Perpetuates the false dichotomy of a linear political spectrum of either liberal or conservative and that in and of itself is one of the reasons for these trends

the graph explicitly takes 1 dimension of the spectrum to look at it in isolation. This is exactly what single graphs do best.

which this survey completely ignores, conveniently

have you looked at the source of the graph?