The Policy Basis of Measured Partisan Animosity in the United States

Abstract Understanding and addressing the consequences of partisan animosity requires knowledge of its foundations. To what extent is animosity between partisan groups motivated by dislike for partisan outgroups per se, policy disagreement, or other social group conflicts? In many circumstances, including extant experimental research, these patterns are observationally equivalent. In a series of vignette evaluation experiments, we estimate effects of shared partisanship when additional information is or is not present, and we benchmark these effects against shared policy preference effects. Partisanship effects are about 71% as large as shared policy preference effects when each is presented in isolation. When an independently randomized party and policy position are presented together, partisanship effects decrease substantially, by about 52%, whereas policy effects remain large, decreasing by about 10%. These results suggest that common measures of partisan animosity may capture programmatic conflict more so than social identity–based partisan hostility.