Skip to main content

Verified by Psychology Today


What Are the Odds That Your Relationship Will Improve?

New research on rank-order relationship stability.

Key points

  • Wouldn't it be great if people could predict the future of their relationship's stability?
  • Researchers answer this key question by examining stability as a function of age and relationship length.
  • The dynamics of most relationships tended not to change over time, but there are ways to make yours work in your favor.

The big question that couples have when their relationship starts out is whether it will last. In the early stages, they may scrutinize every interaction to try to project into the future the fate of their partnership. As the relationship continues, couples may still use everyday interactions as a guide to their crystal ball-gazing into the years ahead.

In your own relationship, you may wish that you could find reliable clues to help see where things are going. You might also, if things aren’t going well, wish that you could set off on a corrective course to fend off what would be a painful and difficult ending, should that become the only recourse to improving your personal happiness.

Over the course of multiple studies on relationship stability, researchers have attempted to identify predictors that hold up to scientific scrutiny. Much of this research is hampered by the obvious fact that it’s impossible to study relationships that no longer exist. By definition, only the couples who remain together can be poked and prodded by researchers trying to understand changing dynamics over time. A second obvious problem in relationship stability research is that age and relationship length are inevitably compounded with each other. While people in later adulthood can and do initiate new relationships, those who’ve been with their partners for 30 or 40 years are very likely to be somewhere in their later 40s and beyond.

What Defines a Stable Relationship?

According to a newly published meta-analysis by the University of Bern’s (and fellow PT contributor) Janina Bühler and colleagues (2022), although large-scale studies have examined average levels of relationship satisfaction over time, these aren’t sufficient to answer the question of what constitutes “stability.” If you think about your own relationship satisfaction, you will more than likely agree that it’s risen and fallen across the duration of your time together with your partner. However, this is only half of the equation. Just as important, according to the German authors, is the question of whether your own satisfaction has risen and fallen in a way similar to other people who’ve been with their partners for the same length of time.

The quality known as rank-order stability captures this idea that people can change in similar ways over time, meaning that everyone’s satisfaction goes up and everyone’s goes down to the same extent across a relationship’s duration, much as all the ships in the ocean rise and fall with the tides. This means that if yours goes up, it could be because something has happened to stimulate a positive change in your relationship, or it could be because it reflects the natural evolution of couple dynamics. If it goes down, similarly, it may be a function of time, not specific forces eroding your particular relationship.

As you think about this rise and fall in rank-order stability, it’s important to remember once again those provisos about research on relationship duration. However, the Bern authors recognized before proceeding that they would have to account for relationship length separately from age of the partners. They, therefore, asked the question, at the outset of their study, “Does rank-order stability of relationship satisfaction vary across adulthood, as a function of age and as a function of relationship duration?” (p. 1138). They also sought to tease out the impact of various external events, including “random contextual factors” (p. 1140).

Individual differences can also play a role in affecting rank-order stability. Some people are just more optimistic, better adjusted, happier, and more likely to attend to their relationship quality. They could also be more securely attached, meaning that they were raised in an early environment that promoted healthy relationship attitudes.

A stable relationship also feeds on itself over time. The interaction between partners helps to shape their individual feelings of satisfaction as each partner boosts (or detracts from) the other’s well-being. Again, however, it’s not just the average that counts, but the rank-order stability. Do most people in relationships find ways to make the dynamics work in their favor, or do some succeed more than others to rise above the tide?

Putting Rank-Order Stability to the Test

Using the method of meta-analysis, in which data from multiple sources (published and unpublished) are consolidated in a statistical model, Bühler and her fellow investigators applied rigorous controls to answer their research questions. They ruled out studies that didn’t meet the criteria of scientific rigor (such as being empirical and quantitative), were not conducted on actual relationships, and most importantly, did not follow couples over time.

From an initial pool of 1,209 studies, they narrowed their analyses to 86 investigations on 148 independent large samples, providing a total number of 153,396 participants (19 to 71 years old). The entire dataset was made available through the Open Science Framework, where anyone can go back and inspect and validate their findings. The studies included in the meta-analysis had to evaluate rank-order stability across no less than two months but lasting as long as 20 years.

Borrowing a concept from similar studies of life satisfaction, Bühler then explored the idea that rank-order relationship stability reflects, perhaps not surprisingly, a “trait-like construct” (p. 1155). Happier people remain happy over time, and the unhappier ones stay unhappy. What’s particularly interesting about this in the context of a couple is that it takes two to keep that rank-order stability where it is. It’s possible that those happy people stay happy in their relationship because they’re so good at resolving conflicts and, furthermore, may have chosen partners with similar personality strengths.

There was one interesting feature of the data that emerged when the authors controlled for the length of time in a given study. The longer the lag between data points, as it turned out, the lower the rank-order stability but the estimate of stability still remained relatively high even in the longest relationships.

What about that age-relationship length problem? The authors attempted to unpack this conundrum, but even their large sample size could not permit this: “the time metrics are… inherently linked to each other.” However, by identifying this problem, the Bern authors provide a useful warning for future researchers because “otherwise, developmental processes in romantic relationships might be misattributed to one of the time metrics” (p. 1157).

What the authors unfortunately could do nothing about is the other inherent problem in relationship duration research. No one can study a relationship between people who are no longer together. As the authors note, “dissolving relationships often show higher levels of stress and dissimilarity” (p. 1157) and therefore could unravel in unpredictable ways over time.

Can You Break the Rank-Order Stability Pattern?

These findings may not be very comforting to you if you’re in a position with your partner that is not very good. You could hope that time might iron out your difficulties, but the Bühler et al. findings imply that sitting back and waiting might not be enough. However, if you want to overcome those potentially overwhelming statistical odds, there may be a silver lining in the German results.

Consider those couples whose personalities and sunny optimism set them out on a positive course over time. Not only did they seem to have a brighter approach to life, but they also had found ways to resolve conflicts constructively. Time might have tested them somewhat, as well as outside factors that impinged on their adaptation, but they could still find ways to cope as individuals and as a couple.

Your route to overcoming what might be a less-than-ideal position in the rank-order hierarchy might also involve talking to your partner about these results. Use the projections over time implied in the findings to figure out how to move up a few notches in the satisfaction domain beyond the point you're at now.

To sum up, change in adulthood, both as an individual and as a couple, isn’t impossible, but the Bern findings suggest that trying to change the course of these trajectories means that you can’t just hope for the best. Fulfillment in life is a process, but it’s one that, informed by these findings, you may be able to both alter and control.

Facebook image: Jacob Lund/Shutterstock


Bühler, J. L., & Orth, U. (2022). Rank-order stability of relationship satisfaction: A meta-analysis of longitudinal studies. Journal of Personality and Social Psychology, 123(5), 1138–1165.