This New York Times article describes an interesting development:
Throughout the 1980s and early ‘90s, women of all economic levels — poor, middle class and rich — steadily gained ground on their male counterparts in the work force. By the mid-‘90s, women earned more than 75 cents for each dollar in hourly pay that men did, up from 65 cents 15 years earlier.
Largely without notice, however, one big group of women has stopped making progress: those with a four-year college degree. The gap between their pay and that of male college graduates has widened slightly since the mid-‘90s.
There are two major problems with the article, and it is likely due to either the reporter analyzing the study incorrectly or the study itself being flawed.
First, the article only seems to lump “female college graduates” against “male college graduates.” Obviously there is a wide spectrum within these two groups — particularly with regards to what an individual studied. Most education majors are women while most future engineers are men. (The latter profession earns much more than the former.) If the study did not take this into account — and we cannot know for sure — then the data is meaningless.
Secondly, any study comparing the wages of men and women is going to provide inconclusive data. One cannot average the salaries of all male [insert profession here], and then compare that with the average salary of their female counterparts and reach a valid conclusion.
Here’s why: to state that gender itself is the reason that women are earning less, you need to isolate that as the only variable in a study. You need to take into account level of education, prior experience, work performance and total hours worked, along with a near-infinite number of other factors that come into play with regards to salary. In other words, if everything is the same except for gender and a difference in salary exists, then one can state that gender is the cause.
If a study can isolate gender — in other words, if it can control for every other factor that helps to determine salary — then that will be a report to take seriously. But I have yet to see a study that does that.
