Opposites rarely attract, and matching personalities matters far less than most compatibility charts claim. That is the honest answer from decades of relationship research. What actually predicts whether two people stay happy together is not how well their personality types “fit,” but the absolute level of three specific traits each person brings: low neuroticism (emotional steadiness), high agreeableness (warmth and cooperation), and high conscientiousness (reliability). A meta-analysis by Malouff and colleagues found that a partner’s neuroticism correlated between -0.26 and -0.19 with relationship satisfaction, while agreeableness sat at 0.11 to 0.20 and conscientiousness at 0.09 to 0.16. Extraversion and openness barely registered. Meanwhile, research by Dyrenforth and colleagues put the contribution of personality similarity at under 0.5% of the variance in relationship satisfaction. In plain terms: who you are matters much more than how well you match.
That runs against most of what dating culture teaches: find your perfect type match, check whether an ENFP pairs well with an INTJ, worry that an introvert and an extrovert are doomed. The data points somewhere simpler — look at emotional stability, kindness, and follow-through, in your partner and in yourself.
The short answer: similarity matters, but far less than you think
Relationship researchers separate personality effects into three buckets, and understanding those buckets clears up most of the confusion about compatibility.
- Actor effects — how your own traits relate to your own satisfaction.
- Partner effects — how your partner’s traits relate to your satisfaction.
- Similarity effects — whether being alike or different as a couple changes anything.
The consistent finding is that actor effects are the largest, partner effects are modest, and similarity effects are close to negligible. Your own emotional stability predicts your relationship happiness better than any matching exercise predicts anything.
Dyrenforth, Kashy, Donnellan, and Lucas reported that a partner’s personality explains roughly 1–3% of the variance in relationship satisfaction, and that similarity between partners consistently explains less than 0.5%. Those are small numbers, and they are worth sitting with. If you have ever ruled someone out because a chart said your types clash, you were acting on a factor that research can barely detect.
The Big Three: why three traits dominate relationship outcomes
The Big Five — also called the Five Factor Model or the OCEAN model — describes personality along five continuous dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Unlike systems that sort people into fixed categories, each trait is a spectrum you can score higher or lower on. In relationship research, three of those five consistently do the heavy lifting.
Neuroticism is the strongest predictor
Neuroticism describes how readily you experience negative emotion — anxiety, irritability, mood swings, emotional reactivity. It is the single best-established personality predictor of relationship trouble, and it works in both directions: your own neuroticism lowers your satisfaction, and your partner’s neuroticism lowers yours.
A meta-analysis by Malouff and colleagues, published in the Journal of Research in Personality, found partner neuroticism correlated between -0.26 and -0.19 with relationship satisfaction. Longitudinal work pushes the point further. Robins, Caspi, and Moffitt followed married adults over 18 years and found that rising neuroticism tracked with declining marital satisfaction, suggesting a feedback loop in which a strained relationship and a reactive temperament make each other worse.
If one trait deserves your attention, it is emotional stability — both yours and your partner’s.
Agreeableness is the trait partners notice most
Agreeableness covers warmth, trust, cooperation, and the tendency to assume good intent rather than picking fights. In the same Malouff meta-analysis, partner agreeableness correlated 0.11 to 0.20 with satisfaction — the strongest positive partner effect in the set.
Most of relationship life is not grand gestures; it is small negotiations about dishes, schedules, money, and tone of voice. Agreeableness is essentially the trait of handling small friction without turning it into a war.
Conscientiousness is the reliability trait
Conscientiousness reflects organization, dependability, impulse control, and follow-through. Partner conscientiousness correlated 0.09 to 0.16 with satisfaction in the Malouff analysis. Its real power shows up over time: conscientious people pay bills on time, keep commitments, share household labor, and do not create preventable crises.
One nuance: Robins and colleagues found that very high initial conscientiousness predicted slightly steeper declines among younger adults, possibly because rigid standards create friction. The broad pattern still favors dependability.
Where extraversion and openness fit
Extraversion (sociability and energy drawn from interaction) and Openness to Experience (curiosity and appetite for novelty) show weak and inconsistent links to relationship satisfaction. Malouff’s numbers put partner extraversion at 0.01 to 0.10 and openness at -0.03 to 0.09 — essentially noise. Whether someone is the life of the party or a homebody tells you very little about whether the relationship will last.
Why “opposites attract” feels true even when it mostly isn’t
The belief survives because it contains a grain of truth wrapped in a large misunderstanding.
People do end up with partners who resemble them — a pattern researchers call assortative mating, the tendency to pair with those similar to oneself at rates above chance. But the resemblance is modest. A meta-analysis by Horwitz and colleagues found actual partner correlations across the Big Five ranged from about 0.08 for extraversion to 0.21 for openness. Those are real but small.
The stronger version of the effect lives in preferences rather than reality. A 2025 study of 538 Norwegian students by Kleppestø and colleagues, titled “The attractive personality: Like me, but better,” found that both sexes preferred partners similar to themselves on Big Five traits — with one blaring exception: everyone wanted lower neuroticism than they had. The ideal partner is recognizable as “like me, but calmer.”
When people say opposites attract, they usually mean differences in hobbies, surface style, or the relief of being with someone who compensates for a weakness. None of those are the traits that predict long-term outcomes.
The problem with 16-type compatibility charts
The MBTI (Myers-Briggs Type Indicator) sorts people into 16 personality types from four letter pairs — INTJ, ENFP, and so on. Compatibility charts ranking all 136 possible pairings circulate constantly online, and very little of that rests on outcome data.
Three issues explain the gap:
- Test-retest reliability. A meaningful share of people receive a different four-letter result when retested weeks later. A label that shifts on retest is a shaky basis for forecasting years of shared life.
- No dose-response with outcomes. Marioles and colleagues followed 426 couples for seven years and found minimal statistical support for the idea that complementary or “opposite” pairings outlast similar ones.
- Untested pairing rules. The charts themselves were built from theory and community consensus, not from tracking which pairings survived. The Myers & Briggs Foundation’s own ethical guidance discourages using type to make decisions about people, and there is no officially endorsed compatibility chart.
Every 16-type compatibility chart in circulation is an unofficial product, not a summary of a body of evidence. That does not make the framework worthless. Type vocabulary is genuinely useful for naming differences in communication style — how much social contact recharges you, whether you plan in concrete details or loose possibilities, whether you resolve disagreement by criteria or by impact on people. Those are real conversational friction points. The error is treating a descriptive vocabulary as a prediction machine.
Where similarity does show up — two specific places
Similarity is not entirely inert. A 2017 study by Weidmann, Schönbrodt, Ledermann, and Grob, published in the Journal of Research in Personality, analyzed 237 couples using dyadic polynomial regression, a method that models both linear and curvilinear similarity patterns. Their concurrent results echoed the broader literature: beyond actor and partner effects, similarity played a negligible role. But two longitudinal effects did emerge.
- Neuroticism gaps. When partners reported dissimilar neuroticism levels, male partners reported lower relationship satisfaction over time. A large stability gap between two people seems harder to live with than two people who are both somewhat reactive.
- Modest shared openness. When both partners reported moderate levels of openness, female partners reported higher satisfaction — a hint that a shared appetite for (but not an extreme pursuit of) novelty supports the relationship.
Notice that both findings are about a gap or a shared level, not about being the same type. What seems to hurt is not difference itself but a large mismatch in emotional reactivity.
Your personality also changes inside the relationship
Treating compatibility as a fixed verdict assumes both people stay the same, and they do not. Personality develops across adulthood, and relationships are one of the forces that shape it.
A study of newlyweds by van Aken and colleagues tracked couples across their first 18 months of marriage and found measurable shifts: both partners declined in agreeableness, husbands in extraversion, wives in openness and neuroticism. Wives whose neuroticism rose saw steeper satisfaction declines for both partners.
The implication cuts both ways. A partner who is prickly today is not permanently so, and a steady one is not guaranteed to stay that way. Stability is a pattern you build together, not a fixed asset one person either has or lacks.
What to actually do with this
If the research points anywhere actionable, it points here:
- Weight emotional steadiness heavily. Not as a dealbreaker test, but as a genuinely relevant signal. Chronic reactivity is the most reliable personality-linked stressor in the literature.
- Pay attention to how someone handles small friction. That is agreeableness in action, and it predicts more than shared interests do.
- Notice whether promises get kept. Follow-through is conscientiousness, and it compounds over years.
- Treat type labels as conversation starters. Useful for naming why you miscommunicate; not useful for deciding whether someone is worth your time.
- Work on your own side of the ledger. Actor effects are the largest in the data. Your traits predict your satisfaction more than your partner’s traits do.
If you want to see where you actually land on these dimensions, taking a structured assessment is a reasonable starting point. Sites such as personalitree.com offer Big Five and 16-type assessments, and looking at your own profile — particularly emotional stability and agreeableness — gives you more usable information about your relationships than any pairing chart will.
Frequently asked questions
Do opposites attract in relationships?
Mostly no. Research on assortative mating finds that partners are slightly more similar than chance would predict, with correlations ranging from about 0.08 for extraversion to 0.21 for openness. A seven-year study of 426 couples found minimal support for the idea that complementary pairings last longer. Differences in hobbies and style are common, but those are not the traits that predict outcomes.
Which personality trait matters most in a partner?
Emotional stability — the low end of neuroticism — has the strongest and most consistent link to relationship satisfaction. Malouff’s meta-analysis found partner neuroticism correlated between -0.26 and -0.19 with satisfaction, a larger effect than any other trait. Agreeableness and conscientiousness follow as the next most relevant.
Does personality similarity predict a happy relationship?
Barely. Dyrenforth and colleagues found similarity explains under 0.5% of the variance in relationship satisfaction. One exception is neuroticism: Weidmann’s 2017 study of 237 couples found that a large gap in emotional reactivity was associated with lower satisfaction for male partners over time.
Are MBTI compatibility charts reliable?
No. There is no large longitudinal study showing that specific 16-type pairings outlast others, and the Myers & Briggs Foundation does not endorse compatibility rankings. The charts are built from theory and community consensus rather than outcome data. The Big Five framework has considerably stronger research support for predicting real-world outcomes.
Can an introvert and an extrovert have a good relationship?
Yes. Extraversion shows weak and inconsistent links to relationship satisfaction, with partner effects around 0.01 to 0.10. What tends to matter is not the difference itself but whether the couple negotiates it explicitly — how often to socialize, how much recovery time each person needs, and who carries the planning load.