I am a market disciple and largely indifferent to outcomes (“disparities”). I have no dedicated belief that particular wage distributions should result from the market.
I take issue with your notion that we should adopt a universal and discriminatory policy solely in order to prevent misguided charges of discrimination from groups who score lower on objective scales of qualification. Aside from being kind of Orwellian (let’s refute stupid charges of non-existent discrimination by rewarding them!), the “no alternative” and “greater good of diversity” (why is diversity per se a good, again???) approach ignores the option (conveniently consistent with the truth) of just saying – you know, one test for everyone is the American way.
We first need to determine if the test is both unbiased and correlates to an ability to successfully perform the functions of the job.
If the test is biased, change the test. If it doesn’t correlate to an ability to successfully perform the functions of the job, then stop making applicants take that test.
But if both of the above are true, the problem isn’t the test and the solution isn’t to lower the standards just so black people can get on the force. Equality shouldn’t be a higher standard than getting a skilled workforce.
Doesn’t the importance of a diverse force outweigh the discrepancy of a few grade points, though?
I’m not talking about appearances, but effectiveness.
I remember when the police forces here began taking women, the requirements had to be changed (height, weight etc.) and there was much wringing of hands about the change, like you see here. But in the end the force was better for the addition of women, as was the community. And, for all the moaning and complaining about reverse discrimination, etc., the force did not fall apart, policing of the community did not suffer and now it’s a total non issue.
People like to make much of nothing, in my experience.
(a) baseline, I don’t care what race my doctor is; but
(b) under an aggressive AA regime, I will worry that a black doctor might be there for reasons other than medical excellence (same could apply to a white doctor cet. par., but AA does not inspire confidence in meritocratic advancement in general).
Curiously, the Department of Justice Civil Rights Division has just released a report concerning the disparate numbers of deadly force against African Americans by the New Orleans Police, but then goes on to criticise the lax hiring standards:
Under current guidelines, the problem is that disparate impact can be construed as prima facie evidence of race-based bias. Since society seems to have generally accepted that there cannot possibly be any genetic differences among race groups, an examination which yields disparate results by race when the test-takers are provided with equal opportunity to study must therefore a biased examination.
This is the thing that got the Ricci case going. The employers were damned if they did adjust for a “biased” test (reverse discrimination) and damned if they did not (regular discrimination). The “proof” that the test was biased was the disparate outcome alone. No further proof of how a particular question could be racially biased was ever offered.
As to how much a standard exam correlates, that’s easier to argue for than to put into practice. Better test scores may be necessary, but not sufficient, to pick the best candidates. In practice if one takes the extremes, almost any job that has a reasonable entrance screening exam will–on average–end up with a higher grade of workers from those who aced the exam versus those who were at the bottom.