Some commenters are criticizing the study in Henry’s original post. Perhaps the study’s analysis is flawed, but I think the finding stands based on other evidence. First, see the graphs in Andy’s follow-up post. Second, here are some graphs that were sent to me by Michael Tesler.
The graph below makes clear that better-educated liberals and conservatives are more polarized on global warming than their less well-educated kindred. The epistemic closure finding, again, is that well-educated (and therefore politically attentive) conservatives are less likely to believe in the scientific consensus regarding global warming than are less well-educated conservatives.
Tesler finds a somewhat similar pattern in attitudes toward evolution:
I say “somewhat similar” because here, better-educated conservatives are not necessarily less likely than less well-educated conservatives to deny facts about evolution. More striking is just how little formal education is associated with correct beliefs among conservatives.
Furthermore, the pattern of findings regarding global warming appears unique to the United States. In a host of countries where elites are not as divided on global warming, the pattern of polarization does not exist.
One explanation for this is familiar to any reader of John Zaller’s The Nature and Origins of Mass Opinion: when political elites take contrasting positions on issues, those positions will be reflected in their fellow partisans in the public, at least among those who are paying enough attention to politics to receive these elite messages.