Saturday, March 25, 2006

The Hopefully Short-Lived Affirmative Action Theory of Male / Female Graduation Rate Gaps

John Tierney has a column($) in the NYTimes today focused in part on the under-representation of men in higher education. This is a real issue, and one where gender imbalance appears to be increasing. (Although after centuries of patriarchy and oppression, it's amazing how quick people are to hit the panic button when any trend starts to run women's way. Last I checked men still, you know, ran the world.)

However, Tierney's take on this issue is strangely peripheral, focused on a recent op-ed from Kenyon College detailing their attempt to rectify the imbalance through gender-specific admissions policies, i.e. affirmative action for men. Tierney says:

"It's not fair to the girls who are rejected despite having higher grades and test scores than the boys who get fat envelopes. It's not fair to the boys, either, if they're not ready to keep up with their classmates. Affirmative action just makes them prone to fail, and is probably one of the reasons that men are more likely than women to drop out of college."

"Probably"? Really?

Fortunately, this is a testable hypothesis. Using data from The Education Trust's College Results Online data tool, we can calculate the graduation rate gap between men and women at every 4-year institution in the nation. We can then group them by admissions selectivity using the popular Barron's Guide ratings. Here's the median male/female graduation rate gap (measured as the female graduation rate minus the male graduation rate) for each selectivity category, in descending order:

Most Competitive: 2.8%
Highly Competitive: 3.9%
Very Competitive: 6.0%
Competitive: 8.5%
Less Competitive: 9.7%
Non-Competitive: 7.2%

Among the "Most Competitive" institutions, for example, female students at the institution with the median gap had a graduation rate 2.8 percentage points higher than male students. The gap favors women in every other category as well. But if affirmative action was really a cause of increased disparities in male/female graduation rates, you would expect to find the largest gaps at the most competitive schools--that is, the relatively small number of schools that have selective admissions and are thus in the position to have an affirmative action policy, gender-focused or otherwise.

In reality, the data show almost exactly the opposite. With the exception of the relatively small number of institutions in the "Non-Competitive" category, tighter admission policies--and thus, opportunities to favor men--are associated with smaller male/female graduation rate disparities. The greater the selectivity, the smaller the gap.

In fact, a perfect example of a highly competitive institution with a below-average male/female graduation rate gap is...Kenyon College, the very institution Tierney cites. Less than three percentage points separate men from women there, which is less than half the norm.

Any gender-based disparity is problematic. But theories as to why should be based on actual data, particularly when it's right there for the asking.

No comments: