There is far, far, far more concentration of a liberal mindset in social science researchers than one would predict from general IQ/openness measures.

30 years ago, the explanation was plausible. These days it's really not.

The difference used to be ~3x. Now it's ~6x. It's ~28x in New England: https://www.insidehighered.com/news/2016/07/05/new-analysis-new-england-colleges-responsible-left-leaning-professoriate

This is the kind of outcome that you'd expect from a process that filters for people of the "correct" outlook, not from an organic process where people with certain traits have an increased tendency to come to similar conclusions.