The main point is not necessarily cohabitation per se. It is the amount of cohabitation and sexual partners.
According to the study women who have serial cohabitations or more sexual partners are more likely to divorce than the women who have only cohabitated with their husband or slept with their future husband before marriage.
Correlation is not causation. And among those who do not cohabitate or have any other sexual partners than their husband, is a large proportion of those who will also never get divorced.
If you are of a culture where that is so, very good!
If you are not of a culture where that is so, the correlation isn't really applicable to your situation as it doesn't differentiate.
And you would perhaps be amazed at how many people think statistical analysis is science, especially statisticians!
You would also perhaps be amazed at how many people think that correlation indicates causation, especially statisticians! I mean, really, they don't have much else. When all you have is a hammer, problems look like nails.