The internet answers this question with lists: eye contact, hair touching, feet pointed toward you, quick replies, laughter, proximity. The problem is not that these things never appear when someone is interested. The problem is that each one can also appear when she is friendly, comfortable, distracted, nervous, polite or simply being herself.
A cue can raise a possibility. It cannot privately settle another person’s intention.
What the flirting study measured
Jeffrey Hall, Chong Xing and Seth Brooks paired 52 men and 52 women for recorded conversations lasting 10 to 12 minutes. All participants described themselves as single and heterosexual, and each pair consisted of strangers. Afterwards, each person privately answered two yes-or-no questions: “Did you flirt?” and “Did your partner flirt?”
Fourteen women said they had flirted. Their male partners agreed in 5 cases, or 36%. Eleven men said they had flirted. Their female partners agreed in 2 cases, or 18%. Across the whole sample, partners agreed with 7 of 25 self-reports of flirting, or 28%. They agreed with 66 of 79 reports of not flirting, or 84%.
The limit that changes the headline
This was agreement with a partner’s yes-or-no self-report in one small college sample. The study did not measure romantic attraction, sexual interest or consent, and it did not establish a sex difference in detection. The 18% figure came from 2 agreements among 11 male self-reports of flirting. It is not a general accuracy score for women.
The finding matters because most self-reported flirting in that setting went unrecognized. It does not create a reverse rule such as “she probably likes you even if you see nothing.” Both conclusions would reach beyond the evidence.
People make errors in both directions
Other studies show a second problem. Men can overestimate women’s sexual interest relative to women’s self-reports. They can also mistake sexual interest for friendliness. Signal-detection work by Coreen Farris and colleagues separated bias from sensitivity and found that the difficulty was not just a thumb on one side of the scale. Distinguishing the two categories was itself imperfect.
In a separate undergraduate speed-meeting study, men overestimated women’s sexual interest and women underestimated men’s, relative to each person’s own report. Later work with 1,226 participants across 3,850 interactions found that men’s projection of their own interest and beliefs about how much a woman liked them explained much of the difference. The misperception is well documented. The complete explanation remains contested.
What common “signs” can tell you
| What you notice | A fair interpretation | What it cannot prove |
|---|---|---|
| She holds eye contact | She is attentive and engaged in this moment. | That she wants a date, touch or sexual contact. |
| She laughs and asks questions | The conversation may be enjoyable and reciprocal. | That friendliness has become romantic interest. |
| She replies quickly | She chose to reply quickly this time. | Why she did, or what she wants next. |
| She stays nearby | She is comfortable remaining in the setting. | That proximity is an invitation to escalate. |
| Several cues appear together | Your hypothesis may become more plausible. | Certainty, permission or consent. |
Context and patterns matter. They can help you decide whether a small invitation would be appropriate. They do not convert observation into access to her private answer.
A better way to find out
When the setting is appropriate and the interaction feels mutual, move from inference to a clear choice.
- Keep the step small. Suggest coffee, exchange numbers or name the next simple thing you would actually like.
- Make it answerable. “Would you like to get coffee Thursday?” is clearer than a vague promise to hang out sometime.
- Leave the exit open. Ask once. Do not demand an explanation, punish a no or keep negotiating.
- Update from what comes back. An engaged yes, a proposed alternative or continuing reciprocity supports another small step. A no ends it. Withdrawal, hesitation or repeated vagueness means step back.
You are not trying to extract certainty. You are creating a moment in which the other person can choose, and then taking that choice seriously.
What if you were wrong?
If you thought she was interested and she declines, the answer is now clearer. Accept it without making the interaction pay for your embarrassment. If you thought she was only being friendly and she makes a direct offer, you have new information. Respond to the offer instead of defending your earlier read.
The goal is not perfect prediction. It is a good response to uncertainty.
Research cited
- Hall, J. A., Xing, C., & Brooks, S. (2015). “Accurately detecting flirting.” Communication Research, 42(7), 939–958.
- Perilloux, C., Easton, J. A., & Buss, D. M. (2012). “The misperception of sexual interest.” Psychological Science, 23(2), 146–151.
- Farris, C., Treat, T. A., Viken, R. J., & McFall, R. M. (2008). “Perceptual mechanisms that characterize gender differences in decoding women’s sexual intent.” Psychological Science, 19(4), 348–354.
- Lee, A. J., et al. (2020). “Sex differences in misperceptions of sexual interest can be explained by sociosexuality and men’s projection of their own sexual interest.” Psychological Science, 31(2), 184–192.
Read the study limits, not just the headline
The first three chapters are free
Part One of You Can’t Read Her examines flirting detection, romantic prediction and retracted dating research. It includes the one-page Field Card.