It’s a shocking and stubborn problem: Women only earn about 20% of the bachelors’ degrees in engineering, computer science and physics. These fields account for 19 of the top 25 college majors ranked by salary by the Georgetown University Center on Education and the Workforce, shown below.
So women have little access to our economy’s most lucrative careers.
Why is this happening? Lara Perez-Felkner, an Assistant Professor in FSU’s Higher Education Program and in the Department of Sociology, has the best answer I’ve seen. Many high school girls with strong math skills don’t believe they are good at math. And that belief keeps them from choosing college majors in engineering, computer science and physics.
The primary result is illustrated below in a slide from a talk on this subject that Lara delivered to the FSU Physics Department last week.
In each plot, the x-axis shows students’ objectively-measured math abilities as determined on a test administered in 10th grade. The y-axis gives a measurement of students’ perceptions of their own math abilities. The gap in perceptions between girls and boys of identical objectively-measured math abilities in 10th grade is shocking. That gap in perceptions narrows somewhat by 12th grade, but it is still very large between boys and girls having the greatest math ability.
Lara goes on to show that the perception of math ability plays a very large role in whether a student decides to major in engineering or computer science. Since boys believe they are good at math and girls don’t, boys choose those majors at much higher rates.
The intervention required to “solve” this problem is to convince girls that they have strong math skills and to encourage them from a young age – starting at middle school – to take high level math and science courses.
Lara’s entire Physics Colloquium talk from last week, which is chock full of statistics on women in engineering, computer science and physics as well as more results from her research on girls’ math ability perceptions, can be downloaded here: