The matching hypothesis is a social psychology theory proposing that people tend to form romantic relationships with partners who are similar to them in physical attractiveness. First introduced by Elaine Walster and colleagues in 1966, the idea draws on a simple principle: when looking for a partner, people assess their own desirability and gravitate toward someone at roughly the same level. Decades of research have largely supported this pattern, though the story turned out to be more complicated than the original researchers expected.
The 1966 Computer Dance Study
The matching hypothesis got its start with one of psychology’s most famous experiments. Walster and her team organized a “Computer Dance” at the University of Minnesota, where students were randomly paired with dates. The researchers predicted that people would prefer partners who matched their own level of attractiveness, applying what’s known as level-of-aspiration theory to dating choices.
The results didn’t cooperate. Regardless of how attractive the participants themselves were, the single biggest factor in whether they liked their date and wanted to see them again was simply how physically attractive that person was. Personality traits, self-acceptance, and academic ability had no measurable effect on compatibility. Everyone, it seemed, just wanted the most attractive partner they could get.
This created an interesting puzzle. The initial experiment suggested people don’t match at all. But later studies, including a 1971 follow-up by Berscheid, Dion, and the Walsters, found that when people had to actively choose a partner rather than being randomly assigned one, matching patterns appeared clearly. The difference was context: in a one-shot random pairing, people aim high. When they have to put themselves on the line and risk rejection, they become more realistic.
How Matching Actually Works
The psychological engine behind matching is self-assessment. People evaluate their own “market value,” so to speak, and then pursue partners whose social desirability is close to their own. This isn’t always a conscious calculation. It often plays out as a gut feeling about who seems approachable or “in your league” versus who feels like a stretch.
Research using both lab experiments and data from online dating sites has confirmed that matching operates on multiple dimensions: self-perceived worth, physical attractiveness, and social popularity. But these factors don’t all kick in at the same stage. Early in the dating process, people may shoot higher. As interactions progress and the possibility of rejection becomes real, matching behavior becomes stronger. The hypothesis doesn’t claim that people settle. It claims that over time, the push and pull of mutual selection naturally sorts people into pairs of similar desirability.
What the Numbers Show
A recent meta-analysis combining data from multiple studies found a meaningful correlation of 0.39 between partners’ physical attractiveness as rated by outside observers. That means couples do, on average, end up with someone at a similar attractiveness level, though it’s far from a perfect match. When researchers looked at how people rated their own attractiveness, the correlation between partners was 0.27, still positive but weaker. This gap makes sense: people aren’t always accurate judges of their own appearance, and self-ratings involve confidence and self-esteem, not just looks.
Interestingly, how attractive you think you are also correlates modestly with how attractive outside observers rate your partner (correlations around 0.14 to 0.18). Self-perception and reality overlap, but imperfectly. People who see themselves as more attractive do tend to end up with partners others consider more attractive, just not as strongly as the objective matching pattern would suggest.
Matching Goes Beyond Looks
While the original hypothesis focused on physical attractiveness, modern research shows matching operates across nearly every measurable characteristic. A large-scale study analyzing over 421 million potential matches on a mobile dating app found that for almost every trait examined, greater similarity between two people meant a higher chance they’d want to meet in person.
The specifics are striking. Two people who shared the same religion were about 97.5% more likely to match effectively than people of different religions. Users who both attended an Ivy League school matched at more than double the overall average rate. Even something as seemingly trivial as using the same type of smartphone was associated with an 8.9% increase in successful matches, despite the fact that neither person could see what phone the other used. People with the same initials matched 11.3% more often than those with different initials.
Shared relationship goals mattered too. When both users said they wanted a committed relationship, their match rate jumped to 0.20% compared to 0.13% when only one person wanted commitment. The one notable exception to the similarity-attracts pattern was introversion: introverts rarely matched well with other introverts. Pairs with at least one extrovert fared significantly better, and two extroverts matched most effectively of all.
Attractiveness Gaps and Relationship Stability
The matching hypothesis implies something about what happens when partners aren’t well matched, and research backs this up. A series of studies found that people rated as more attractive in their high school yearbook photos went on to have shorter marriages and higher divorce rates. The same pattern held for high-profile celebrities. More attractive individuals showed less tendency to mentally dismiss appealing alternatives to their current partner, a psychological behavior that normally acts as a kind of relationship glue.
The mechanism seems to be that highly attractive people have more options, which creates more opportunities for dissatisfaction. When someone attractive is in a relationship that hits a rough patch, they’re less likely to brush off the attention of an attractive alternative. This doesn’t mean attractive people can’t sustain relationships. It means that when a gap exists between partners, the more attractive partner faces more external pull, and the relationship requires more active maintenance to stay stable.
Why Matching Matters in Everyday Dating
The practical takeaway from the matching hypothesis isn’t that you should carefully calculate your attractiveness score and limit yourself accordingly. It’s that the dating market naturally sorts people through a process of mutual selection. You pursue people you find attractive; they accept or decline based partly on how attractive they find you. Over many interactions, this process tends to produce couples of roughly similar desirability.
This also explains the common experience of feeling like someone is “out of your league.” That instinct reflects a real, if imperfect, self-assessment process. People who aim far above their own attractiveness level face more rejection, which gradually recalibrates their expectations. People who aim below it tend to find relationships easily but may not feel fully satisfied. The matching hypothesis suggests the sweet spot is somewhere in the middle, where both partners feel they’ve done well without feeling they’re constantly at risk of losing the other person.
The evidence from dating apps confirms this plays out at scale, even in digital environments where swiping feels impulsive. Beneath the surface, people are drawn to similarity across dozens of dimensions, from education and religion to relationship goals, forming patterns that are remarkably consistent with what Walster proposed nearly 60 years ago.