Stay or leave your relationship: how to decide

Why this question is so hard

Because it pits two futures against each other, and you don’t know either one. Staying means betting on a change you haven’t seen yet. Leaving means betting on a life you haven’t lived yet. The human brain is badly equipped for this bet: it overestimates what it could lose compared with what it could gain, it counts the years already invested as if they obliged you to keep going, and it changes its mind with fatigue. On a Monday night after an argument, the answer is clear. On a Sunday morning after a good moment, it is just as clear, in the opposite direction. Ambivalence is not a failure of clear thinking. It is the normal state of someone who still cares about two things.

There is also a reason no one mentions: the question “stay or leave” is often asked too early, as a shortcut. It avoids a more uncomfortable question: “What exactly is wrong, and have we really tried to change it?” According to an often-cited estimate from Notarius and Buongiorno, couples wait an average of six years after serious problems appear before asking for help. Six years of wondering whether to leave, without really trying to stay.

What research says about people who stay

One American study is worth knowing, along with its limits. In 2002, Linda Waite and her colleagues followed adults who said they were unhappy in their marriage over five years. Among those who stayed together, about two thirds said they were happy five years later. And on average, those who divorced were no happier than those who stayed. The couples who bounced back didn’t do it by magic: the original problem (money, work, kids, illness) resolved with time, or they worked on it, or they changed the way they talked about it.

The limits, because they matter: this is American data from the 1990s, it shows correlations rather than proof of cause and effect, and it does not apply to relationships where there is violence or control. What the study really says is more modest and more useful: “I’m unhappy in my relationship today” does not predict “I’ll be unhappy in my relationship in five years.” A couple’s unhappiness is often a season, not a permanent state.

Another, lesser-known finding comes from Paul Amato. Studying divorces, he found that a large share came from low-conflict couples: no violent fights, no betrayal, but boredom, distance, the feeling of living side by side. These are often the couples who type “stay or leave” into a search bar, and they are also the ones with the most room to maneuver, because nothing is broken. It’s switched off, which is not the same thing.

Finally, Scott Stanley and Howard Markman distinguish two kinds of commitment. Dedication: I stay because I want this relationship and a future with this person. Constraint: I stay because of the kids, the apartment, the money, what other people think. Every relationship mixes the two, but when only constraint is left, you’re no longer saving a relationship. You’re managing a situation. So the honest question is not “can I leave?” but “is there still dedication, or only constraints?”

The questions that don’t help

“Do I still love them?” Romantic feeling fluctuates with fatigue, safety, and the latest argument. After months of tension, many people feel nothing, then feel it again once safety returns. It’s a poor thermometer for a decision this big.
“Could I find someone better?” No one knows, and the question compares a real person, with visible flaws, to an imaginary one with none.
“What will people say?” People will say something either way, and they won’t be living your life.

The questions that actually help

Am I safe? If you’re afraid of your partner, if there have been blows, threats, or control over your money, your outings, or your relationships, none of the other questions in this article apply. This is not a relationship to save. It is a situation to get out of, with help. The page on toxic relationships lists resources. In France, call 3919 (free, anonymous) or 17 in immediate danger. Elsewhere, contact your local domestic violence helpline.
What exactly makes me want to leave? A behavior (“my partner doesn’t listen to me anymore”), a pattern (“we argue all the time”), or the person themselves (“I no longer respect them”)? The first two can be worked on, often successfully. The third is of a different nature. Be honest about which one it is.
Have we really tried, or have we just waited? Waiting for it to pass, making an effort for a week, issuing an ultimatum: none of that is trying. Trying means naming the problem, changing something specific, setting up a framework for talking about it, and often asking a third party for help. If the answer is “no, we’ve never really tried,” then the decision to leave would be made on incomplete information.
What would I lose that I couldn’t get back, and what would I keep that I no longer want? Write both lists. Not to count points, but to see what really weighs the most. Many people discover that what they want to leave is a way of functioning, not a person.
If nothing changes in six months, what will I do? This question turns permanent anxiety into a dated decision. It gives the trial a length, and an honest exit if the trial fails.

Before deciding, measure. The MyLoveLab test shows, in 15 minutes, where your relationship is solid and where it leaks, across 7 dimensions. It is not a verdict: it helps you see whether everything is switched off, or just one thing.

See where my relationship stands

“Relationship test: stay or leave”: what a test can and can’t do

No test can decide for you, and those that promise a verdict in ten questions are making fun of you. What a serious test can do is answer the previous question: is everything broken, or just one thing? A couple scoring 30 out of 100 across all seven dimensions is not in the same situation as a couple scoring 80 on five dimensions and 20 on intimacy or conflict. In the first case, the question of staying is a real one. In the second, what you have is an identified problem, with a name, and known ways to work on it. That is exactly what our test measures, and nothing more: it will never say “stay” or “leave.”

A way not to decide alone in your head: the fixed-length trial

Permanent ambivalence wears you down more than any decision, whichever one it is. What helps is turning “I don’t know” into “we’ll try this, until this date, and then we’ll decide.” It’s not an ultimatum, because you both choose it and the commitment is about behaviors, not a threat. And it can go either way: in three months you may find that it’s coming back, or that it won’t.

Exercise to try tonight

Six months, three changes

To propose calmly to your partner, on a day without an argument, as a proposal and not a sentence.

Each of you names one specific behavior you’d like to see change in the other, and one you commit to changing in yourself. Specific and observable: “let’s have dinner without phones three nights a week,” not “be more present.”
Set a review date six months out, and write it down somewhere. Add a twenty-minute check-in every week to see where things stand, without reopening the whole past.
Decide together on outside help, a couples therapist or a marriage counselor, for at least a few sessions. It’s the step couples skip the most, and the one that changes the most.
Rule during the trial: no threats to leave, in any argument. If the question comes back, it gets noted for the review, not thrown into the conversation.
On the review date, each of you answers one question honestly: “Has there been movement?” Not “is everything fixed?” Movement, or none.

This trial assumes two people who are safe and on equal footing. It is never a tool in a relationship that involves fear, control, or violence.

What if your partner doesn’t want to try?

It’s common, and it’s information. One person alone can change a lot in a relationship: stop the criticism, respond to small requests, set a calm tone. Often, the other person starts to move once the atmosphere changes. But one person can’t carry a relationship indefinitely, and a partner’s repeated refusal to try anything, even after weeks of visible effort on your side, is one of the things to put on the scale. Your method didn’t fail. It’s an answer you received.

If you decide to leave

It’s not a failure, and it’s not the purpose of this article to hold you back. Just two things. The first is about children, because it’s the most common reason to stay: the work of Paul Amato and Alan Booth shows that what harms children most is the conflict they witness, not the separation itself. On average, children of high-conflict couples do better after a separation, and children of low-conflict couples often do worse. In other words, staying “for the kids” only makes sense if you can stay without making them live in tension. The second: leaving well is work too, and family mediation exists for that. A couples therapist can also guide you through a separation. The two aren’t contradictory.

And whatever you decide, you’ll make the decision with today’s information, as the person you are today. That is all anyone can ask of a decision. No one makes this one with certainty, and those who tell you they did are reconstructing it after the fact.

Frequently asked questions

Can you save a relationship when you no longer have feelings?

Often, yes, when “no more feelings” really means “numb”: after months of tension, the body turns down the volume, and for many people the feelings come back once safety returns. That’s different from “I no longer respect them” or “I no longer want this person,” which are less often reversible. The useful question isn’t what you feel tonight, but whether respect is still there.

How long should you give it before deciding?

No study sets a length, but a length fixed in advance is better than none: it turns anxiety into a trial. Six months with concrete changes and outside help is enough to see whether there’s movement. Without concrete change, six months or six years won’t make a difference.

Should you stay for the kids?

Paul Amato’s research shows that what matters for children is the level of conflict they’re exposed to, more than whether their parents stay together or separate. Staying in constant tension hurts them more than a calm separation. Staying while managing to lower the conflict, or separating while keeping it low, are both good answers. Violence, however, is never a situation to stay in.

References

Waite, L. J., Browning, D., Doherty, W. J., Gallagher, M., Luo, Y., & Stanley, S. M. (2002). Does Divorce Make People Happy? Findings from a Study of Unhappy Marriages. Institute for American Values.
Amato, P. R., & Hohmann-Marriott, B. (2007). A comparison of high- and low-distress marriages that end in divorce. Journal of Marriage and Family, 69(3), 621-638.
Amato, P. R., & Booth, A. (1997). A Generation at Risk: Growing Up in an Era of Family Upheaval. Harvard University Press.
Stanley, S. M., & Markman, H. J. (1992). Assessing commitment in personal relationships. Journal of Marriage and the Family, 54(3), 595-608.
Notarius, C., & Buongiorno, J. (1992). Wait time until professional treatment in marital therapy. Unpublished manuscript, Catholic University of America, cited by Gottman (1999).
Gottman, J. M., & Silver, N. (1999). The Seven Principles for Making Marriage Work. Crown.