Compared to the US population, the proportion of PhDs in their sample seems inflated, and the proportion of people with only a high school diploma is lower then

Compared to the US population, the proportion of PhDs in their sample seems inflated, and the proportion of people with only a high school diploma is lower then expected. I wonder how much one can trust a self-reported PhD degree, especially from a survey collected on the internet.