Algorithms Fail Tremendously at Finding Us Love
Love can’t be calculated mathematically. That’s the conclusion of a new study published in Psychological Science, which reveals just how terrible formulas can be at predicting compatibility between singles.
Researchers created and tested a standard questionnaire-based algorithm, similar to what most dating sites use. To determine how algorithms fared at setting up couples who’d be interested in a second date, researchers used 100 “self-reported traits” (a liking for horror films, for example) to match daters based on mutual interests. Researchers then sent these singles on four-minute dates and checked in with them after it was over.
Researchers began believeing they’d be able to predict interest in a second date based on shared mutual interests, (which, logically speaking, would indicate they’d be compatible) but results proved otherwise. Responses were unpredictable and shared mutual interest were no sign that a couple was interested in pursuing a relationship.
“It predicted 0 percent [of the matches.] Some of the models we ran got a negative percentage, which means you’re better off just guessing,” lead author Samantha Joel said. “I thought we would be able to predict at least some portion of the variance—like extroverts or liberals like each other.”
“It’s a very elusive, mysterious thing. I don’t think people even know themselves what it is about a specific person,” Robin Edelstein, a psychologist at the University of Michigan says. “I don’t know if it’s about specific questions or specific traits.“
Past research published in the same journal drew a similar conclusion. “To date, there is no compelling evidence that any online dating matching algorithm actually works,” lead researcher Eli J. Finkel concluded. “If dating sites want to claim that their matching algorithm is scientifically valid, they need to adhere to the standards of science, which is something they have uniformly failed to do.”
Finkel insists it’s unlikely these algorithms work even in principle, given the limitations of matching procedures. “Developers of matching algorithms have tended to focus on the information that is easy for them to assess, like a similarity in personality and attitudes, rather than the information that relationship science has found to be crucial for predicting long-term relationship well-being. As a result, these algorithms are unlikely to be effective.”
The enduring upside is that dating sites do let singles meet quickly and easily, though experts suggest this convenience has come to overwhelm daters. “For years, the online dating industry has ignored actual relationship science in favor of unsubstantiated claims and buzzwords, like ‘matching algorithms,’ that merely sound scientific,” Finkel said. So, I don’t know what to tell you. Love is a strange thing that only you can know for certain. In that regard, we can’t help but lean on the words of lifestyle guru and The Hillsstar Lauren Conrad, who famously said, “Love is not a maybe thing. You know when you love someone,”
