April 19, 2004

O tempora, o mores

Rivka at Respectful of Otters offers a lucid explanation, based on sampling bias, for the fact that many teachers agree with Camille Paglia's jeremiad on the degraded state of today's youth.

Kids these days aren't like Paglia's friends in college were. But most kids at the time that Paglia was in college probably weren't like her friends. The educational path that ends in a Ph.D. and a professorship typically involves early segregation from average people, and when Paglia was growing up in the age of "ability tracking," that was even more true. Grad school selects for precisely the kinds of people who, as Paglia describes her college buddies, like to have intense discussions of literature at midnight. (Well, my program selected for people who liked to get drunk and mock inferior research methods, but the principle is the same.) No doubt there were plenty of other students at SUNY Binghamton in the 1960s who only cared about football and drinking and their future business careers, and who whined when they were assigned books that were too hard.

Paglia didn't have to hang out with those guys then, but now their direct descendants have shown up in her classes and she's stuck with them. The middle of the bell curve is much fatter than the ends, so she's probably got a lot more beer-and-Cliffs-notes kids in class than she has proto-intellectuals. That doesn't mean that America's youth have gotten dumber, it just means that she's being forced to encounter a wider spectrum of America's youth than she ever had to know when she was one.

At the University of the Arts, Paglia's classes may refer to IMDb more often than to Cliff's Notes, and may prefer other substances to beer. However, the principle is correct and applies to teachers in general, as well as to anyone else whose work forces them to deal with a more random sample of the population than their school-days affinity groups did.

I'm sure that there really are some significant cultural differences among people from different places and times. However, I like to think that my own specific beliefs about such differences (for instance between kids today and kids 40 or 50 years ago) are better supported by evidence. In particular, I try to be suspicious of generational generalizations based on nothing but my own personal observation, because of the sampling bias that Rivka describes, as well as the mythologizing effects of selective memory.

Posted by Mark Liberman at April 19, 2004 09:54 AM