Assuming you're comparing arranged marriages to "love marriages" I wouldn't even bother. In the US the divorce rate is somewhere near or above 50%.

Logically "love marriages" make plenty of sense but the data proves (in the US at least) that it doesn't work. At least not once the human factor is taken into account.

People from cultures where arranged marriage is not acceptable seem to have a very hostile view on it. There's good and bad but having come from that culture as I grew older I began to see the benefits.

My point is, I'm not aware of any evidence that arranged marriages are in general a crappy deal for both sides. I found that paper while looking for a full text of this one http://psycnet.apa.org/psycinfo/1985-19991-001 which demonstrates that objective measures of love increase over time with arranged marriages and decrease with marriages for love. These papers have small sample sets because they're exploratory and pointing directions for future research, which as far as I can tell, never happened on a large scale.