Thinking about Goodbye's post: what matters is what prepares girls best for the society that they are going to enter, and the world of work that they are best suited for.
So if the school equips girls with the STEM qualifications that they need for engineering, but none of the other skills that they might need in that world (including the ability to work effectively in mixed or otherwise all-male teams), then that is not preparing them fully.
But equally, a school that prepares its girls very well for working in mixed or all-male teams but which explicitly or implicitly makes it less likely that they would study STEM subjects is also not preparing them fully.
I do also agree with PP who suggest that digging deeper into the data to compare like with like would be informative, as girls' schools are more common in some sectors rather than others.
So e.g. STEM subject takeup in:
- Girls' grammars vs fully mixed grammars
- Girls genuine comps vs fully mixed genuine comps
- Girls' private selective vs fully mixed private selective
- Girls' private non-selective vs fully mixed private non-selective (I realise that all private schools are selective by parental income, but i mean those where the entry exams are a formality or where there are few applicants per place)
- Girls' secondary moderns (the 'other' school in partial or full grammar counties) vs fully mixed secondary moderns.
- Girls' faith schools vs fully mixed faith schools
i suspect the fact that many girls' schools are private, grammar or the other' school in grammar counties [confusingly sometimes called comprehensives, but without a comprehensive intake], or may be faith schools, can skew the data, because the vast majority of true comprehensives are mixed.