The link between partnership and mental well-being: When cohabitation is common, does marriage still matter?

The decline in marriage and increase in cohabitation raises questions about whether marriage still provides benefits to well-being in mid-life. We use the British Cohort Study 1970 to examine the extent to which partnership experiences influence positive mental health at age 42. Cohabitation has become common for this cohort; for example, around 21% of those currently in a partnership are cohabiting. Given the rich background variables measured in childhood, this prospective longitudinal dataset is ideal for addressing issues of selection using propensity score analysis. Results show that men and women have significantly higher well-being scores if they are currently in a partnership, even after matching on similar childhood background characteristics. However, differences in well-being scores between cohabiting and married people disappear when matching people with similar childhood background characteristics, especially for more committed unions. We also find differences by specific domains of childhood characteristics: matching people on psychological attributes eliminated significant differences for people in long-lasting relationships and with children, while matching on family background and educational aspirations eliminated differences for fewer types of couples. This demonstrates the importance of early childhood conditions for understanding the relationship between cohabitation, marriage, and mental well-being.