a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by wasoxygen
wasoxygen  ·  3201 days ago  ·  link  ·    ·  parent  ·  post: Half of American whites see no racism around them

I believe that racism is a real problem in the United States today. It seems clear to me that people have different outcomes in life depending on race; black people in particular have a lower standard of living than white people (though the "black" and "white" labels make me uncomfortable).

Why do I believe this?

Well, because I have heard it repeated many, many times. Also, it is clear that racism was an institutionalized reality in the past: Article 1, Section 2, Paragraph 3 of the Constitution records that for purposes of democracy, ten black people counted as much as six white people.

Neither of these are good reasons to believe that racism affects outcomes today. Popularity of an idea does not make it true, and there has been change in law and custom reflecting the more enlightened values of our time.

It makes me uncomfortable to think that I don't have good evidence for my beliefs, so I request some help to ground my ideas more firmly with evidence. Hubski is a place where I can ask "What's so bad about slavery?" and get a thoughtful response, though b_b identified and then avoided the central question needed to settle that issue.

I found myself enjoying the way a new Hubski user talked about banking. Seemed like a sensible person. Then b_b linked to a comment in which the user questioned whether racism effectively hurts minorities, now that racism is widely shunned. This sentiment alone was enough for me to completely give up on that user as a credible source.

A friend suggested that I may have lost faith too soon; that "for all practical purposes" racism in this country has been beaten. I was rather taken aback, and mentioned that I felt I had an embarassingly large number of convincing arguments to the contrary, and that "sentencing disparity would make the point as clearly and indisputably as any."

My friend said "Judges are highly educated and cosmopolitan; they are unlikely to be racially biased."

This struck me as the purest naiveté. Judges are human beings. I read Kahneman. People ooze with bias, and the most pernicious discrimination can happen while the biased person is unaware. I offered vaguely-remembered research conclusions that judges give harsher sentences before lunch than after, on Mondays rather than Fridays. Who knows, maybe the defendant's smell affects judgements; everyone knows you had better dress sharp when you appear in court.

My friend characterized my laundry list of biases as "a little bit of an own goal for you." If we accept that judges are affected by these biases, and that they are all unreasonable and unjustified, "in order to prove racial bias you must control for all of these other biases." My friend added prior offenses, apparent contrition at trial, whether or not victim impact statements were given, the state where the offense occurred, whether mandatory sentencing guidelines were in place at the time of the offense, et multa cetera as additional factors that require control in any study that concludes racial bias affected judgements.

One more factor: income, which enables a defendant to get better representation which can lead to a lighter sentence. This seems like one that would be easy to measure.

I believe that being black in the United States is correlated to having lower income. If we find that black people are given harsher sentences for the same crimes, we can't be sure it was race or income that led to that outcome without examining evidence. I did a little research, but I couldn't find any studies of sentencing disparity that controlled for income.

There are many other arguments to demonstrate the reality of racism. I would like to come to a conclusion about this one specific question before considering others. It was clear in the GMO story that people were reluctant to examine evidence contrary to their beliefs, preferring to change the subject. "So the scientists found that GM corn didn't clearly cause cancer." "But what about all the pesticides?" "Well, the GM crop requires less pesticide, because it is already resistant to the parasite. That's why they modified it." "But the seeds are sterile!" "They're not, in fact." "But it's not natural!"

If we find that low income, rather than race, is the more important factor in explaining sentencing disparity, that does not mean there is no problem, of course. It does mean that we could direct efforts at reform more effectively, by trying to help low-income defendents of all races.

So here is my question: If you believe that race is an important factor in sentencing, can you find evidence that excludes other explanations, especially low income?

If you do not believe that race is an important factor in sentencing, what other area would you suggest investigating to find evidence of racism?





b_b  ·  3201 days ago  ·  link  ·  

    So here is my question: If you believe that race is an important factor in sentencing, can you find evidence that excludes other explanations, especially low income?

Yes.

(And of course this is just sentencing. It says nothing of the bias in arrest rates to begin with.)

wasoxygen  ·  3196 days ago  ·  link  ·  

Thank you, this was a careful and interesting study. This is the conclusion I draw from it:

Some of 70 judges examined in Cook County, Illinois, showed statistically significant variation in incarceration rates, although not in sentence lengths, for non-drug crimes, but not drug crimes, and this variation was associated with race, or some factor that correlates with race, and it is not clear if the variation favors whites over African American defendants, the reverse, or a mixture of both.

Section 2, "Literature Review," gives an overview of the "great deal of scholarship investigating the role of race in the courtroom" and describes many problems in earlier studies.

Section 6, "Potential Confounding Factors and Analysis by Crime Category," recognizes that differential judicial treatment could be explained by factors that correlate with race.

    African Americans may commit different crimes than whites, and judges may have different sentencing policies for different crimes. For example, suppose that some judges are stricter on sentencing for violent crimes than they are for other crimes. Suppose also that African Americans commit more violent crimes. This correlation would then lead to the appearance of heterogeneity in racial gaps in sentencing even if judges were race blind.

Other factors not considered (the application of DNA testing is mentioned) "could in principle generate the type of variation we observe if these unobserved features vary systematically across racial groups and judges differ in their treatment of these characteristics." I wonder if education level could be such a factor. If defendants of one race more often appear uneducated, a biased judge might be inclined to "teach them a lesson," whereas a more articulate defendant simply "made a mistake."

With enough data, it would be possible to test for these factors. This study examined over 600,000 felony cases, but after applying strict rules on what cases would provide good evidence, only 34,227 cases remained. When these are considered separately by type of crime, "each category contains a relatively small number of observations. In performing this analysis (data not shown), we find no evidence for excess heterogeneity in racial gap in any crime category. This result is almost certainly due to a lack of power."

The authors, following the important rule that you must not fool yourself, required a high standard of evidence, and the conclusions they were able to justify are correspondingly weak. Race, or something that correlates with race, makes a difference to some judges in some cases.

    While these confounding factors are still potentially a concern, the approach in this paper advances the field in light of previous work, because now the unobservable case characteristics would have to be correlated with the defendant’s race and elicit differential treatment across judges.

As usual, more research is advised.

I imagined that sentencing disparity would be an obvious place to look for evidence of racism, but in my view this evidence is not very persuasive. Do you have any suggestions as to where else we might look?

wasoxygen  ·  3196 days ago  ·  link  ·  

As an aside, b_b, the link you sent me is a good example of the importance of the lesson that "you must not fool yourself—and you are the easiest person to fool."

I didn't read the article until now, and I see it both amplifies the conclusions of the study and diminishes the reservations. The researchers have "demonstrated conclusively" that "racial bias affects judicial sentencing;" they "offer statistical proof" that judges vary in their treatment of race.

When the researchers say "We find evidence ... providing support for the model ..." the copywriter translates this to "race is affecting sentencing decisions" in the same paragraph.

Most egregiously, "the technique eliminates the problem of 'unobserved variables,'" which the researchers took pains to point out as a problem with their study.

Many people will read the press release and not look closely at the study. It's always a good idea to cut out the middleman!

b_b  ·  3196 days ago  ·  link  ·  

Interestingly, the data only say that racial bias exists, but not whom it is working against (or for).

However, we do have evidence that black people were targeted for sentencing disparity in mandatory minimum sentencing laws. The Supreme Court has ruled against these laws (specifically harsher sentences for crack than powder cocaine), but I don't think we're completely rid of them.

wasoxygen  ·  3201 days ago  ·  link  ·  

Thanks! Sending to Kindle now.