Calculating Mate Value: A “Weighty” Task

By Gary Brase, Sydni Huxman & Jordann Brandner


Most of the time, you can only have one. Whether it is a college, a car, a home, or a relationship partner, people make difficult decisions about which options to pursue and which to not pursue. A big part of what makes these decisions difficult is that the options differ along many different traits. Possible relationship partners, for example, can vary on traits such as personality, intelligence, similarity of beliefs, health, physical attractiveness, facial attractiveness, financial status, and social status. These are all different traits, measured on totally different scales, yet somehow they must be combined into an overall evaluation and decision. There is a term for this overall evaluation: mate value. The decision about which possible partner to choose (“mate choice”) becomes easy once the mate values of the possible candidates are evaluated– choose the one with the highest mate value (assuming they will also choose you). That bypasses actually explaining much of the tricky work in this process, though. How is mate value calculated?

It turns out that there are many different possible ways to combine these qualitatively distinct traits to get an overall evaluation. Some strategies may allow high values on a trait to make up for being low on another trait, known as compensatory strategies. Others may be more non-compensatory, where traits can be “deal breakers” and a low value can eliminate a person from consideration. For example, if mate values is calculated with a compensatory calculation, extreme intelligence in a potential partner could make up for a poor financial condition. On the other hand, if mate value is based on a non-compensatory calculation, then no amount of intelligence can compensate for unacceptably poor finances. Yet, other calculations may be some mixture of the two, combining some compensatory aspects and some non-compensatory aspects.

In our recent research, we tested the performance of over seven different plausible ways that mate value could be calculated, across three different studies that used different types of information presentations. Our results show that the compensatory weighted additive model – where all the traits are combined, each adjusted for importance – outperformed all the others. The other models tested included an equal weights model, a “take-the-best” model, an aspiration models, a threshold model, Euclidean distance models, and correlational models. The last two models were also evaluated using participant’s rating of their “ideal” partner’s traits and their “realistic” partner’s traits (because you don’t always get everything you ideally want!).

One study had people choose “Person A” versus “Person B” based on tables that summarized each person’s as being either high or low on each of multiple traits (Personality, Intelligence, Similarity, Health, Body Attractiveness, Face Attractiveness, Finances, and Social Status). Each person in the study repeated this choice between two possible partners for 45 pairs of profiles, which were specifically designed to contrast the different possible mate value calculations. A second study used the same design, but had bars to show how high each possible partner was on each of the traits. A third study was designed to be closer to the format of dating sites; it used photos for each possible partner and descriptions of how well the person rated on each trait.

Regardless of how information was presented to people, the weighted additive model outperformed all the other options in accounting for the choices people made. Almost intuitively, it seems realistic that in the world of dating, high values on one trait may be able to make-up for other traits that they were lacking. Certainly, from an evolutionary standpoint, it might make sense that our ancestors would be willing to accept a partner who was extremely physically fit and healthy so that they could produce healthy offspring, even if they did not totally align in their belief systems or personality types. Part of what makes this research challenging is that all of the methods for integrating traits to calculate an overall mate value perform better than chance. The weighted additive model, though, did the best by predicting more of the choices than other models.


Almost intuitively, it seems realistic that in the world of dating, high values on one trait may be able to make-up for other traits that they were lacking.


Dating and choosing a partner is a complex process in the real world, and there are other factors involved beyond these studies. For example, potential partners (unlike cars and houses) have to reciprocate a positive choice. There are also possibilities that mate value evaluations incorporate other traits, or differ across demographic factors (such factors could include individual differences within populations or across populations – including non-WEIRD populations). There is also fascinating recent work on how people integrate the different traits of potential friends, extending this work to an additional, rich topic. Although there is much more work to be done in this arena, there is now strong evidence for the compensatory nature of traits in human dating decisions.



Read the paper: “Weighting” to find the right person: compensatory trait integrating versus alternative models to assess mate value