A number of us who follow such things have been getting lots of questions about the news stories over the last two weeks reporting that research now proves that living together before marriage doesn’t cause divorce. A sampling of these stories are here, here and here. The headlines are stark and definitive: New Research Says Cohabitation Doesn’t Lead to Higher Likelihood of Divorce.
The reason this seems to be such big news is that a boat-load of published research over the past few decades by mainstream scholars has consistently shown that living together before marriage corresponds with a couple’s elevated risk of divorce. And dramatically so. This finding has been so significant and consistent that it’s earned itself a name among sociologists: the cohabitation effect. And the debate in the field has not been whether it exists, but why it exists. I give a thorough overview of this research in my recent book, The Ring Makes All the Difference.
So what do we make of this new study that supposedly proves that all this other research is wrong? A good place to start is to read the study, which would have been a good idea for the journalists who reported on it. The author of this article says unequivocally that her study proves “that cohabitation doesn’t cause divorce and probably never did.”
Before addressing the study itself, it is important appreciate a problem that the newspapers made in their reporting as well as this author in her quote above. In reading a great majority of all the published studies over the decades on the relationship between cohabitation and divorce, I don’t know of one that asserts that cohabitation causes divorce. The cohabitation effect asserts that premarital cohabitation seems to be associated with weaker, less healthy and shorter-lasting relationship both before and after marriage. As stated above, there has been little contest on whether the cohabitation effect is real and this study doesn’t challenge that fact. The big question among scholars is why; what is the linkage in this association? The author of this article, having done her Ph.D. dissertation in this field, knows the theory does not assume causation.
Her study does take an interesting and important approach to this topic in that it examines how age at cohabitation and later marriage impact the cohabitation effect as we know that couples who get married in their mid-twenties tend to have happier, more successful marriages. So it could be assumed that cohabiting couples who do marry at later ages would have lower likelihood of divorce than those cohabitors who married younger. And this is largely what she found. But note how it’s explained in her abstract:
“Analyses…revealed that age at coresidence [moving in together] explained a substantial portion of the higher marital dissolution rate of premarital cohabitors.”
She also begins her explanation in the conclusion of her article similarly,
“The findings discussed in this article indicated that the previously found association between premarital cohabitation and divorce in earlier decades can in part be attributed to the age at which premarital cohabitors began residing. These findings also suggest that the measurement of age has a considerable effect on the observed relationship between cohabitation and divorce.” (p. 366)
So what is she saying? That when you adjust for age at cohabiting and subsequent marriage, we find the risk of divorce is reduced. It is an interesting finding and not altogether surprising. And it clearly admits and affirms the cohabitation effect. But what it does NOT say – anywhere in the study, implied or outright – is “that cohabitation doesn’t cause divorce and probably never did” as stated in the author’s press release from which most journalists based their stories.
And what else must be appreciated is that in most sciences, one new study – or even a small handful – does not overthrow the larger body of data that exists on a particular point. But this is precisely how most journalists played it: Newest study owns the field. A contrary finding must be replicated by many more studies over a longer period of time as we have in the literature leading to the truism of the cohabitation effect. As time goes on and more research is done regarding age at cohabitation, we might find that this study is onto something.
But not today. There is no research that even hints at the conclusion that cohabitors as a whole do not face a greater likelihood of relational instability and divorce once they do marry.