1/n Data from an unpublished 2016 survey experiment shows that white participants assigned to read an article about whites' declining majority status (racial threat condition) did not react with greater Trump support (if anything, attitudes moved in the opposite direction). 
2/n Nor was the (null) effect conditional on party identification.
@epkaufm I know there are other studies that attest to the relationship. But might it be less robust than previously acknowledged?
@epkaufm Perhaps most surprising, it also had no effect on support for cutting immigration
