Wednesday, April 23, 2003 :::
Let me start by conceding that Kate Malcolm is absolutely right when she says that the antidote to spun numbers is "push the numbers further". That allowed, a (possibly) theoretical question for her:
Suppose black students, for whatever combination of reasons, are being lost in the third grade. By the time they make it through college they are hopelessly (I mean that at least almost literally for our supposition here) worse prepared than white students (ignoring the technicolors as unnecessarily complicating). (Er, I use "technicolors" to mean people who aren't black or white. I hope that's clear.) There are no "diamonds in the rough" here; if they perform to the best of their abilities in whatever new environment, the white students will, as a group, outperform the black students as a group.
The question, then, is whether we let in a black student who will genuinely do worse than a white student we're passing over. (Clearly, whatever's happening in the third grade should be addressed; if we know exactly what to do, that solves this problem fifteen years from now. Consider that the question is asked of now, or that we don't know what to do in the third grade.)
My feeling is that continuing unfairness created by society should be halted, but that, for unfairness that was previously suffered, from a practical standpoint it doesn't matter whether it was a matter of society failing to educate the children or lowering their self-esteem, or whether it was nature endowing some (black and white) children with more intelligence than other (black and white) children. We make the most of what we have.
If unfairness is inflicted by individuals on individuals, restitution (as possible) and/or punishment is appropriate; individuals can be deterred. A society can not be deterred by itself, restraining itself from creating unfairness in fear of retribution by its future self. Compensation strikes us as morally valuable, but, if we stand back, makes no more sense for a caused tragedy than one of nature. (Steve, you have my David Friedman book; can you find me a quote in there, something about torts being a poor way to try to implement insurance?)
Following Kate's example, I'll reiterate that what we really need here, back in the real world, are more data. Are we losing black children in the third grade? If so, what do we do about it? (Something similar to what we should do about bullying?) Is that the only place the problem is taking place? Either way, though, my question is one that I think needs to be addressed. The most troubling thing to me about systematic variation of outcomes is not the variation of outcomes per se — some students will always do better than others, and I decline for the moment to be troubled even by a correlation, in and of itself, with race — but that does strongly suggest that there is something feasable that could be done to improve black students to the current white levels, and that we've missed opportunities to raise acheivement does bother me.
::: posted by dWj at 11:59 AM
