Where am I leaping at the DOJ? I am pointing out that the main stat being offered in the media reports and most here as smoking gun evidence of widespread racism is meaningless and uninterpretable without factoring the many well established covariates with race that are also well known to impact (and should impact) the probability of a search.
How in the world can you write
(I spelled it write this time... oh damn it!) off 16,000 of 21,000 with outstanding arrest warrants? You'd expect numbers like that in a prison.
Please clarify what and whom each of those numbers refer. Also, keep in mind that blacks have 2-5 times the national arrest and conviction rates for most crimes, and evidence from victimization reports confirm that these largely reflect actual differences in engaging in criminal acts, not just being more likely to be caught. So, whatever your warrants stats refer to, you need to show that it is not in line with what is expected from national stats on criminality.
This exact same stat was put forth in the main "Ferguson" thread, and shown to be meaningless then to most of the same folks parroting it yet again now.
What is odd is that the same people leaping to parrot these meaningless stats as "evidence" of racism are the first to completely discount similar types of stats when they seem to work against their pet theory.
Actually, there is a lot of talk about the initial reports of the DoJ report, which more than likely looked over more than that single statistic allegedly being parroted.
My comments have been directed at the frequent referenced to search and arrest disparities as in themselves evidence of racism. They are not. The emails and other evidence of actual racism, do make those statistical disparities any more interpretable as racism, because they remain just what is predicted whether or not the searching cops were racist.
IF you take a shower every single day, then on the day that your neighbor gets killed, is the fact that you took a shower evidence that you killed him? No.
IF I find blood on your clothes, that is evidence. However, even then, is that same shower you took now evidence that you killed him? No.
You would have taken a shower whether or not you killed him, and the added fact of the knife has no impact on that probability.
Uninterpretable correlations with countless confounds do not magically become evidence for a specific causal claim, just because unrelated facts support that cause. The correlation remains uninterpretable until some of those alternative factors can be ruled out.
So far, I haven't seen any evidence addressing those confounds, namely are blacks without warrants or records more likely to be searched than whites without warrants or records?
The DOJ has that data and unless they are pathetically incompetent know that such results are far better evidence of racism than what is being reported. Can you or anyone touting this report as valid evidence show me where they did that very simple analysis, controlling for prior record and warrants?
IF they don't include that analyses, then either they are so intellectually incompetent that the whole report is toilet paper, or they are deliberately suppressing those analyses because they do not support the narrative they wanted to tell. I wouldn't know whether to bet on gross incompetence or corruption with the DOJ, but it would have to be at least one of those.