Sliding vs. Deciding

How Cohabitation Changes Marriage

My colleague Scott Stanley and I put out a report in the summer of 2014 that was called “Before I Do,” sponsored by the National Marriage Project at the University of Virginia; that report is the foundation for this article.[1] A generation or two ago, people formed relationships and made commitments differently than they do now. We were interested in looking at the ways dating and commitment sequences have changed over the years, and how those sequences might be related to later marital quality.

One of the perspectives Scott Stanley and I have been working on is what we call “sliding versus deciding.” This concept refers, in part, to the number of choices young people have today. This variety of choices might be one of the biggest differences between dating today and dating a generation or two ago; now people have many more options, not just in the partners that they choose, but also in the paths that might or might not lead them to marriage. Our general premise is that we can expect better outcomes if people make conscious decisions rather than sliding into new circumstances.

“Sliding versus deciding” summarizes the distinction between “dedication” and “constraint” commitment. “Commitment” usually implies the idea of a relationship having a long-term future, and that is what we call “dedication.” It is a sense that the couple is working together as a team; there is the expectation of a future together and of planning for the future. The flip side of dedication commitment is “constraint” commitment. Constraint commitment comes from things that build up and make it harder to leave the relationship. Some examples of constraints are buying a home together, having a child together (that one does not happen to be predictive of whether or not couples stay together if they are unmarried, interestingly), or adopting a pet together—things that might make it harder to end a relationship regardless of how committed or dedicated you feel to that relationship. In simpler terms, constraint commitment is sliding; dedication is deciding. 

A generation or two ago, deciding was the norm. You felt love toward another person, you felt attracted to another person, you decided to be more committed to another person, commitment built, and then you built constraints. It was after you made a commitment that you moved in together, had a child together, changed your career, moved across the country, bought a house together, adopted pets. In that case, it did not really matter that you had taken on those extra constraints because you already felt dedicated to this person. But when you slide through new circumstances or relationship transitions like moving in together or having sex in a relationship, you break up that traditional sequence. For example, you might feel attracted to someone, you might feel like you love someone, and then you start building constraints without really developing a sense of dedication to the relationship. You are still on this track towards staying together, however, because of those constraints. We think what happens when people slide into relationship transitions is that they may start building constraints before they have a chance to think about whether they want to be committed or dedicated to this person and this relationship; and this constraint before commitment could cause problems later on.

To look at some of these questions about experiences and sequences before marriage, Scott and I used a study that we had conducted with our colleague Howard Markman at the University of Denver, a study that was initially funded by the National Institute of Child Health and Human Development. We call it the relationship development study.[2] About 1,300 people started this study. They were recruited nationally, all of them unmarried English-speaking people in the U.S. They completed surveys 30-35 pages long in 11 waves 4 to 6 months apart. We were interested in looking at these unmarried couples and following them over time to see what happened. (I am very grateful that these participants stuck with us long enough to complete this study because it was fairly time consuming and a little intrusive.) We focused only on people in opposite-sex relationships in this study. When the study started, about two-thirds of participants were dating, and the other third had already moved in with their partners. For this report, we chose people who got married during the course of the study so that we could examine the histories of those relationships and those individuals, maybe even before they got into the relationship that turned into marriage, and then we could look at their marital quality after marriage.

Here are some of the results in basically a y=mx+b kind of equation. We used multilevel modeling so that we could aggregate the information that people had given us about themselves over these 11 waves of data to predict marital quality. We measured marital quality with a brief version of a widely used instrument called the Dyadic Adjustments Scale. This instrument asks people to rate their marital happiness, how often they confide in one another, how often they think things are going well, and how often they have thought of breaking up or getting a divorce. In this sample, people were relatively happy, reporting fairly high marital quality, which makes sense, since most of them had just gotten married. In order to continue with the study, they had to get married during the study period; and usually newlyweds have the highest level of marital quality they will ever experience. Marital quality tends to decline after marriage, but why wouldn’t it? You are not going to get married if you are not pretty sure you are happy in the relationship.

We analyzed a number of background characteristics before we started looking at the main questions about different relationship experiences and transitions. Most of our findings did not vary much when control variables were added. Often, the association was a little weaker with controls, but there were very few instances when a finding became non-significant. One of the messages from our results is that not many background characteristics were related to marital quality, at least in these first couple of years into marriage. What we do see are associations between factors that are significant, significance that has been well replicated. For example, we know that people who have higher levels of education tend to report higher marital quality, and they are less likely to get a divorce. We know that people who were living with both of their biological parents when they were 14 years old also report higher marital quality later on. (There is nothing magical about the number 14; it is just the age that was used in other research.) We know that people who come from more stable families tend to have more stability in their own relationships later on. Unfortunately, we also see some important gaps between African Americans and Caucasians in terms of stability and quality of relationships. In this study, we do not see that religiousness was associated with marital quality. That contradicts a number of findings in this field, however, and merits some explanation. I think the reason we do not see an association between religion and marital quality in this study is because we are looking at such an early segment of marital quality; the effects of religiousness might show up later in marriage, especially when couples are having children together. Similarly, we did not find that income was associated with marital quality in this study, and that also contradicts other findings, especially about the way income is related to the risk for divorce. Here again, I think that is likely because we were looking at marital quality right after marriage. 

In terms of the main findings, we focused on two broad categories of experiences people might have before getting married. One category is experiences from prior relationships. The other category is experiences with the person whom they eventually marry, the couple’s history before marriage. Based partly on the findings discussed earlier regarding background characteristics but also on the general literature, we controlled for a number of variables related to marital quality. We also looked at some moderators, factors that might make the findings different for different people. We wanted to know if any of those associations might be different for men and women or if they might be different for people with and without a college level of education. 

Regarding individuals’ prior relationship experiences, here is what we have found. Most of the people in the study were entering a first marriage. People who were entering a first marriage reported higher marital quality than people who were entering a second or third or even fourth (there was one fourth marriage in this sample). Couples with children from prior relationships had lower marital quality later on. We also looked at their experiences with past sexual partners. People who had sex only with their future spouse and no one else reported higher marital quality later on. About 23% of this sample had sex only with the person they married. (Unfortunately, we did not ask this question quite correctly to discern if they had sex before they got married or not. There has not been much research on that in a number of years; the timing of sex in relationships and what that might mean for marital quality later on is still a very important question.) Almost 40% of the sample had lived with a prior partner before they got together with the person they married; having had other cohabiting partners was associated with lower marital quality later on.

There were a couple of gender differences regarding prior relationship experiences as well as regarding children from prior relationships. Having children from a prior marriage made a bigger difference for women in terms of their later marital quality than it did for men. This can probably be explained in part by the fact that women are more likely to have custody of those children, so the woman’s children are more likely to be involved in the new marriage. Women’s own marital quality may be more affected by their own children than by their partner’s children.

We also saw a gender difference when we looked at the number of prior sex partners these individuals had before they got together with the person they married. While it was true that men who had sex only with their future spouse experienced higher marital quality, among those who had had multiple partners, the number did not make as much difference as it did for women. (This troubles me somewhat; I do not love the messages these findings are sending to men versus women about number of sexual partners and what that might mean for them. It seems that there is still a double standard.) About one-fourth of the sample, as mentioned earlier, had sex only with their future spouse; another one-fourth of the sample had had sex with two to four partners before marriage; and about one-third had had sex with five to ten partners before marriage. The findings show that the group of women who had had sex with more than ten partners before marriage reported the lowest marital quality later on. The average in the sample, about eight sexual partners before marriage, was similar for men and women. However, the median in this sample was five—the mean is skewed by people who reported many, many sexual partners before marriage. What we see here is that, at least for women, the more partners they have had, the lower their marital quality later on.

One of the most interesting things about these findings, which in many ways were new to this field, is that prior relationship experience really matters, but in a somewhat counterintuitive way. If you are hiring an architect, you want to hire someone who has a lot of experience. If you are going to see a doctor, you probably want a doctor who has a lot of experience in your illness. But in terms of relationships, we are seeing that the opposite is true. People with more experience might end up having more trouble later on in their marriages. There are a couple of potential explanations for this. One is that, if you have a lot of experience, you also have a greater sense of what the alternatives are, and you have more comparisons to make to other people. It also is true that the more experience you have in relationships, the more experience you have breaking up. That experience with breaking up might make it seem easier to break up later in a marriage, or it may make you think about breaking up more and question the quality of the relationship more. We also see that more experience means you are more likely to have children before you get married, and we saw the effect of that earlier.

More than 40% of babies born in the U.S. today are born to unmarried parents. We used to say they were born to single mothers, but they are not “single” mothers, for the most part. The mothers are often partnered with the baby’s father or with someone else when the baby is born. We also know that those families tend to be quite fragile, and we are seeing some of that fragility carried over in these findings about marriage as well: children from prior relationships tend to be difficult on a marriage. Having children is hard, and it is especially hard to start a marriage already having children.

It is interesting to think about the messages young people hear today about relationship experience, messages like, “Don’t settle down too soon.” “Make sure you get everything out of your system.” “Your 20s are a time for great exploration.” “What happens in Vegas stays in Vegas.” “Those things won’t affect your future marital quality or outcomes.” In fact, what we are seeing here seems to be the opposite. Those experiences do impact us in some important ways and may lead to us having a more difficult time later on in marriage. Scott Stanley talks about the “duct-tape hypothesis,” which is the idea that, if you have a piece of duct tape and you keep sticking it to things, it gets less sticky over time. Applied to relationships, the idea is that the more relationships you have, the harder it may be to really commit to another partner going forward. 

After looking at prior relationships, we looked at experiences the couple had together before they got married. For example, we looked at how long they had been together. Remarkably, in this sample, time together was not associated with their marital quality later on. There may be some extremes. Getting married very quickly could be a negative, or, in some cases, a positive. On the other hand, for people who have been together a very long time, maybe there would be some circumstances where that was a negative. There is no magic number for how long you need to be together before getting married, at least according to these data.

We also looked at the age at which a couple married and found that getting married older was associated with lower marital quality, which again goes against some conventional wisdom. 

We asked people if their relationship started with “hooking up,” and we let them self-define what “hooking up” meant. About 32% of the sample did start that way, and we found that those couples had lower marital quality later on.

Couples who had children together already or who were pregnant before they got married also had lower marital quality. This finding may reflect the fact that having kids is hard, but it also may be a result of sliding into marriage. Hooking up, having sex before getting into a relationship, before feeling committed in a relationship, and having children or becoming pregnant before getting married may also reflect that sliding mentality to some degree. 

However, education level was also related to how strongly being pregnant before marriage was associated  with later marital quality. People who had a college degree were much less likely to be in the top marital quality group if they had a child together before they got married. For people who did not have a college degree, we hardly saw any difference at all in their marital quality based on whether they had a child before they got married. That difference is important for us to think about in future work and in public policy as well. In part, what this reflects are the norms. People who do not graduate from high school are much more likely to have a child outside of marriage than people who graduate from college. It is really quite outside the norm for people who do graduate from college to have a baby or become pregnant, or to let anyone know about that before getting married. Those social norms might affect later marital quality. 

We also tried to look at some potential red flags in a relationship before getting married. We asked people in the study if they had ever had a sexual relationship with someone other than their partner while they were dating that person; in other words, had this person ever cheated on their partner before getting married. In our sample, 16% said yes, but this was only very weakly related to marital quality later on. Again, this is not a great message for us to be telling young people. Many dating relationships where there is infidelity end long before they would turn into marriage. About 10% of our sample knew that their partner had had a sexual relationship with someone else while the two of them were dating, and that was related to lower marital quality later on.

We also asked many questions about physical aggression in these relationships, and this data is staggering. Of people who got married in this sample, 53% reported that there had at some point been physical aggression in their relationship. It could be as minor as throwing something at your partner that could hurt, all the way up to very severe aggression, coercion, or sexual coercion. In marriage and couple research, we tend to ignore aggression, but this number says that we cannot. This is something we really need to be talking about and helping to educate people about: how to make sure conflict does not get to that level and how to get people out of very unsafe relationships. I imagine only a few of these in our study really represented very unsafe relationships, but there are some unsafe relationships within this sample. Probably not surprisingly, people who reported some physical aggression in their relationships also reported lower marital quality later on.

Aggression is also involved in some other findings that are coming out of our research lab at the University of Denver. A graduate student there has been looking at how infidelity and aggression in one relationship are associated with infidelity and aggression in the next relationship. She finds that you are much more likely to experience infidelity or aggression if you have had that in your just-prior relationship. This fits with the idea that the more experience you have, the more likely you are to repeat some of those negative patterns. Part of what is interesting here is that it goes both ways. That is, if you cheated in your last relationship, you are more likely to cheat in your next relationship; but also, unfortunately, if you were cheated on in your last relationship, you are also more likely to be cheated on in your next relationship.

We asked a number of questions about commitment and cohabitation. We asked people about the timing of moving in together and whether they had committed to marriage before they moved in together. About 70% of this sample had lived together before they got married. We found that people who had lived together before they were committed to marriage together, before they had a mutual and clear plan to get married, reported lower marital quality later on. About 70−75% of people now live together before they get married. The most common answer to the question “Why did you move in together?” is “It just kind of happened.” And that it was more convenient. What we are seeing here is that, if a couple has already made the commitment to marry, if they are already dedicated to one another, moving in together is not associated with lower marital quality. However, moving in is associated with lower marital quality among those couples that slide into living together but then maybe build a number of those constraints. For some of them, it may be the constraints that lead them to get married, when they otherwise maybe would not have married this partner after all. That is also an important finding to consider, because a lot of young people today really like the idea of collecting lots of data before making the big decision of getting married. What better way to do that than by moving in with this person? You want to know if this person is going to leave the toilet seat up, or you want to know how she handles money. You want to find out all these things about this person. This desire may come from a really good place, but the problem is that moving in together may put couples on a track toward getting married that is hard to get off. It is harder to end a relationship once you have moved in together, even if the relationship fails that test, essentially.

We asked people who did live with their partners before they got married how it happened. Was it a slide, or did you make the decision together? We found that people who said they made a decision together reported higher marital quality, which supports the idea that deciding is generally associated with better outcomes. We also asked people at every wave to rate their own commitment on a 1−7 scale, and then immediately afterward to rate their partner’s commitment on the same 1−7 scale. Interestingly, across those waves, before they got married, if they ever thought they were more committed than their partner was, they had lower marital quality later on. The results reflect the idea that, if there are major differences between a couple in how they like to make decisions and what commitment means to them, those differences may continue to cause problems. 

We asked about the wedding and whether people got any kind of premarital education. An amazing 43% of people said that they got some kind of premarital education together. That education could range from something like meeting with the pastor at the church one time to talk about wedding plans to something much more intensive, like a 30-hour workshop on relationships. There is good evidence that premarital education—especially the kind that teaches couples communication skills and gets them talking about differences and expectations—is associated with a lower risk for divorce later on. We asked if people had a wedding, and about 90% said yes. Those who were in the minority on that question reported lower marital quality later on. We also asked how many people attended the wedding. The mean number of people was 116 in this sample, and the more people you had at your wedding, the higher your marital quality later on. 

Further research that has come out recently has looked at this question of wedding attendance more carefully, and they also find that the more people who attend your wedding, the higher your marital quality and the lower your risk for divorce—and it is not related to how much money you spend on the wedding.[3] There are some really good theoretical reasons to think that the number of people who attend might actually be important. One reason is that the more public your commitment is, the more likely you are to follow through on that. If you stand up and tell 100 or 200 people that you are promising to spend the rest of your life with this person in sickness and in health, it is going to be a little harder for you to break that promise because it has been such a public commitment. The other reason is that the large group likely reflects a greater social system and social network that supports your marriage. It may be that people who have more wedding attendees simply also have greater social networks. There may also be a causal relationship there as well, because it means that this entire audience of people has also committed to you to help protect your marriage and support you in this marriage. We were a little nervous about the implications of these findings at first, and I think this is a great area to do more research to really understand what these dynamics mean before we start sending the message, “Go have a big wedding and make your parents spend lots of money on it.” This is, however, an interesting finding to think about in terms of sliding versus deciding, of commitment and what it means. 

So what does this all mean for educating people about relationships? One thing is clear: there are some experiences and background characteristics people have that they cannot change. We cannot go back and change whether we were living with both our biological parents at age 14, but there may be some ways we can change the dynamics that those kinds of background characteristics initiate. We also see that there are many things young people have some control over that may be related to later marital quality and outcomes. We really need to start thinking about ways that we can impact people and help them make good relationship decisions earlier. The field of relationship education has focused on couples and premarital couples. It seems important to teach people how to communicate better, but I think we could have a much greater impact if we helped people think about their relationship experiences before they have them—when they are teenagers, when they are young adults, when they are in the middle of making some of these decisions or sliding through things. These are the times to think about whether they should hook up with somebody, whether they should get pregnant, whether they should move in with someone. There is a great amount of education that we could be doing long before a couple is about to walk down the aisle.

Galena Rhoades, Ph.D., is research associate professor and associate clinical professor in the department of psychology at the University of Denver.

[1] Galena K. Rhoades and Scott M. Stanley, “Before ‘I Do’: What Do Premarital Experiences Have to Do With Marital Quality Among Today’s Young Adults?” The National Marriage Project, the University of Virginia, 2014, available at

[2] Scott Stanley et al., “The Relationship Development Study,” university of Denver, available at

[3] Andrew M. Francis and Hugo M. Mialon, “‘A Diamond is Forever’ and Other Fairy Tales: The Relationship between Wedding Expenses and Marriage Duration,” Economic Inquiry 53.4 (October 2015): 1919-1930.