The idea that men would prefer college educated women because they can make more money after marriage seems unlikely. Marriage creates a monopsony job market for women because of their geographic immobility, decimating their future wages regardless of how skilled or well-educated they are. Employers know that women aren't able to move to another region to find a better job since they tend to stay with their husbands, so they are able to pay them much less. Economic research comparing women in cottage industries (where they can work for any employer from home) versus women working traditional jobs backs this up. If a man really wanted a women who would be able to contribute money to the relationship, it would be far more economically rational to look for a woman who was already wealthy rather than a women who could potentially earn money due to her education.
This was an interesting article but I am highly skeptical about that "96 percent," "over 96 percent" statistic. There are a slew of arguments about related problems that contradict many of the claims in this article.