The more formal point you're making here is that the survey data social psychologists work with can be explained by an infinite number of hypothesis, and the "s

The more formal point you're making here is that the survey data social psychologists work with can be explained by an infinite number of hypothesis, and the "statistical testing" they do is basically pseudoscience. All they ask is "is my favoured hypothesis more likely than random?" and that's true, but it says nothing about lots of other explanations. Here everything can be explained by lines of trust, and resilience to information from low-trust sources. Indeed, it's highly rational to highly doubt sources you do not trust, because you are almost never in a position to validation information. Either way, the explanations they choose lead to a clearly partisan political narrative, way outside of the scope of their survey studies, that science washes attacks on trump and his supporters.