Premarital cohabitation, marital quality and risk of divorce: Testing the selection hypothesis

Research in the past decade has centered around questions concerning the impact of cohabitation on marriage. A number of studies confirm that people who cohabit prior to marriage report lower levels of marital quality and experience higher rates of divorce than those who do not cohabit prior to marriage. Researchers have failed to agree, however, on why marriages preceded by cohabitation tend to be less successful. One possibility is that people who cohabit are also the same people who are more likely to divorce (i.e. Selection Hypothesis). It is also possible that the experience of cohabitation increases the risk of divorce, perhaps by causing changes in values and attitudes which are associated with it (i.e. Causation Hypothesis). Determining what causes greater instability of marriages preceded by cohabitation is important for many reasons. Studies show that many cohabitating relationships involve children who may suffer the consequences of family disruption. Policy makers and professionals can benefit from this research as they strive to create family policy which will strengthen marriages and also try to help troubled relationships. Although previous work has examined areas of difference among cohabitors and noncohabitors, we still lack information as to why these differences exist. The purpose of this study is to compare these two explanations for differences in marital quality between those who cohabit and those who do not. Data are taken from the National Survey of Families and Households, a nationally representative survey of adults age 19 and up, in the United States. The sample for the present study includes three-hundred-sixteen respondents age 35 and younger, who were never married and had never cohabited at the time of NSFH1 and who were married, divorced or separated from their first marriage at NSFH2. In this sample, evidence that cohabitation is associated with reduced marital quality is weak to nonexistent. Cohabitation was not a significant predictor of marital happiness, marital interaction or marital instability. The only outcome measures significantly related to cohabitation were marital disagreements, marital success and the index of marital instability. In the cases of these variables, some seems to be explained by selection. I conclude that the selection hypothesis may become less useful as an explanation for lower marital quality and higher rates of divorce among cohabitors as younger cohorts of couples continue to cohabit at high rates. While cohabitation appears to be related to levels of discord in a marriage, it does not appear to be related to measures of marital happiness. Future research should examine the reasons why couples choose to cohabit, as well as the personal characteristics of partners in cohabiting relationships. These factors could have an impact on the level of disagreements among cohabiting couples