I stumbled upon a great piece by John Tierney in The New York Times. Tierney summarizes research on gender differences measured across cultures.

Four take-away points:
  1. Men and women answer personality tests differently.
  2. The differences exhibit variation across cultures. In other words, men and women answer personality tests differently, but these differences vary by culture; the differences aren't uniform. 
  3. The magnitude of the differences is smaller in traditional, agricultural cultures than in modern, post-industrial cultures. In other words, men and women answer personality tests more in the same way in agricultural societies, but more differently in post-industrial societies.
  4. We can posit a change over time: as societies urbanize, men and women answer personality tests in increasing different ways.  

There has been considerable research in this area (Tierney offers some references), most recently an article (highlighted by Tierney) in the Journal of Personality and Social Psychology by David P. Schmidt, Anu Realo, Jüri Allik, and Martin Voracek. What struck me most in this scholarly article was the researchers' use of the Big Five Inventory, which measures personality traits such as "agreeableness" or "extraversion", not more popular notions like "caring" or "competitiveness".

I confess that much of the article went over my head, but I'm uncomfortable with Tierney's reference to a gender "gap". I'm also puzzled whether differences in personality tests answers really correspond to differences in personality. In other words, I'm not sure just what this research measures and compares, and how much can be read –especially by the lay reader– into the results.