The theory says you quietly rate yourself, quietly rate everyone else, and aim for someone at roughly your own level. It sounds plausible and slightly bleak. Then someone tested it against what people actually do on a dating site, and the results did not cooperate.
📘 What you need to know
The matching hypothesis (Walster, 1966) proposes that people choose partners of a similar level of physical attractiveness to themselves.
It is a cognitive process: self-assessment plus assessment of others, both based on physical appearance.
It connects to self-worth and social status. A partner perceived as less attractive may reduce your social status; one perceived as more attractive may raise it.
There is a balancing act: aiming higher carries a greater risk of rejection, aiming lower reduces that risk but may not be fully satisfying.
Repeated rejection can damage self-esteem, so over time people adjust their expectations towards partners they think are realistically attainable.
Taylor et al. (2011) tested it using real online dating behaviour and found no support for it.
The theory is often criticised as simplistic and reductionist — the “I am a 5 so I seek a 5” problem.
The balancing act
The mechanism depends on people learning from rejection. If they do not adjust — and Taylor’s initiators did not seem to — the theory loses its engine.
This is one of the rare topics where the headline study contradicts the theory it was testing. Do not hide that. An essay that explains a disconfirming result and then asks what it means is worth far more than one that pretends the evidence lined up.
Research support — or rather, against: Taylor et al. (2011)
Aim: to investigate the matching hypothesis using real-life online dating behaviour.
Participants: 60 heterosexual male and 60 heterosexual female profiles chosen at random from an online dating site. All were initiators — they had started contact with other users.
Procedure: researchers tracked the initiators’ real online activity and recorded who replied (reciprocating contacts) and who did not (non-reciprocating contacts). Up to six profile photographs were collected for each initiator and each of their contacts, giving 966 photographs in total (527 female, 439 male). Independent blind researchers rated every photograph on a 7-point attractiveness scale from −3 to +3. Calculations compared each initiator’s mean rating with the mean rating of the people they contacted, and separated reciprocating from non-reciprocating contacts.
Results: the findings did not support the matching hypothesis. Initiators’ own attractiveness showed no correlation with the mean attractiveness of those they contacted, and initiators consistently approached people rated more attractive than themselves.
Conclusion: the matching hypothesis cannot fully account for real-life attraction and partner choice in online dating.
Why online dating changes the sums. The theory’s engine is fear of rejection, and rejection online is cheap: no public embarrassment, no witnesses, and you can message twenty people in an evening. If the cost of aiming high collapses, the reason to match disappears. That is a better explanation of Taylor’s result than “the theory is simply wrong”, and it is the kind of reasoning that scores.
Evaluation
Strengths
Equally attractive couples do exist, which gives the theory at least anecdotal support.
It takes cognition seriously. The theory highlights how thinking — self-assessment, weighing risk, adjusting expectations — may determine partner preference, which fits our capacity for higher-level thinking.
Taylor’s design used blind raters and real behaviour. Independent raters who did not know the hypothesis removes a major source of bias, and tracking actual contacts beats asking people who they would date.
Limitations
The evidence does not support it. The most direct real-world test found no correlation at all.
Taylor’s sample is small for the claim being made, which limits how robust and reliable the findings are.
Simplistic and reductionist. Reducing human attraction to a balancing act — “I am a 5 so I seek a 5” — quantifies something that resists being quantified.
It ignores everything else. Humour, kindness, shared interests, status and simple proximity do not appear anywhere in the model.
Online behaviour may not be dating behaviour. Sending a first message is not the same as forming a relationship, so the test may not fully capture partner choice.
EXAM QUESTION
Evaluate one theory of the formation of personal relationships. [22]
Step 1: state the hypothesis and its cognitive basisWalster (1966); self-assessment plus assessment of others based on appearance.Step 2: explain the mechanism, not just the claimRisk of rejection versus satisfaction; repeated rejection adjusts expectations downwards.Step 3: use Taylor honestly966 photographs, blind raters, 7-point scale, no correlation, initiators aimed above their own rating.Step 4: explain the disconfirmation before dismissing the theoryOnline rejection is cheap, so the mechanism may simply not apply in that context. Then add reductionism and the missing variables.Top answers explain why a theory failed, not just that it did
Link to concepts
Causality
What counts as attractive is highly subjective and prone to individual differences, which weakens the scientific validity of the theory before any data is collected. Attraction does not lend itself easily to experimental testing, and is arguably better suited to qualitative research that asks people what actually drew them to someone.
Responsibility
Research that rates people’s physical attractiveness has to be handled with real care. If participants learn how they were rated, and the rating is lower than they expected, that could cause embarrassment or distress. Anonymity matters too: it would have been essential for Taylor’s team to ensure the dating site photographs were never published and that no clue was given as to whose images had been used.
💡 Exam tip
Give the study’s actual numbers: 120 profiles, 966 photographs, a 7-point scale from −3 to +3.
Say the raters were blind and explain what that prevents.
Do not report Taylor as supporting the hypothesis. It did the opposite, and examiners notice when students bend it.
The cheap-rejection explanation for the online result is the strongest analytical point on this page.
Use “cognitive process” deliberately — this is a cognitive theory, not a biological one.
Contrast with the dopamine page: one explains attraction chemically, the other by calculation. Neither covers the whole thing.
⚠ Common mix-up
Saying Taylor supported the matching hypothesis. The initiators contacted people rated more attractive than themselves.
Treating “no correlation” as a weak result. A clear absence of correlation is a strong finding against the theory.
Confusing the matching hypothesis with “opposites attract”. It predicts similarity, not difference.
Assuming attractiveness ratings are objective. They are averaged judgements, not measurements.
Forgetting the reciprocating and non-reciprocating split. Separating those groups is central to the design.
Applying the theory to all relationships. It is about physical attractiveness at the formation stage only.
Up next: Self-Disclosure and Growing Closer — attraction gets a relationship started, but something else has to happen for it to deepen.
Want this explained one-to-one?
Book a free session with an experienced IB Psychology tutor and get your trickiest topics made simple.
