Why are women underrepresented in Science, Technology, Engineering, and Math  (STEM) careers? I’ve posted a couple of articles about this, which taught me that people have very strong opinions about the topic.

Unfortunately, this is one of those areas where the strongest opinions and the loudest voices come from personal experiences. Detailed academic study of the problem often takes a back seat to arguing about life histories.

But this is not to say that there aren’t plenty of people working on the problem.  A recent study from the University of Chicago highlights just how intractable this condition might be.

Catch-22

The Chicago group looked at boys and girls learning mathematics in first and second grade. What they found was that having a female teacher with math anxiety affected girls’ performance, but not boys.  Because 90% of elementary school teachers in the US are women, the study posits that we have a self-perpetuating problem with women and math: anxious female math teachers produce underperforming girls.

Of course, if the study just presented the simple facts it would be pretty boring, so by necessity they do some interesting extrapolation:

Teachers' anxiety might undermine female students' confidence in learning mathematics throughout their years of schooling and also decrease their performance in other subjects, such as science and engineering, which are dependent on mathematical understanding.

I find their recommendations long on good intent, but short on specifics:

The authors suggest that elementary teacher preparation programs could be strengthened by requiring more mathematics preparation for future teachers as well as by addressing issues of math attitudes and anxiety in these teachers.

The question of “addressing issues” is where I have the big problem.  It basically says that to solve the performance problems of women in mathematics, we need to solve the performance problems of women in mathematics.  Although we often solve problems recursively in mathematics, this doesn’t seem to be a fit.

No Answers, More Questions

There are lots of anomalies in the STEM problem – one would be medicine, where each year the US med school graduating class has far more men than women. Why do the risk factors that affect engineering and mathematics not apply to medicine? Is that teacher bias and anxiety that is so measurable in mathematics absent in science class? What if turns out that more science classes are taught by men?

And what of computer science? This career choice has a large imbalance, with evidence that it is growing. CS degree programs are not exactly math-heavy, and the career itself can be essentially math-free. There is clearly more to the problem than just arithmetic.

I think there is only one piece of safe footing in this whole area of research: whatever your pet theory is about gender imbalance in STEM, there exists data that shows it is incorrect, incomplete, or just plain wrong.