Girls do better than boys on skills-based BTEC tests

examby albertogp123

Outnumbered science girls beat boys
[Via BBC News | Science/Nature | World Edition]

Girls who take certain skills-based science and technology qualifications outperform boys, suggest figures from an awarding body.


The Business and Technology Education Council in Britain administers a series of certification tests for various vocations. 

It turns out that, although females make up a minority of those taking STEM tests, the percentage that gain distinction is much higher than males.

For example, only 5% of the people taking the Level 2 engineering test were female but 37% gained a distinction while only 20% of the boys did.

They did about the same on the harder Level 3 tests.

Of course, this is probably more an example of self-selection. When so few take the test ( 810 girls did), it is much liklier that only the ‘best’ females do, with the majority taking other subjects.

That is, the extreme right of the bell curve is who takes the test. As more girls take the test, it will most likely revert to the same mean as boys.

So, I do not believe this yet shows that girls are smarter than boys but it does show that they are most likely just as capable.

2 thoughts on “Girls do better than boys on skills-based BTEC tests

  1. You can go a little further with this. Some people argue that the “right edge of the bell curve” is what actually matters in STEM performance. That is, what average students do doesn’t matter because exceptional students are the ones who make the most socially important contributions.

    It’s well known that girls on average *do* outperform boys on basic science and math. The above argument is often used to advance an argument that men are inherently better suited to science that women, even thought girls appear to be better suited than boys.

    This data suggests that, even if you accept this questionable theory, real data don’t give you much support.

    1. The questionable argument is partially based on the fact that males have a greater variance than women, to the right edge for them extends further than for women *as does the left edge). The problem arises from making the conjecture that socially (or scientifically) important contributions come only from those extreme numbers. I would expect other factors – like getting anyone to listen to your great ideas – may be as important (if not more so).

      In reality, it is not a good idea to separate out such complex things by a single axis or distribution. Perhaps at least 4 dimensions?

Comments are closed.