It's security, stupid! Voters' perceptions of immigrants as a security risk predicted support for Donald Trump in the 2016 US presidential election

We analyzed two datasets to determine the predictive validity of four explanations of support for Donald Trump during the 2016 US presidential election: (a) security concerns regarding immigrants, (b) economic concerns regarding immigrants, (c) cultural concerns regarding immigrants, and (d) social dominance orientation. Results of a two-phase study (N = 354) suggested that perceiving immigrants as a security concern was predictive of increased support for and greater odds of voting for Donald Trump three weeks later. Perceiving immigrants as an economic threat predicted odds of voting for Donald Trump, but only among liberals and there was no evidence of cultural concern or social dominance orientation (SDO) predicting support for Donald Trump or odds of voting for Trump. A follow-up analysis of the cross-sectional ANES survey corroborated that security concerns were an important correlate of voting for Trump, but also that SDO was correlated with having previously voted for Donald Trump. While our two-phase study has the benefit of prediction, the cross-sectional ANES data does not-"predictors" in these data were collected up to two months post-election.