In my culture, it is quite common to rate people on a 10-point scale of physical attractiveness. I think a lot of people don’t think very much about this scale when they’re using it. For example, it’s from 1 to 10 inclusive, so 5.5 is the middle of the scale.

I often suspect that people are using it as a uniform (flat) distribution. The subjects seem to be “9s” and “10s” more often than I would expect from a normal distribution with reasonable variance. And not surprisingly, the most common subjects below the middle are notably unattractive. I treat it as a normal distribution: chances are you’re a 5 or a 6.

OkCupid is an online dating site that I love for its four-factor Dating Persona Test. The site was built by four Harvard math grads. Kyla told me that they’ve started crunching numbers on users activity and publishing the results. (Statistical analysis is more legitimate than speed dating research, because they don’t pretend they’re in a laboratory where they can control all the variables.)

One of their best results is how men users rate the attractiveness of womens’ profile pictures and messages from men to women:

Men rate women on a normal distribution and disproportionately hit on attractive women. The old story about all the guys ignoring the most attractive woman because they assume she’s out of their league is confirmed.

Women, on the other hand, think all men are ugly:

But women will message ugly guys, as long as they’re not too ugly.