Do the Things People Say They Want Actually Predict Who They're Attracted To?

Often, no, or at least far less than you’d expect. In a landmark speed-dating study, what people said they wanted before meeting anyone failed to predict who they were actually drawn to face to face. The gap between the stated preference and the lived one is one of the most interesting findings in dating research, and it quietly undermines a lot of how we talk about attraction.

The study everyone cites

In 2008, Paul Eastwick and Eli Finkel ran a speed-dating experiment with a follow-up over the weeks after. Beforehand, participants reported the usual pattern you’d predict from decades of surveys: on paper, men emphasised physical attractiveness more and women emphasised earning prospects more. Then people actually met.

What they found is the striking part. Those stated ideals, measured before the event, didn’t predict what actually drove people’s romantic interest once they were in the room. The traditional differences that showed up so clearly on the questionnaire largely stopped mattering in the live interaction. What people said they wanted and what moved them turned out to be two different things.

The honest nuance

This isn’t a settled, one-line fact, and it would be dishonest to present it as one. Later work, including a 2015 replication and a broader debate about the predictive validity of ideal partner preferences, has produced mixed results, with some studies finding that stated preferences do carry some predictive weight in certain contexts. So the careful version is this: what you say you want in the abstract is a weak guide to what actually moves you when you meet a real person. Not meaningless, but not to be trusted the way we tend to trust it.

Why the gap exists

Part of the answer is that we don’t have great access to our own inner workings. Eastwick and Finkel connected their result directly to Nisbett and Wilson’s classic point that people often can’t accurately report the real reasons behind their own reactions. You can sincerely believe you have a type and still be reliably surprised by who you fall for.

What this means for dating

If stated preferences are a shaky guide, then advice and matching built on them is standing on sand. The more reliable signal is behaviour: what actually happens between two people, and how that behaviour gets read. That’s the thing worth measuring, and it’s what cwēn is built on.

See what a behaviour-first approach looks like at askcwen.com.

Quick answers

So do preferences matter at all? They matter, but less than we assume, and mostly not in the tidy way we state them. What you say you want in the abstract predicts your real-world attraction weakly and inconsistently.

What predicts attraction better than stated preferences? Actual behaviour and interaction. How two people respond to each other in real situations carries more signal than either person’s pre-written list of wants.