A Spatial Exploration of the Halo Effect in the 2016 U.S. Presidential Election

Abstract

The outcome of the 2016 presidential election in the United States was partly influenced by factors such as social marginalization and anti-immigrant sentiment, both of which have been associated with the global rise in far-right voting (FRV) outcomes. Sociological hypotheses such as group threat and group contact have been suggested as potential contextual factors in the relationship between immigrant share and far-right support; more recently, the halo effect has been used as a spatial mechanism to explore these relationships. Briefly, the halo effect describes increased FRV in ethnically homogeneous areas that are near ethnically diverse areas. This portentous trend has significant international implications, but differences in the spatial distribution and composition of immigrant share and the spatial scale of analysis have produced inconsistent empirical results. This study addressed the spatial distribution of the halo effect in the 2016 U.S. presidential election and used geographically weighted regression to explore spatial nonstationarity in the relationships between socioeconomic factors and Republican (GOP) support. Spatial nonstationarity is observed when a relationship is not globally consistent such as in large or complex areas where local processes might interact with or overwhelm global processes. We found evidence of spatial nonstationarity in the relationships between GOP support and many socioeconomic factors, including the halo effect. In addition to providing a better understanding of the spatial context of electoral dynamics and their interactions at a local scale, these results indicate that spatial nonstationarity in these relationships might be partially responsible for inconsistent results in previous studies.