It is just partisan cognitive pollution served up to look like it is sort of sciencey. But it is a wonderful example of how to studiously ignore context in order to buttress an ideological position and also avoid having to confront unpleasant alternative interpretations.
For example:
The question is whether this reflects truly held, empirically based views (Pew) or negotiating positions. Whites view things as much better than blacks. On the other hand, if you take the perspective of negotiating positions between group identities, whites are arguing "Things aren't so bad. You don't need reparations." whereas blacks are arguing "Things are really bad." So which is it - real views or negotiating positions?
Then there is the partisan sniping - Pew gives the public the opportunity to score points off of Trump by asking whether Trump has made race relations worse. You read that 56% think he has made race relations worse and that sounds dreadful. But then you do a reality check by going over to Gallup and see what the trend has been for the past couple of decades.
What does this tell you?
For fifteen years 70% or so of the population, white and black, thought race relations were somewhat good or very good. But early in the second term of Obama, those numbers plunge to something like 50% or so. And this is way before Trump was on the nomination trail. Blacks have been registering a steady and ongoing decline in race relations since 2013. Whites stabilized at the end of the Obama administration and it has remained stable since then.
While it has been a staple on the left that Trump is a racist, there is little empirical evidence to support that position. And if you want to speak about what has caused a perceived decline in race relations, it has nothing to do with Trump and everything to do with Obama.
But that is not the conversation the mainstream media wants to have.
It is interesting that Pew, from what I can tell from a quick scan, never raises, much less addresses, the systemic skew in their research. Since blacks tend to vote monolithically for Democrats (85-90% though Trump is trying to change that), in many respects, in every question about Trump, Pew is confusing race and party. Race is just a proxy for party. This is a contextual issue that needed to be addressed in the construction of the sampling population and size in order to draw the conclusions they wish to draw.
If Pew wanted to deal with this rigorously, they needed to sample an equal sized population of liberal/Democratic blacks and conservative/Republican blacks. To stretch the argument, if all blacks are Democrat and all Democrats believe Trump to be a racist, then all you are saying is that all blacks are voting a party line. My suspicion is that the 5-10% of conservative/Republican blacks have a markedly different viewpoint of life under Trump than the 80-90% of liberal/Democratic blacks. And again, this has nothing to do with Trump, it started under Obama.
But Pew apparently wasn't going for accurate polling, they were going for ideological polling.
Or rather, critical theory/critical race theory as the following illustrates.
For some, this illustrates white privilege. Whites have the luxury of never having to consider their racial identity because they are the majority of the population.
But for those who subscribe to John Roberts perspective,
The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.this illustrates that the racists, i.e. people who view the world through the lens of race, are minorities given how much they define themselves by race.
For the logical empiricists of the world, Pew seems to be the one practicing bias by adopting the postmodernist/critical theory framing of their questions rather than the traditional majority perspective of human universalism.
Pew is finding racism by defining it into existence. Some sort of sociological mutant of Heisenberg's uncertainty principle.
Even though I am insulting Pew for its deliberate or unconscious biased framing, there are some interesting aspects to the survey. This, in particular.
Rather than dwell on the surface elements, I want to consider this chart given what we know about statistics, Jussim's research on stereotype accuracy, and normal distributions.
The first point is the relative probability, given the normal distribution of many traits, within any individual identity for any other attribute.
Pew is focusing everything through the lens of race, but the questions could be applied through multiple other identities/attributes such as
IQSome generic renditions of Pew's approach making it attribute blind.
Education Attainment
Unskilled worker, blue collar, white collar
Income
Hours worked per week
Height
Weight
Physical attractiveness
Religion
Sex
Familial status
Social Economic Status
etc.
Most adults have +/- views of the country's progress on XNow translate these using alternate valid identities to begin to see that these surveys are not quite telling us what we think they are saying.
People with attribute non-X are more likely to see advantages in being X
Most say it is now more common people to express Xist or insensitive views.
About half of people with attribute X say being X has hurt their ability to get ahead.
X are far more likely than non-X to say discrimination is a major obstacle for X people.
X and non-X differ widely in views of how non-X are treated.
X are more likely than other groups to see their X as central to their identity.
Most high education adults have negative views of the country's progress on educationIn other words, the truth of the proposition has nothing to do with the opinion of the individuals with the particular trait being emphasized. This goes to directly to Jussim's stereotype accuracy issue.
People with low education achievement are more likely to see advantages in being highly educated.
Most religious people say it is now more common for people to express anti-religious prejudice or insensitive views.
About half of people who are overweight say being overweight has hurt their ability to get ahead.
Low labor force participation rate people are far more likely than high labor force participation rate people to say discrimination is a major obstacle for low labor force participation rate people people.
Men and women differ widely in views of how women are treated.
High education people are more likely than other groups to see their high education as central to their identity.
In addition, there is a need to standardize what is being compared in order to achieve apples to apples comparisons.
For example the black high school graduation rate is 69% while that of whites is 86%. In order to assess some aspects of the race dimension, we need to normalize the base of comparison. What happens to "About half of blacks say being black has hurt their ability to get ahead" when we normalize so we are comparing like to like. We need to compare blacks and white by education attainment level in order to isolate what might possibly be discrimination versus what might be an artifact of poor education.
We know that high school graduation is one of the key strategies for life success. Without a HS diploma, things are much harder. Of course the perception of discrimination is likely to be higher in the population with only 69% graduation versus the population with 86% graduation. Because things are harder and it is more reassuring to attribute those hard =outcomes to discrimination than to not having worked hard enough in school. You have to normalize.
Without normalization, you end up misinterpreting root causes and trying to solve problems that are not the real causal factor.
Finally there is the interesting angle of distribution curves. Even within a monolithic race group or religious or other, there will be perceived bias and discrimination.
You ask a white person with low income and no HS diploma whether people feel increasingly empowered to say cruel and insensitive things about them and of course the answer is yes. It has become a biannual rite of electoral passage to insult and demean those less well educated.
My overall point is that this Pew study is not designed to answer the interesting questions but it is designed to obtain the clickable answers.
Cognitive pollution of the ideological kind. Especially trying to tie the decline in racial civility to the current president when it is broadly known and recognized that this occurred under Obama. Pathetic.
Posts
No comments:
Post a Comment