Study: Trump fans are much angrier about housing assistance when person black

[Spectral]

It's amazing what people wind up using as the takeaway from article. This seems like the punchline of the study to me:

We find no direct treatment effects of the white
vs. black racial cue on these dependent variables: policy opposition (b = 0.01, p>0.10), anger (b
= 0.01, p>0.10), and individual blame (b = -0.01, p>0.10).

So basically, they found nothing. Luckily, our intrepid researchers knew that there must be racism at play, so they kept working hard:

To establish that our experimental treatment did, however, polarize respondents on the
basis of racial attitudes, we estimated a series of models interacting treatment assignment with
racial resentment, ethnocentrism, and perceptions of white disadvantage (along with control
variables). These results show that our manipulation of race did lead to distinct reactions among
racial liberals and conservatives, as measured by multiple indicators of racial bias (see Appendix
Tables A2-A4).

Holy shit, look at the phrasing bolded there. Maybe it's just bad writing, but the way they wrote this up makes it look like they really did think, "well, what can we do to call them racists anyway?". Anyway, they were able to use sufficient statistical manipulation to find some completely unimpressive effect sizes that kinda-sorta support their claim that Trump supporters are racist.

To me, it just looks like another example of Andrew Gelman's garden of forking paths, with the additional twist that the authors were wildly enthusiastic about finding some path that arrives at their desired conclusion.

[Ahhdurr]

No one would have ever guessed that Dumpsters are racist. Unless you read their posts here.

[primalmatter]

No one would have ever guessed that Dumpsters are racist. Unless you read their posts here.

Least they are not in love with the concept of the "white man's burden."

[Lenonis]

Is there really anyone left who would deny that at least some of Trump's base is racist? I figured everyone, even Trump supporters, are down with that fact now given the irrefutable evidence.

[Grapemask]

It's amazing what people wind up using as the takeaway from article. This seems like the punchline of the study to me:

So basically, they found nothing. Luckily, our intrepid researchers knew that there must be racism at play, so they kept working hard:

Holy shit, look at the phrasing bolded there. Maybe it's just bad writing, but the way they wrote this up makes it look like they really did think, "well, what can we do to call them racists anyway?". Anyway, they were able to use sufficient statistical manipulation to find some completely unimpressive effect sizes that kinda-sorta support their claim that Trump supporters are racist.

To me, it just looks like another example of Andrew Gelman's garden of forking paths, with the additional twist that the authors were wildly enthusiastic about finding some path that arrives at their desired conclusion.

I'm not sure if you simply stopped reading after you felt your biased confirmed, but the rest of the study explains what that garbled heap of poor technical writing is about.

Data interpretation is complex, and it's explaining that there was not direct 1:1 correspondence between only a racial cue and these feelings (that is, if they only considered racial cues and no other factors, the data could not be parsed to reach any conclusion). This was how they created one of their bases for their study. They then ran paired comparisons, connecting racial cues with other factors (including Clinton support, as well as a dozen other variables), but found that the pairing of (Trump Favorability x Racial Cue) created responses in the data set significantly above and beyond other pairings.

[Spectral]

I'm not sure if you simply stopped reading after you felt your biased confirmed, but the rest of the study explains what that garbled heap of poor technical writing is about.

Data interpretation is complex, and it's explaining that there was not direct 1:1 correspondence between only a racial cue and these feelings (that is, if they only considered racial cues and no other factors, the data could not be parsed to reach any conclusion). This was how they created one of their bases for their study. They then ran paired comparisons, connecting racial cues with other factors (including Clinton support, as well as a dozen other variables), but found that the pairing of (Trump Favorability x Racial Cue) created responses in the data set significantly above and beyond other pairings.

You get that this just isn't actual hypothesis-driven research though, right? They had a hypothesis, it was wrong, so they just kept data-mining until they found a way to torture their data to have a snippet that sorta corresponded to their original idea. It's a shitty piece of work done by ideologues that know enough about statistics to be dangerous, but not enough to actually do science.

Post hoc model creation is a pretty bad sin when it comes to abuse of statistics.

[Garnier Fructis]

It's amazing what people wind up using as the takeaway from article. This seems like the punchline of the study to me:

So basically, they found nothing. Luckily, our intrepid researchers knew that there must be racism at play, so they kept working hard:

Holy shit, look at the phrasing bolded there. Maybe it's just bad writing, but the way they wrote this up makes it look like they really did think, "well, what can we do to call them racists anyway?". Anyway, they were able to use sufficient statistical manipulation to find some completely unimpressive effect sizes that kinda-sorta support their claim that Trump supporters are racist.

To me, it just looks like another example of Andrew Gelman's garden of forking paths, with the additional twist that the authors were wildly enthusiastic about finding some path that arrives at their desired conclusion.

What I took this to mean is that they basically just 'rediscovered' implicit bias. Which is something everyone has to a degree, whether they're actually racist or not. So not exactly a meaningful discovery.

[Cthulhu 2020]

What I took this to mean is that they basically just 'rediscovered' implicit bias. Which is something everyone has to a degree, whether they're actually racist or not. So not exactly a meaningful discovery.

Nobody should be at all surprised that Trump voters get angry at the idea of black people receiving housing assistance.

And yet some still are, and some are even aghast at the suggestion it seems.

[Deleted]

Is there really anyone left who would deny that at least some of Trump's base is racist? I figured everyone, even Trump supporters, are down with that fact now given the irrefutable evidence.

Meh we still have people believing god created Earth in 7 days so....

[Spectral]

What I took this to mean is that they basically just 'rediscovered' implicit bias. Which is something everyone has to a degree, whether they're actually racist or not. So not exactly a meaningful discovery.

I think it's important to keep in mind that the work done on implicit bias isn't very good either - it has regularly struggled with replication issues, quality issues, and isn't actually predictive of behavior anyway. This NY Mag writeup is good, worth reading in whole, but one relevant excerpt:

A pile of scholarly work, some of it published in top psychology journals and most of it ignored by the media, suggests that the IAT falls far short of the quality-control standards normally expected of psychological instruments. The IAT, this research suggests, is a noisy, unreliable measure that correlates far too weakly with any real-world outcomes to be used to predict individuals’ behavior — even the test’s creators have now admitted as such. The history of the test suggests it was released to the public and excitedly publicized long before it had been fully validated in the rigorous, careful way normally demanded by the field of psychology. In fact, there’s a case to be made that Harvard shouldn’t be administering the test in its current form, in light of its shortcomings and its potential to mislead people about their own biases. There’s also a case to be made that the IAT went viral not for solid scientific reasons, but simply because it tells us such a simple, pat story about how racism works and can be fixed: that deep down, we’re all a little — or a lot — racist, and that if we measure and study this individual-level racism enough, progress toward equality will ensue.

Of course, people do have biases, but the social "science" on the matter is pretty low quality.

[Garnier Fructis]

Of course, people do have biases, but the social "science" on the matter is pretty low quality.

I don't dispute that. That people do have biases was really all I was going for. I've never been sold on the idea that this can predict anything meaningful.

[Rukh]

I don't dispute that. That people do have biases was really all I was going for. I've never been sold on the idea that this can predict anything meaningful.

He says in a thread about predicting something meaningful. :P

[Brubear]

I wonder what would happen if they did this with say, showing a black person? Or even more none-Trump supporters.

I mean I get it, Trump supporters are racist, we all know that. But from what we know of this study its horribly biased. It set out to prove a point and lo and behold, its point is proven because they took careful control of who they showed this stuff to.

From the Vox article.

"In contrast, favorability toward Clinton did not significantly change respondents’ views on any of these issues when primed with racial cues.

The researchers concluded, “These findings indicate that responses to the racial cue varied as a function of feelings about Donald Trump — but not feelings about Hillary Clinton — during the 2016 presidential election.”

A chart showing greater support for Trump led to more opposition for housing assistance programs, among other issues.
The researchers, which generally controlled for a host of socioeconomic and demographic variables, also tested to see whether measures of polarization or ideology changed the findings. They found little to no effect."

[Mormolyce]

I'm afraid the days of hiding this behind plausible deniability are over.

[Theodarzna]

From the Vox article.

"In contrast, favorability toward Clinton did not significantly change respondents’ views on any of these issues when primed with racial cues.

The researchers concluded, “These findings indicate that responses to the racial cue varied as a function of feelings about Donald Trump — but not feelings about Hillary Clinton — during the 2016 presidential election.”

A chart showing greater support for Trump led to more opposition for housing assistance programs, among other issues.
The researchers, which generally controlled for a host of socioeconomic and demographic variables, also tested to see whether measures of polarization or ideology changed the findings. They found little to no effect."

I'm still waiting if that study is replicated across many ethnic groups however. I.E. Black tax payers, Hispanic housing, Asian tax payers, black housing, Black tax payers, white housing.

[paralleluniverse]

It's amazing what people wind up using as the takeaway from article. This seems like the punchline of the study to me:

So basically, they found nothing. Luckily, our intrepid researchers knew that there must be racism at play, so they kept working hard:

Holy shit, look at the phrasing bolded there. Maybe it's just bad writing, but the way they wrote this up makes it look like they really did think, "well, what can we do to call them racists anyway?". Anyway, they were able to use sufficient statistical manipulation to find some completely unimpressive effect sizes that kinda-sorta support their claim that Trump supporters are racist.

To me, it just looks like another example of Andrew Gelman's garden of forking paths, with the additional twist that the authors were wildly enthusiastic about finding some path that arrives at their desired conclusion.

This post is bogus.

The quote that you've taken from the paper says that the racial cue alone has no statistically significant effect on support, anger, and blame.

However, the racial cue DOES have a statistically significant effect for Trump supporters as can be seen in Table 1, Page 9, which is what the article correctly reported.

In addition, this is not p-hacking. P-hacking involves a study without a pre-determined scope that measures hundreds of response variables in order to fluke out some statistically significant finding. This study has a pre-determined scope of looking at 3 response variables, support, anger and blame, and finds a statistically significant result for Trump supporters on all 3 variables.

The amount of statistically illiterate, erroneous spin here is embarrassing for you, and done to muddy the well-known fact that many Trump supporters are racists. Fail.

[jibberbox85]

Link to study article is discussing.
https://drive.google.com/file/d/0BzC...hQSW0tVDQ/view

https://www.vox.com/identities/2017/...porters-racist


Found this to be rather interesting. Going to read through the actual study when I have time, but reading over the conclusion it seems that it really is the person being black that's the deciding factor for Trump voters on whether a person is deserving of housing assistance.

The problem with this study is that it's rather bias, and doesn't cover the topic from all angles. With that in mind, most people across all spectrum's have some sort of bias that are due to environmental, and or news/social medial influences. This is nothing new, and to link it with racism makes this article even more so the reason why people are always looking over their backs.

[paralleluniverse]

Btw, findings similar to this study have been known for a long time:

The second paper, by the economists Alberto Alesina, Edward Glaeser, and Bruce Sacerdote, was titled “Why Doesn’t the United States Have a European-style Welfare State?” Its authors — who are not, by the way, especially liberal — explored a number of hypotheses, but eventually concluded that race is central, because in America programs that help the needy are all too often seen as programs that help Those People: “Within the United States, race is the single most important predictor of support for welfare. America’s troubled race relations are clearly a major reason for the absence of an American welfare state.”

Source: https://www.nytimes.com/2015/06/22/o...ng-shadow.html

[rda]

One caveat to the study: It’s not clear if the results are reflective of higher support due to a white racial cue or lower support due to the black racial cue. It’s possible that it’s not that Trump supporters are made less likely to support housing assistance when exposed to the image of a black man, but that they are made more likely to support housing assistance when exposed to the image of a white man. The researchers said that they will need further study to clarify this, along with questions of how racial cues affect views on other policy issues among Trump supporters.

I am going to translate this for you and others:

1. The researchers wanted to find racial bias amongst Trump voters.

2. They measured a couple of things, one of them seemed promising.

3. They wrote the paper and published it. "Peer-reviewed" it too and got congratz on job well done amongst the researches of the same kind.

4. Then someone pointed out to them that they did a shoddy and hasty job, and that this manifests in a number of ways. For example, even if the measurements have been carried correctly and the methodology is all correct bla bla bla bla bla, even then their interpretation is simply not the only one possible (!). What they thought was distaste for one group could be a preference for another group. (If you are at all close to the science, you understand that this is such a stupid rookie mistake that the "research" likely has many more.)

5. They hastily corrected the abstract to point out that their results might be bogus and washed their hands.

Seriously, I don't care much about Trump. But this is just BS. This isn't science, this is fishing for facts supporting the conclusion you want.

[PACOX]

if anything this is a response to the attack on white males i've witnessed in this country in the last 5 years or so. when did it become so fashionable to hate old rich white men? its like the coolest thing right now.

Maybe its blow back?

Its dumb to generalize an entire group like that for sure. People are fucking lazy though. I think we all know who they are talking about but criticism can do without the stereotyping.