The Monkey Cage

1. Re. intelligence: IQ research shows that whites test higher than blacks. Thus the "stereotype" is accurate.

2. Re. hardworking: Blacks have higher rates of unemployment than whites, so they almost certainly work fewer hours per week on average. Thus the "stereotype" is accurate.

3. Re. trustworthy: Blacks have much higher rates of crime than whites, so one could infer that they are less trustworthy. Thus the "stereotype" is accurate.

So what's the story here? That Americans are well informed about group differences?

The funny thing is . . . if we're going with stereotypes, isn't the stereotype that Hispanics are harder-working than whites, willing to take the jobs we won't take, etc?

At the risk of pointing out the obvious BB, I'm not sure that your arguments hold water if IQ tests are potentially race biased and the other comparisons do not stand up to basic controls for economic status.

More generally, this reminds me of one of my favorite findings in political science. In the article "The New Racism" by Paul Sniderman et al., they found that when conservatives and liberals had to consider giving government aid to a hard working African-American, the conservatives were *more* likely to give aid than the liberals. The conjecture was that they didn't believe such a person existed.

"I'm not sure that your arguments hold water if IQ tests are potentially race biased"

What exactly do you imagine is on an IQ test? Cultural trivia?

"and the other comparisons do not stand up to basic controls for economic status."

The survey didn't ask respondents to control for economic status. Thus, they answered appropriately.

"What exactly do you imagine is on an IQ test? Cultural trivia?"

I'm not a psychologist and therefore not an expert on IQ tests, but I'm under the impression there is a rather large literature on this exact subject. You'd have to ask them what they mean by cultural biases.

"The survey didn't ask respondents to control for economic status. Thus, they answered appropriately."

I wasn't commenting on the respondents, but on your suggestion that these are somehow true objective facts -- blacks are lazier than whites, etc. Simply put, you cannot easily disentangle the effects of poverty from the effects of race. You may observe blacks have higher unemployment rates, but that says next to nothing about the role of race in producing employment. The "'stereotype' is accurate" only if you live in a bivariate, unidimensional world. Most of us don't.

If you do want to talk about the survey respondents, it at least minimally suggests that they are willing to use race to characterize people over more powerful predictors like class. Besides that, its not a true-false knowledge question anyway, so you can't "answer correctly."

"You may observe blacks have higher unemployment rates, but that says next to nothing about the role of race in producing employment. The "'stereotype' is accurate" only if you live in a bivariate, unidimensional world. Most of us don't."

Your argument is completely spurious. The survey doesn't ask about causality. Nor have I implied anything about causality when stating that "the stereotype is accurate." The only "objective reality" at stake here is correlative, not causal.

Why does this matter? Because Medvic makes the following inference from his data: "These are simply astonishing numbers and indicate significant (and probably deep-seated) racist stereotyping." That inference is unsupportable. To conclude that the results are due to racism, he would have to show that the survey respondent's beliefs are false. He hasn't done that.

It's bad social science. Period.

just because a test says it measures intelligence does not actually mean it measures intelligence. First, what is intelligence? is it a single definable attribute, a grouping of somewhat similar attributes, a large number of not particularly related attributes?

the original IQ tests were designed to identify students that weren't well served by the type of education they were getting so that these students could receive alternate education. Their use to rank people was a primarily American invention led by Goddard and terman who believed that intelligence was a single, innate ability.

Further, saying unemployment is higher in one group only says the lower unemployment groups works more not necessarily harder.

BB,

The point is simply that being poor or unemployed does not mean that a person is "lazy". Blacks have higher poverty rates partly because of the residual effects of segregation and discrimination. A person can be extremely hardworking and still be poor and in need of government assistance.

Likewise, crime is related to poverty as well as institutional racism. People with money have lower conviction rates even when they are accused of the same crimes, and profiling can become a self-fulfilling prophesy by targeting people based on color. So a higher crime rate among poor (which happen to be disproportionately African American) citizens does not make blacks as a race inherently untrustworthy.

The stereotypes that blacks are lazy and untrustworthy emerges from a flawed logic that is primarily rooted in racist sentiments about the inferiority of blacks as a race. Citing an objective fact about the state of African American communities is one thing. Using those facts to infer supposed character flaws about the entire race in an attempt to make that race appear inherently (even genetically, in the case of IQ) inferior is racist. You seem to be confusing those two things.

To ADG:

Your strongest argument is about the "lazy" dimension, since that word encourages respondents to speculate about motive (which is opaque) rather than outcome (which is well known, as measured by unemployment). But for you to be right, it would have to be true that whites and blacks at the same income level have identical work ethics. I doubt this is true: consider, for example, the John McWhorter thesis about how black students consider working and studying hard to be "acting white" and therefore to be avoided. That'd be a clear cultural bias among US blacks in favor of "laziness."

Your argument about crime (re. the trustworthiness dimension of the survey) is 100% factually wrong: see http://www.city-journal.org/2008/18_2_criminal_justice_system.html

As I pointed out to Scott, I have made no claim about "inherent" or "genetic" tendencies, nor have I said anything about causality until this post (re. McWhorter). Your imputing racism to me is despicable.

To BillC:

Suppose you're right: suppose IQ tests only half of what "really" constitutes intelligence. So what? How does it change how people should have responded to the survey? Do you think that blacks are substantially better than whites on the untested dimension of intelligence, so that it all evens out? If so, what's the content of that untested dimension?

BB,

Wow, you cite a conservative magazine article from a publication that also discusses Obama's "post-masculine charisma" as some kind of serious citation for your claim. And you complain about "bad social science" in the meantime?

It's well-established that income and wealth is a major factor when it comes to conviction rates and incarceration rates. People who can afford top-notch lawyers win their cases more than those who rely on public defenders. I'm not sure what is controversial about this statement, but your City Journal op-ed did nothing to contradict that basic fact.

No BB, I'm saying

1. IQ tests do not test real intelligence in any meaningful way, so making a statement about them explaining 1/2 or 1/3 or any fraction of intelligence is pointless.

2. even if one grants some meaningfulness to whatever it is that IQ tests measure, the median or mean is a poor measure of any differences between test groups. Any difference determination without consideration of the deviations around that score is incorrect. if I remember correctly, there were no statistical differences between the populations

ADG:

Wow, nice comeback. I cite a well-researched article, and you respond with a) a guilt-by-association argument, and b) bald assertion unsupported by either facts or citation. Very convincing!

BillC:

Evidently you think IQ scores are completely random, uncorrelated with anything? You're just making things up as well. No evidence, no citation, no expertise.

BB:

Regarding IQ and Race, see

The Black-White Test Score Gap. Brooking Inst Press, 1998.

For a good review of the literature (most of it experimental) on the influence of the color of skin (as opposed to knowledge about "reality") in informing opinions, see

Mendelberg, Tali. 2008. "Racial Priming: Issues in Research Design and Interpretation." Perspectives on Politics. pp 135-140

See also:
Peffley, Mark, and Jon Hurwitz. 2007. "Persuasion and Resistance: Race and the Death Penalty in America." American Journal of Political Science.

One finding from the study is that whites (on average, of course) are more likely to support the use of the death penalty when the race of the suspect - which is experimentally controlled - is black. Nothing else varies, which is inconstsent with your claim that whites are only acting as well informed citizens.

I'm sure Sigelman can point you to more reading on this if you're really that interested.

JMA:

How do the Brookings book or Mendelberg article contradict anything I've said?

Your point from Peffley and Hurwitz is, as far as I can tell, equally irrelevant. I haven't claimed that no one is racist; rather, I've claimed that the survey responses are factually accurate and therefore don't permit any inferences about racism.

You're arguing entirely against strawmen.

BB:

I guess it depends on how the questions are worded. It's not clear in the post how they were. Most questions I'm familiar with either ask people to rate "most black/white people" or ask them to rate black/white people "in general". Given that most black people are not criminals, jobless, or morons, your arguments don't seem to work. That is, if most black people have jobs, why do 70%+ white people rate them as not hardworking and not trustworthy? The better question to get at your argument would ask "what percent of blacks/whites are hardworking/intelligent/trustworthy.

Just because a larger percentage of blacks (as opposed to whites) are in jail and jobless does not lead to the conclusion that most blacks or blacks in general are not trustworthy or hardworking. Right?