A friend sent this article the other day, about the factors which predict whether a marriage will last.
One factor is how long you were dating before the marriage. Unsurprisingly, the longer you dated, the less likely you are to divorce.
Another factor is how much money you make. The more (up to a level of $125,000+ a year), the less likely the marriage is to end up in divorce.
The frequency with which you go to church is a little more complicated. People who attend "sometimes" are 10% more likely to get divorced than those who never go, but those who attend "regularly" are 46% less likely.
Your attitude toward your spouse is also important: men who care more about their wive's looks are more likely to end up divorced, and women who care more about how much money their husband makes are more likely to as well. I have to wonder how they measured this. I would think the majority of wives would care about what kind of provider their husband would be, and the overwhelming majority of men care about what their wives looked like.
The number of people who attended the wedding is also a factor: the more who attend, the less likely a divorce. But here's a surprise: the more you spend on your wedding, the more likely a divorce. (One would think there would be a positive correlation between the number of guests and the cost of the wedding. One would also think there'd be a correlation between income -- a positive indicator -- and the cost of the wedding.)
Couples who had honeymoons are less likely to divorce. (That's probably correlated with how much money there is.)
In any case, all those factors are interesting, but I can think of several more which should have a stronger impact.
The greater the difference between the spouses' IQ's, the more likely a divorce. One will feel bored and disgusted, the other condescended to and resentful.
The frequency of sex, both early on and later, must correlate. My guess, couples who has sex frequently early on are slightly more likely to remain together. (But if sex is all that the marriage is about, that could have a negative correlation with stability.) But I would think the far stronger correlation would be with couples who continue to have regular sex past the ten year mark: they are more likely to remain together since it means they can still stand each other.
I would think the age at which the partners married would have a lot to do with it: couples who marry early, in their impetuous youths, would seem more likely to grow apart as they age and (theoretically) mature.
The number of children in the family must have an extremely strong effect on the stability of a marriage. When was the last time you heard of a couple with six children getting divorced?
The existence of a mental disorder must also have an extremely strong negative impact. It's very hard being married to a sociopath, or someone who's bipolar, or has borderline disorder. I've heard that 80% of marriages where one partner has Aspergers end in divorce, though I can't confirm that. And even the presence of a garden variety narcissist must the marriage difficult for the partner: it's hard being married to someone who will never, ever admit he's wrong.
I would think the strongest negative indicator must be the existence of a prenup, which presupposes that the marriage was more rental than purchase, right from the start.