No article, quiz, or study can tell you whether to leave your relationship — and you should be suspicious of anything that claims it can. What research can do is show how people actually arrive at this decision: what predicts staying, what predicts leaving, and why the process feels so tangled even for thoughtful people. Understanding how the decision works doesn’t make it for you, but it can make your own thinking clearer.
This article is educational. It summarizes what psychological research finds about stay-or-leave decisions in romantic relationships. It is not counseling, it cannot judge your relationship, and it offers no verdict on what you should do. Only you know the full texture of your life together — the research below is a lens, not an answer key.
Table of Contents
Why the Decision Is So Hard
Few decisions carry as much weight in as many directions at once. Leaving affects not just feelings but housing, finances, friendships, family, and your sense of who you are. Research on stay/leave decisions highlights a feature that makes them uniquely difficult: they are interdependent. Samantha Joel and colleagues (2018) found that people weigh not only their own satisfaction but also their partner’s perspective — people who believed their partner was highly committed were more likely to stay themselves, even after accounting for their own satisfaction. Staying “for the partner” is a real, measurable motive, not a weakness.
Layered on top of that are quieter pressures. There is the pull of everything already invested — the years, the shared home, the life built together — which can make staying feel obligatory regardless of how the relationship feels now. There is fear of regret, which cuts both ways: people imagine regretting a mistaken departure and regretting another year spent in the wrong place. And there is genuine uncertainty: no trial run, no undo button, which is part of why the decision so often stalls.
What Predicts Staying vs Leaving
When researchers follow couples over time, a few factors consistently shape who stays and who goes. Satisfaction matters — people who feel their needs are met are more likely to remain — but it is far from the whole story. Some people stay in relationships they describe as unhappy, and some leave relationships that look fine from the outside. The gaps between those outcomes are where the interesting psychology lives.
The Investment Model Lens
The most influential account of why people stay or leave is Caryl Rusbult’s (1980) investment model. Rusbult proposed that commitment to a relationship — the psychological state that keeps people in it — rests on three distinct factors.
The first is satisfaction: how positively you evaluate the relationship and how well it meets your needs. The second is quality of alternatives: how appealing your life looks outside this relationship, whether that means another partner, being single, or your wider circumstances. The third is investment size: the resources attached to the relationship that would be lost or diminished if it ended — shared time, a home, a social network, money, children, and even the identity you’ve built as part of a couple.
The model’s most important insight is that these factors can pull in different directions. Someone can be dissatisfied yet remain deeply committed because investments are large and alternatives look poor. In other words, commitment is not the same as happiness — and staying is not proof that a relationship is good, just as leaving is not proof it was worthless. Reading your own situation through these three lenses can reveal why you feel pulled the way you do.
Ambivalence Is Normal
If you feel certain on Monday and doubtful on Friday, the research says you are not broken, dishonest, or secretly decided. Joel and colleagues (2021) studied stay/leave ambivalence directly and found that feeling torn is a common feature of major relationship decisions — and that higher ambivalence predicts exactly that pattern of day-to-day fluctuation in intentions. The wavering is proportional to what’s at stake.
This matters because ambivalence is often misread. Some people treat doubt as a verdict: “If I were meant to stay, I wouldn’t question it.” Others treat a good week as proof the doubts were imaginary. Neither reading is supported. Ambivalence appears to be what a mind looks like when it is honestly processing a decision with real costs on both sides. It may resolve with time and information, or it may persist — but its presence alone doesn’t tell you which way it will resolve.
Deciding Well (Not Deciding “Right”)
Because no study can certify a “right” answer, a more useful goal is deciding well: making the decision in a way you can stand behind regardless of outcome. Research on how these decisions unfold suggests a few features of deciding well.
Deciding well is informed — it involves honestly appraising the relationship as it is, including patterns you’ve noticed over time. (Our article on relationship red flags covers the patterns researchers and clinicians flag as worth noticing.) It is honest about motives: the investment model can help you see whether you’re staying from commitment and care, or mainly from sunk costs and fear of poor alternatives — both are human, but they’re different foundations. And where circumstances allow, deciding well is unhurried: intentions fluctuate when ambivalence is high (Joel et al., 2021), so a decision made at the peak of a bad week may not reflect your considered view.
A Note on Safety
Everything above assumes a relationship where deliberation is safe. If a relationship involves fear, coercion, or violence — if you monitor your words to avoid frightening reactions, or feel controlled or unsafe — the frameworks in this article don’t apply in the same way. In that situation, the priority is safety, not more analysis, and the appropriate next step is professional or support resources equipped to help with exactly that. This distinction matters more than any model in this article.
Key Takeaways
- Research cannot make this decision for you, but it does reveal how people decide — and understanding the process can clarify your own thinking.
- Stay/leave decisions are interdependent: concern for a partner, and beliefs about their commitment, measurably influence whether people stay (Joel, Impett, Spielmann, & MacDonald, 2018).
- Rusbult’s (1980) investment model holds that commitment depends on satisfaction, quality of alternatives, and investment size — which is why commitment and happiness are not the same thing.
- Ambivalence is a normal feature of big relationship decisions, and it predicts day-to-day fluctuation in intentions rather than revealing a hidden answer (Joel et al., 2021).
- The realistic goal is deciding well — informed, honest about motives, and unhurried where possible — not finding a provably “right” answer.
References
- Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16(2), 172–186. https://doi.org/10.1016/0022-1031(80)90007-4
- Joel, S., Impett, E. A., Spielmann, S. S., & MacDonald, G. (2018). How interdependent are stay/leave decisions? On staying in the relationship for the partner. Journal of Personality and Social Psychology, 115(5), 805–824. https://doi.org/10.1037/pspi0000139
- Joel, S., Stanton, S. C. E., Page-Gould, E., & MacDonald, G. (2021). One foot out the door: Stay/leave ambivalence predicts day-to-day fluctuations in breakup intentions. European Journal of Social Psychology, 51(2), 294–312. https://doi.org/10.1002/ejsp.2739
Keep Exploring
- The Psychology of Breakups — if the decision is made: why endings hurt and what recovery looks like.
- Boundaries in Relationships — sometimes the issue is a limit that was never set, not the relationship itself.
- Relationship Red Flags — patterns worth noticing when you’re weighing what you’re seeing.
How do I know if I should leave my relationship?
No test or article can answer that for you, and research doesn’t offer a threshold score. What studies do show is what shapes the decision: your satisfaction, how your alternatives look, the size of your investments, and your concern for your partner (Rusbult, 1980; Joel, Impett, Spielmann, & MacDonald, 2018). Examining those honestly can clarify why you lean the way you do — the verdict itself remains yours.
Is it normal to feel unsure about leaving?
Yes — very. Research on stay/leave ambivalence finds that feeling torn is common during major relationship decisions, and that ambivalence predicts intentions that fluctuate from day to day (Joel et al., 2021). Uncertainty reflects the size of what’s at stake; it is not, by itself, proof that you should stay or go.
Why do people stay in unhappy relationships?
The investment model gives the clearest answer: commitment is driven not only by satisfaction but also by investments that would be lost on leaving and by how poor the alternatives look (Rusbult, 1980). Large shared investments — years, a home, a network, an identity — can sustain commitment even when happiness is low. Concern for a partner’s feelings adds further weight (Joel, Impett, Spielmann, & MacDonald, 2018).
Should I stay for my partner’s sake?
Research confirms that people do exactly this: believing a partner is highly committed makes people more likely to stay, beyond their own satisfaction (Joel, Impett, Spielmann, & MacDonald, 2018). Whether that motive should carry the day in your case is a values question no study can settle — but it’s a real and common reason, not a failure of nerve.
What if I’m afraid to be alone?
That fear is widespread and understandable, and the investment model helps explain it: being alone is one of the “alternatives” people implicitly weigh, and if it looks bleak, staying looks better by comparison (Rusbult, 1980). Noticing that the fear is part of the calculation — rather than a verdict on the relationship itself — can help you separate “how do I feel about this relationship?” from “how do I feel about being single?”