I really think that the 1% prevalence number thrown around for INTPs in the general population has to be an underestimate. I'm wondering how they came up with

I really think that the 1% prevalence number thrown around for INTPs in the general population has to be an underestimate. I'm wondering how they came up with that number. Where did they get their test sample? Surely if you go around asking people to take personality tests at random, their likelihood of saying "yes" vs "no, sod off" is going to be a lot stronger for some personality types than others.