Print Post
  • Why are some relationship choices linked to poorer marital outcomes? Tweet This
  • Strong selection-based explanations of behavior imply that people have little power over their choices. Tweet This

Our new National Marriage Project report, “Before ‘I Do,’” suggests that the adage “What happens in Vegas stays in Vegas” is false when applied to relationships before marriage: namely, what happens before “I Do” does matter for later success in marriage. We find, for instance, that the following experiences are associated with lower marital quality:

  • Having a child before marriagei
  • Starting a relationship by “hooking up”
  • Having had a number of sexual partners before marriageii
  • Cohabiting with multiple partners before marriage (serial cohabitation)iii

Social scientists do not argue so much about the fact of such associations as the meaning of them—do they reflect causation or can they be dismissed with another adage, “correlation does not equal causation”? In more sophisticated critiques, the question is the degree to which such findings reflect what social scientists call “selection effects.” That is, some suggest that the apparent link between such behaviors and lower-quality marriages is simply an artifact of certain personality characteristics (e.g., a propensity to risky behavior), difficult family backgrounds (e.g., parental divorce), or socio-demographic disadvantages that make some people more likely to report these behaviors and more likely to have lower-quality marriages.

For instance, economist and Bloomberg View columnist Noah Smith dismissed our new report by tweeting:

@WilcoxNMP @USATODAY @SharonJayson Study is biased for many reasons. Selection bias. Different uses of the word "hookup". Reverse causality.

–Noah Smith (@Noahpinion) August 19, 2014

We acknowledge that it is challenging to establish the degree to which findings like these in the social sciences reflect causality. Further, we do believe that selection plays an important role in understanding romantic, marital, and family outcomes and that what goes into selection—including income, education, poverty, and family environment—clearly impacts how, for some, personal choices get restricted by environments. Nevertheless, we want to suggest reasons for caution when making conclusions about personal choice and the role of selection.

Consider the link between premarital births and marital outcomes. Selection plays an important role in who is most likely to have a child before marrying (either with another partner or with the person he or she ends up marrying). All can likely agree about the challenges this presents. For example, having a child from a prior relationship can impact the options an individual has in searching for the best possible partner. Or, for a couple with a child before marriage, the tasks required in taking care of the child may interfere with the establishment of positive relationship routines. Regardless of these obvious points, does evidence for the role of selection mean people have no choice but to live out their predispositions? We do not think so.

Understanding Selection

The history of research on cohabitation provides a clear example of the importance of understanding selection.iv Decades of findings showed that cohabiting with one’s eventual spouse before marriage is associated with poorer marital outcomes.v Findings in this area have recently become more nuanced (for more information, see this recent post), but from the discovery of this link through the present, researchers have realized that selection plays an important role in understanding how cohabitation is associated with outcomes in marriage.vi

Selection variables have typically been operationalized as relatively static, socio-demographic variables, such as religiosity, attitudes about divorce, education level, income, and family background. Historically, the people who were most likely to live together before marriage also had characteristics that made them more likely to struggle in marriage than those who hadn’t cohabited. That means at least part of the negative link between cohabitation and marital outcomes was driven by background characteristics rather than by the effects of cohabitation itself. Some people seem to argue that this link and other links between certain behaviors and poor relationship outcomes are entirely explained by selection.

It is impossible to prove with certainty that the associations between early relationship choices and marital outcomes involve causality. Doing so would require experimental designs that randomly assign people to various relationship experiences. It would be a further reach beyond existing methods to demonstrate that people have free will and can make choices. Nevertheless, as we argue in the “Before ‘I Do’” report, there are good reasons to believe that causality does come into play.

What Choice Do You Have?

Strong selection-based explanations for how predictors relate to outcomes come with a cost: they imply that the choices of individuals have little bearing on their outcomes. This type of determinism disavows the ability of individuals to affect their own destinies. If we overemphasize selection, we risk undermining personal agency and implying that people are dependent entirely on outside help and influence. This message is disempowering.

We propose that, although selection matters in relationship outcomes, so too do our choices and beliefs. Here’s an example of how those factors may play out in the life of one individual.

Suppose a person, “Jennifer,” is at high risk for poor relationship outcomes due to certain demographic characteristics. In addition, she has already cohabited with two partners, and she has a son from one of these prior relationships.

Now suppose that Jennifer has recently come to realize that living with someone increases the difficulty of breaking up. She subsequently meets a man she finds attractive but in contrast to her past pattern, she decides to go more slowly and she chooses not to have him move in with her and her child—thinking, “at least not yet.” She wants such an important step to be a mutual decision, not an accident.vii  Her relatively poor economic options make this hard to do, but she holds out.

After two months of dating, Jennifer has figured out that this man is not a good fit for her: he’s controlling and he drinks a lot, and she is concerned about how he has treated her son. She decides for sure not to let the man move in and, then, soon after, she decides to break off the relationship.

Handling Selection Cautiously

A strong selection perspective doesn’t really address how having a child with one boyfriend affected Jennifer’s long-term chances of forming good romantic relationships, or how her decision about her latest relationship could alter important aspects of the rest of her life and that of her son. Jennifer still faces a high risk of lower relationship and marital quality later on, but she was able to exercise her own choice at a moment when much was at stake. Giving her the support to continue making choices that protect her and her child’s interests in the future could have similar beneficial effects.

Studying selection characteristics, such as those that affect people like Jennifer, shows us which people may need the most help in overcoming the risks built into their lives. At the same time, we believe that researchers should handle the subject of selection with caution and carefully consider the messages given to individuals about how they live their lives. If a behavior (1) is identifiable, (2) is reliably associated with negative outcomes, and (3) is plausibly one a person has some control over, we believe that it is best not to dismiss the importance of that behavior. Even when such behaviors clearly involve selection, we should help people understand what aspects of their destiny can be impacted by personal choice.

Whatever their prior life experiences and demographic characteristics, understanding the potential impact of relationship experiences on outcomes can empower individuals to make choices that bring them closer to achieving their aspirations.

Scott Stanley, Ph.D., is a Senior Fellow at the Institute for Family Studies and a Research Professor and Co-Director of the Center for Marital and Family Studies at the University of Denver. Galena Rhoades, Ph.D., is a Research Associate Professor in the Psychology Department at the University of Denver and a psychologist in private practice.


i. Waite, L. J., & Lillard, L. A. (1991). Children and marital disruption. American Journal of Sociology, 96(4), 930.; Tach, L., & Halpern-Meekin, S. (2009). How does premarital cohabitation affect trajectories of marital quality? Journal of Marriage and Family, 71(2), 298-317.

ii. Teachman, J. D. (2003). Premarital sex, premarital cohabitation, and the risk of subsequent marital dissolution among women. Journal of Marriage and Family, 65(2), 444-455.; Busby, D. M., Willoughby, B. J., & Carroll, J. S. (2013). Sowing wild oats: Valuable experience or a field full of weeds?. Personal Relationships, 20(4), 706-718./pere.12009.

iii. Lichter, D., & Qian, Z. (2008). Serial cohabitation and the marital life course. Journal of Marriage & Family, 70, 861-878.; Lichter, D.T., Turner, R.N., Sassler, S. (2010). National estimates of the rise in serial cohabitation. Social Science Research, 39, 754-765.

iv. For example, see: Smock, P.J. (2000). Cohabitation in the United States: An appraisal of research themes, findings, and implications. Annual Review of Sociology, 26, 1-20.; see also Brown, S. L., & Booth, A. (1996). Cohabitation versus marriage: A comparison of relationship quality. Journal of Marriage and Family, 58, 668-678.

v. Stanley, S. M., Rhoades, G. K., & Markman, H. J. (2006). Sliding vs. Deciding: Inertia and the premarital cohabitation effect. Family Relations, 55, 499–509.

vi. For example: Lillard, L. A., Brien, M. J., & Waite, L. J. (1995). Premarital cohabitation and subsequent marital dissolution: A matter of self-selection? Demography, 32, 437-457.

vii. For more on this distinction, see: Lindsay, J. M. (2000). An ambiguous commitment: Moving into a cohabiting relationship. Journal of Family Studies, 6(1), 120-134.; Manning, W. D., & Smock, P. J. (2005).  Measuring and modeling cohabitation: New perspectives from qualitative data.  Journal of Marriage and Family, 67, 989-1002.; Stanley, S. M., Rhoades, G. K., & Markman, H. J. (2006). Sliding vs. Deciding: Inertia and the premarital cohabitation effect. Family Relations, 55, 499-509.