One of the ways we know that racism is a giant all-encompassing social factor is the Implicit Association Test. Psychologists ask subjects to quickly identify whether words or photos are members of certain gerrymandered categories, like either a white persons face or a positive emotion or either a black persons face and a negative emotion. Then they compare to a different set of gerrymandered categories, like either a black persons face or a positive emotion or either a white persons face or a negative emotion. If subjects have more trouble (as measured in latency time) connecting white people to negative things than they do white people to positive things, then they probably have subconscious positive associations with white people. You can try it yourself here.
Of course, what the test famously found was that even white people who claimed to have no racist attitudes at all usually had positive associations with white people and negative associations with black people on the test. There are very many claims and counterclaims about the precise meaning of this, but it ended up being a big part of the evidence in favor of the current consensus that all white people are at least a little racist.
Anyway, three months ago, someone finally had the bright idea of doing an Implicit Association Test with political parties, and they found that peoples unconscious partisan biases were half again as strong as their unconscious racial biases (h/t Bloomberg. For example, if you are a white Democrat, your unconscious bias against blacks (as measured by something called a d-score) is 0.16, but your unconscious bias against Republicans will be 0.23. The Cohens d for racial bias was 0.61, by the book a moderate effect size; for party it was 0.95, a large effect size.
Okay, fine, but we know race has real world consequences. Like, there have been several studies where people sent out a bunch of identical resumes except sometimes with a black persons photo and other times with a white persons photo, and it was noticed that employers were much more likely to invite the fictional white candidates for interviews. So just some stupid Implicit Association Test results cant compare to that, right?
Iyengar and Westwood also decided to do the resume test for parties. They asked subjects to decide which of several candidates should get a scholarship (subjects were told this was a genuine decision for the university the researchers were affiliated with). Some resumes had photos of black people, others of white people. And some students listed their experience in Young Democrats of America, others in Young Republicans of America.
Once again, discrimination on the basis of party was much stronger than discrimination on the basis of race. The size of the race effect for white people was only 56-44 (and in the reverse of the expected direction); the size of the party effect was about 80-20 for Democrats and 69-31 for Republicans.