How well do people do on estimates like this compared to estimates about other subjects?
Because it kind of seems like a lot of the respondents are just plain innumerate.
For instance, on the open borders question, the plurality of both party respondents
answered that members of the opposing party had the most extreme possible view.
That's obviously not going to be true on average for anything resembling a normal distribution
no matter how polarized you think the parties are.
Also, I bet respondents were confused about the meaning of "open borders."
I've encountered people who somehow think that
current (or pre 2016) US immigration policies qualify as "open borders"
which would make the estimate that most democrats support "open borders" much closer to accurate.
