[3JPLW]

Man, they took the least-informative numbers from that article. Much more interesting is the difference from their predicted divorce rate (which corrects for race, gender, and income, but not age, unfortunately).

Quite a few of the engineering professions have 50-60% the divorce rate of the national average, but after accounting for the above three covariances it's not quite as striking (putting them at 60-90% of expected). Agricultural engineers are still way down at 12% of expected, but they don't list the N for non-LE occupations.

Here's the actual article, but it's behind a pay wall. http://link.springer.com/article/10.1007%2Fs11896-009-9057-8

Google scholar has a link to a publicly accessible version: http://faculty-course.insead.edu/popescu/UDJCore/2010/handou...