How about this?
Some highlights that illustrate EXACTLY THE POINTS I WAS MAKING EARLIER:
"This pattern of results supports the contention that self-selection into the two pathways to marriage helps to explain some of the difference in marital separation rates."
"The analysis also indicated that the elevated risk of marital separation linked with indirect marriages is disappearing with more recent marriage cohorts._ For example, before any controls were entered, the difference between direct and indirect marriages in the probability of separation was 17 per cent for those who married in the early 1970s and 5 per cent for those who married in the early 1990s._ When the controls were introduced, the difference in probabilities of separation for direct and indirect marriages fell to 11 per cent for the earlier cohort (1970–74) and to –1 per cent for the recent cohort (1990-94)."
---
You should read the conclusions, and you'll see that addressing my points would have been fruitful - these fine scientists introduced the control elements I asked for, and found the differences decreased (although some cases retained significance, the actual figures were astonishingly low) or DISAPPEARED ENTIRELY. Bottom line: if you're not living together before marriage simply because you're afraid of divorce, you might want to check your figures.
I am far too busy to jack around with this, but I can't help but deny you the banal "where's your figures????" cop out.