Not news: "Trump supporters are racist." News: Now with scientific evidence

[Associate Justice PiT]

Anyone else notice that the white one has the words "foreclosure", and the black one appears to say "reclosure". Not saying this affected the experiment but I just randomly noticed that.

The other flaw is that statistical significance is generally accepted at the 5% level. This means that if 30 out of 100 trump supporters were for housing assistance with the white dude, then 25 out of 100 were supportive with the black dude.

Do 5/100 people represent the whole? No, they do not.

     When they say 5% level, what they mean is that the odds that the difference are explained by random variance is less than 5% (note that statistical significance does not mean that it is impossible that the difference is explained by pure chance, but rather that it is improbable). It is not to say that there is a 5% margin of error.

[Associate Justice PiT]

     Anyway, I notice from the study that people opposed to Trump were more likely to support the program when presented with the version with the black person (this is also in contrast with the ideal situation, wherein government programs are regarded in a purely race-blind fashion). This makes me wonder if these reactions on both sides are related to the politicization of race and the poor state of race relations in recent years. The study notes that support for Trump captures this far more effectively than partisanship or ideology and thereby dismisses the notion, though I would posit that it is a mistake to think of support for Trump as a dimension unlike partisanship or ideology; rather, it is likely better correlated with positions on certain political issues than either of those are.

[Associate Justice PiT]

     Anyway, I notice from the study that people opposed to Trump were more likely to support the program when presented with the version with the black person (this is also in contrast with the ideal situation, wherein government programs are regarded in a purely race-blind fashion). This makes me wonder if these reactions on both sides are related to the politicization of race and the poor state of race relations in recent years. The study notes that support for Trump captures this far more effectively than partisanship or ideology and thereby dismisses the notion, though I would posit that it is a mistake to think of support for Trump as a dimension unlike partisanship or ideology; rather, it is likely better correlated with positions on certain political issues than either of those are.

While I see your point it's important to note that  the finding is that the more someone supports Trump the more supportive they are of the policy if it's associated with helping a white person rather than a black person. You can of course reformulate this single result as "the more someone opposes Trump the more supportive they are of the policy if it's associated with helping a black person rather than a white person"

There is no way from the study to distinguish between the explanation "people who support Trump dislike black people" versus "people who oppose Trump dislike white people" and so on. It's also not exactly implying that people support the policy in a racist way, since it's subconscious priming.

     I am not reformulating that result, but rather referencing another finding from the study that the Vox article references. Quoting from the Discussion and Conclusions section of this study:

You must be logged in to read this quote.

     As you point out quite correctly, the study cannot accurately make claims as to why these discrepancies happen or how people think about issues. The observations that opposite race-based discrepancies exist on both ends of the support spectrum is what makes me think that it might be capturing conditioning arising from attitudes towards certain political issues (particularly a variety of issues surrounding race relations that have gained prominence in recent years) as well as more conventional subconscious racism. That is only a hypothesis, however.

[Associate Justice PiT]

    Anyway, I notice from the study that people opposed to Trump were more likely to support the program when presented with the version with the black person (this is also in contrast with the ideal situation, wherein government programs are regarded in a purely race-blind fashion). This makes me wonder if these reactions on both sides are related to the politicization of race and the poor state of race relations in recent years. The study notes that support for Trump captures this far more effectively than partisanship or ideology and thereby dismisses the notion, though I would posit that it is a mistake to think of support for Trump as a dimension unlike partisanship or ideology; rather, it is likely better correlated with positions on certain political issues than either of those are.

While I see your point it's important to note that  the finding is that the more someone supports Trump the more supportive they are of the policy if it's associated with helping a white person rather than a black person. You can of course reformulate this single result as "the more someone opposes Trump the more supportive they are of the policy if it's associated with helping a black person rather than a white person"

There is no way from the study to distinguish between the explanation "people who support Trump dislike black people" versus "people who oppose Trump dislike white people" and so on. It's also not exactly implying that people support the policy in a racist way, since it's subconscious priming.

     I am not reformulating that result, but rather referencing another finding from the study that the Vox article references. Quoting from the Discussion and Conclusions section of this study:

You must be logged in to read this quote.

     As you point out quite correctly, the study cannot accurately make claims as to why these discrepancies happen or how people think about issues. The observations that opposite race-based discrepancies exist on both ends of the support spectrum is what makes me think that it might be capturing conditioning arising from attitudes towards certain political issues (particularly a variety of issues surrounding race relations that have gained prominence in recent years) as well as more conventional subconscious racism. That is only a hypothesis, however.

I did get what you mean and I thought I had made an explicit mention of it. Must have forgotten to put that in, sorry for the unnecessary confusion. I was trying to clarify the difference between your point and the thing I was saying for the benefit of other people since I was initially confused by that myself. Didn't work out so well.

     Oh, not a problem. When a news article references a study I try to at least skim the study, since journalists obviously aren't academics and don't have the best credentials to speak to the work (of course I don't either). I mistakenly thought the detail I was speaking to was more buried than it was because I stopped reading the article and just went to the study.