Perhaps the takeaway should be that these sorts of studies need a control group. Now it may be difficult to find a population that would be good to serve as a c

Perhaps the takeaway should be that these sorts of studies need a control group. Now it may be difficult to find a population that would be good to serve as a control group, which I think points toward the idea that these sort of studies should be used to examine the differences between groups, rather than to make claims about any group in absolute terms. I also think that studies of the form "people who disagree with me are stupid because they think X" are fundamentally dumb. This particular issue is just a small part of why I think that.