develop the capacity of ‘preferential detachment’ (an ability to share information across different areas of the brain – which presumably aids powers of synthesis and sophisticated evaluation) about a year or two ahead of boys
5. Most importantly, these arguments have a dangerous tendency to ignore the fact that all statements about boys and girls are generalisations. Even if you accept that there are some general behavioural differences between boys and girls, a significant minority of girls and boys will fail to conform to these expectations. Therefore, if a single-sex school commits itself too firmly to the concept that it can save its teachers a differentiation challenge by focusing firmly on certain styles of teaching it will almost definitely be failing a significant proportion of its pupils. In short, the assumption that single-sex schooling can lead to an effective reduction in personality based differentiation challenges does not bear close scrutiny.

The weaknesses of co-ed arguments

1. Proponents of co-ed systems have an alarming tendency not to address the actual issue of quality of education and educational outcomes. They tend to be more concerned with wider social issues. Thus many papers arguing for co-education rest more on the argument that single-sex schooling perpetuates sexism in the wider society than on any analysis of educational outcomes.
2. Like their opponents (who assume attitudes among pupils and teachers that are thirty years out of date) proponents of co-ed education tend to imagine schooling within the context of a world long gone. Thus, the classic argument is that single-sex schooling fails to prepare young people to work, live and socialise with the opposite sex. This seems to assume that all single-sex school pupils board for whole terms, have no boyfriends or girlfriends, no social network beyond school, and never undertake co-ed enrichment or co-curricular activities.
3. As with their opponents, co-ed proponents like to select data-sets precisely designed to prove their points (the favourite for co-ed–proponents being Scandinavia).

So, what practical conclusions can we actually draw?

Here, the last word should go to John Hattie, author of the most widely respected overview of research into the actual effects of different factors on school-age learning, and to the February 2014 American Psychological Association analysis of this issue which (to my knowledge) is the largest and most robust meta-analysis ever carried out on single-sex versus co-ed schooling.

Hattie’s conclusions are that…

“The variability between boys and girls is very large – and much, much greater than the average difference between boys and girls”

And…

“Overall, there is very little compelling evidence of a compositional effect related to whether a class is single or mixed-sex … There are more powerful effects due to the quality of teaching and teacher expectations than to whether a class is one sex or mixed.”

The American Psychological Association meta-analysis report (in the February 2014 issue of the journal Psychological Bulletin) concluded that, both in terms of learning outcomes and pastoral wellbeing “there’s basically no difference” between pupils in co-ed and single sex environments and that any advantages accruing from single-sex education “are trivial and, in many cases, non-existent.”

In short, the most robust and wide-ranging research points to an educational effect in favour of single-sex schooling of zero to marginally small. Any claim by a school that its gender make-up (whether single-sex or coeducational) is a significant factor in the educational performance of its pupils should therefore be approached with considerable scepticism.