Affirmative Action in the US: is it still needed?

google is your friend

at least pretend you're interested enough to look into it

well now we know you didn't follow any of the links I posted

For a test to be regarded as "unbiased" or "fair", differential item functioning should not take place. Mean scores on the test's items should be alike for subgroups of interests who are alike with respect to overall ability on the construct of interest.

I don't know if subsequent Supreme Court decisions addressed the issue.

But they are both standards, however imperfect, that do not discriminate by race and gender. No standard is ever going to be perfect, and if universities find better standards than those, good for them! For purposes of this discussion, I'll settle for the best standard universities can find that doesn't discriminate by race gender, and so forth.

It depends. Why did the minority applicants fail the exam?

Well, if that's what you'll settle for, then a rock-paper-scissors contest would work just as well.

Thomas wrote:

It depends. Why did the minority applicants fail the exam?

Why does that matter? It was the same test for everyone, just like the SAT. Being a conscientious mayor, however, you can be confident that the test itself is not biased.

If the test fails applicants who would make good police officers, it's a bullshit test, regardless of the applicant's race, just as rock-paper-scissors would be.

But if it's a valid, no-bullshit test, the minority applicants' failure would indicate that they wouldn't make for good police officers. I would refrain from hiring bad police officers just because they belong to minorities.

That's why it makes a difference, and that's why my first question would be: How did the police officers fail the test? What exactly did they fail at?

Why so? The police department doesn't have an unlimited number of jobs available. If there are 20 slots to fill, then wouldn't it be fair to take the 20 top-scoring applicants on the exam, even if the top 50 would have made good officers? After all, that's how you'd handle college admissions.

I see. It's not that you support tests, you just support good tests.

For instance, if candidate A fails the test to become a police officer, how would you determine if he would nevertheless have made a good police officer?

I'm still not seeing why that would make a difference. You can assume that thy failed because they did not fulfill the requirements for passing.

For instance, if candidate A fails the test to become a police officer, how would you determine if he would nevertheless have made a good police officer?

This apparent mismatch between what the city tests for and what the city needs is what gives me probable course to search for flaws in the test. Not unfairness to the applicants.

Yes I do. Am I sensing a tone of sarcasm in your voice here, or is that just me?

joefromchicago wrote:

For instance, if candidate A fails the test to become a police officer, how would you determine if he would nevertheless have made a good police officer?

I don't have an algorithm for answering that question in the abstract and in all generality. If you have a specific case in mind, now would be a helpful time to talk about the specifics. What was the test, and what did the minority applicants fail at?

I wouldn't. I would look around for better tests

But how do you determine if the test is weeding out good applicants if the only means you have to measure a candidate's goodness is the test? You need something else -- a "meta-test" -- to test the test.

Thomas wrote:

I wouldn't. I would look around for better tests

Let me get this straight: you would replace the current test with a test that allows more minority applicants to pass? How is that not affirmative action?

Because the policy objective is better law enforcement for the city's residents, not fairer job opportunities for the city's would-be police officers. If I could provide the citizens better law enforcement with an all-White police force, I would. But that's not the scenario you stipulated --- realistically so, I think --- and I'm running with your stipulated facts.

Just because a broken clock happens to be right twice a day, that doesn't make it an accurate clock. And just because good policy happens to meet the demands of affirmative-action advocates every now and then, that doesn't make it affirmative action. You know that. You've been giving Occom Bill the correlation-is-not-causation speech for years.

I'm astounded.

But from the point of view of a disappointed white applicant, who would have been hired under the old test, it makes very little difference whether your goal is to increase minority representation on the police force or to provide citizens with better law enforcement. To him, you just substituted a test that was fair with one that wasn't.

This isn't an instance of correlation-versus-causation. When universities take student diversity into consideration when making admission decisions, you oppose that. But when a city takes crime-prevention into consideration -- and that indirectly involves considerations of race -- then you're in favor of that. But there's not a whole lot of difference between the two situations.

Why does calling something "affirmative action" make a difference if, in the end, it achieves the same result?

It's a good thing, then, that I don't care about the White applicant's disappointment, just as I don't care about the Black applicant's disappointment with falling short on the SAT test. If that coincides with what an equal-opportunity disappointer would do, I don't care about that either.

The difference is that in the scenario you're describing, minority police officers are in a better position to win trust in the neighborhoods it's policing. That has direct consequences for how well they're doing their job. But in universities, in general, there is no such benefit to studying. Being white rather than Asian, or Black rather than White, will not make you a better mathematician, scientist, or engineer. There might be exceptions around the humanities, but as a general rule, I see no benefit comparable to the police-force scenario as you described it.

Maybe you should care. After all, racial discrimination is illegal.

But then if your only objection to affirmative action is that it shouldn't be used where it does no good, then you're not really opposed to affirmative action per se, you're just opposed to it in certain situations.

But in universities, in general, there is no such benefit to studying. Being white

But what Freedle found in 2003 has now been confirmed independently by the new study: that some kinds of verbal questions have a DIF for black and white students. On some of the easier verbal questions, the two studies found that a DIF favored white students. On some of the most difficult verbal questions, the DIF favored black students. Freedle's theory about why this would be the case was that easier questions are likely reflected in the cultural expressions that are used commonly in the dominant (white) society, so white students have an edge based not on education or study skills or aptitude, but because they are most likely growing up around white people. The more difficult words are more likely to be learned, not just absorbed.

While the studies found gains for both black and white students on parts of the SAT, the white advantage is larger such that the studies suggest scores for black students are being held down by the way the test is scored and that a shift to favor the more difficult questions would benefit black test-takers.

On average, he said, black students were performing only slightly above matched-ability whites on hard questions. But averages did not submit applications to colleges. Individual students did. Some of those individuals, he discovered, would have gotten a boost of a hundred points or more on the SAT if the score was weighted toward the hard items. He proposed that the College Board offer a supplement to SAT scores, called the Revised-SAT, or R-SAT, which would be calculated based only on the hard items. This, he said, would "greatly increase the number of high-scoring minority individuals."

In their paper, Santelices and Wilson rule out Freedle's suggestion that the bias he found in the test might affect all kinds of multiple-choice questions, or minorities other than blacks. But they did find it in sentence completion and reading comprehension sections of the SAT.

Saul Geiser was the director of research in Atkinson's office originally given the assignment to look into Freedle's theory. Eventually he arranged for Santelices, then a doctoral candidate at UC Berkeley, to do the research as her PhD thesis, working with Wilson, a UC Berkeley psychometrician who had also been asked to look at Freedle's work.

Geiser said he thinks the two researchers did a good job. He does not agree with Bunin's criticisms of their work. He said he, like Freedle, wants more more research on why blacks and whites answer these questions differently, so that any unfair disadvantages for blacks can be removed.

He said he thought the College Board, in particular, should "get over the denial" of any merit to what Freedle has discovered. That may take a while. The College Board, after all, may be right that the SAT is unbiased.

But the new paper means more researchers are likely to go more deeply into what Freedle has found, and eventually settle the question of what should be done about it.

On some of the easier verbal questions, the two studies found that a DIF favored white students. On some of the most difficult verbal questions, the DIF favored black students.