Ah, this feels interesting. I would have said I’m a below average dancer, but above average person at dancing. The difference being whether I compare to people

Ah, this feels interesting. I would have said I’m a below average dancer, but above average person at dancing. The difference being whether I compare to people who I observe dancing, or consider everyone. Differently for sex, I assume that this is possibly bimodal (none versus some), so this could provide a measure of the actual skew in the underlying distribution (difference between mean and median) and not only the perception of bias.