Some majors tend to lend to a lot of money. Some majors don’t. For some reason females tend to cluster in certain majors and males in others. Using data from the 2009 American Community Survey (Aside: I’m so excited that they finally added major to this survey! I really think education research makes a great deal more sense once you recognize not all majors have equal market impact.) I looked at the concentration of majors by gender for people 25 to 35.
These pictures give rise to a couple of questions. Why do women choose to major in such low paying fields of study? (The top fifteen female majors earn, on average, 39% less than the top 15 male majors.) Do they value non-pecuniary benefits more than men? Do they plan on not working? Why are male majors so tightly clustered? (The top fifteen majors for men account for 52% of all men, while the top 15 only account for 29% of women.) Do men focus on the “money majors” while women pursue a wider variety of educational inputs?
Furthermore, this doesn’t appear to be just discrimination or social pressure. I think most people would agree that women have been increasingly less discouraged from entering male dominated fields. But when you look at majors by age it looks like female participation in the highest paying fields the increase has been flat for 50 years.
High paying majors are frequently more math-y than low paying ones so you could claim that this is a result of men having slightly higher average math scores and higher variance. But I don’t think that brings much to bear here since the difference in variance only creates large disparities at 3 or 4 standard deviations which I don’t think is enough to matter at the level of undergraduate degrees. Maybe it is. I guess we could look at SAT scores and planned majors but I certainly moved around a lot from freshmen to graduate, and so did most of the people I know, so I’m not sure how convincing that data would be.
Males do seem to score higher on the SAT (average of 500 for females and 534 for males) but I don’t know how much of the disparity it can explain.
Does anyone have any other ideas that might explain these data?