The overall conclusion to me still stands:
Sotomayor will be confirmed.
Most likely.
I think that’s an understatement. The republicans don’t have the votes even if they were minded to make a stand on a well-qualified and relatively moderate (she’s nowhere near a liberal scalia) nominee.
Again: Has anyone actually SEEN this test? Has anyone done an analysis of the questions to show where racial bias might have played a part?
I ask because the numbers of people involved are just too small for the result itself to be a statistically meaningful indicator of bias. From this Slate article:
Three out of eight blacks and hispanics passed the first test, and 16 out of 25 whites passed. Yes, the white rate was higher than the black/hispanic rate, but with such a small sample size it’s not inconceivable that other factors than race accounted for this. If one more black had passed, they would have had a pass rate of 50%, vs 61% for the white candidates. Even even two more black candidates out of eight had passed, they would have had a higher pass rate than the white candidates.
It seems to me that in a case with such a small sample size, the only way to show bias would be to appeal to the test itself and examine the questions that were asked. Also important would be to know if the questions themselves were directly related to the job.
Finally, to show that the test was biased, I would want to see the results of the tests used to hire the firefighters in the first place. Was there a racial quota system in the original hiring? Were the standards different for black and white? What were the respective educational levels of the candidates?
Would it matter to anyone to know that the average high school grade of the black candidates was 68%, and the average of the white candidates was 80%? I’m not saying that that’s the case, as I have no idea. What I am saying is that racial discrimination is a hell of a lot more complex than simply looking at the raw numbers of people who pass a test - especially when the results aren’t that far out of whack and the sample size is quite small.
Unless they successfully filibuster.
And, as I said before, the democrats have the votes. Republicans cannot successfully filibuster without enough votes to survive cloture.
Al Franken will be seated soon (almost certainly before the floor vote on confirmation). Once he is seated, there will be 60 democratic (or independents who caucus with the Democrats) senators. 60 votes is enough for cloture even if no republican votes against the filibuster. Plus, I could name three or four senators (Snowe, Collins spring to mind) that would make up for any sudden defection.
So while yes, a successful filibuster would stop the nomination, the way the votes stand makes a successful filibuster extremely unlikely. From the statements I’ve read, even an attempt to filibuster is unlikely (which makes sense–it’s foolish to try something so desperate if it’s doomed to failure).
I stand by my initial statement.
Yes, I agree. She’s nowhere near a liberal who has a distinct, complete, consistent, honest, and well-considered legal theory which is reliably followed without veering off into or expressing whatever personal opinions she may have. 
Actually, I don’t think she’s a bad judge, just not a great thinker, either. Her politics mostly don’t interest me. I find some of her stupid ideas contemptible, but
Disoarate Impact cases don’t always require any kind of scientific validity, however. The plain fact is that few if any lawyers know statistical correlation from a hole in the ground. And the government bureaucrats who deal in race are known to ignore it when convenient.
But I, too, would like to see the test. With Disparate Impact, the tested material has to be as close to direct relevance as possible to pass under the law. Several companies, for example, have been successfully sued for DI because they administered “general skills” tests. Their minority applicants may have been uneducated or simply less educated, and therefore didn’t pass often enough. Redesigned tests which focused more on directly job-related factors are usually OK.
IIRC, didn’t the city (or perhaps another) get counter-sued after it passed several candidates who failed, leaving some (white, but also asian and hispanic) who passed the admissions test out?
Cite, S’il vous plait? Her decision in the New Haven case adheres to circuit precedent–so it’s no example.
But I suppose you do have examples of where her judging veered off into personal opinions? Or her “stupid ideas”–not in speeches, but in her work as a judge (if you say people can’t distinguish, then you have a real problem with Scalia).
You ought to find this easy, given that she has a longer judicial history than ANYONE now on the court did when confirmed (Justice Breyer had fourteen years as a judge before his nomination to the Supreme Court (1st. Cir. in 1980, S.ct. in 1994— Sotomayor was nominated to S.D.N.Y in 1992 and the 2d. Cir in 1997).
Want to share with the class?
True, but isn’t this basically the same thing? The EEOC four-fifths rule states
So sure, we can’t have hard-and-fast quotas. But these so-called antidiscrimination laws set a sort of metaquota, a criterion for criteria that mandates, just as strictly, that you have to have equality of outcome. (By the way, have you noticed how ignorant the Four-Fifths rule is simply in terms of simple statistical numeracy? Personnel selection differences should be measured in standard deviations, not in percentages.)
But that puts an enormous burden of proof on the employers, since it’s extremely difficult to prove that any nontrivial test is related to job performance, especially to a hostile advocacy group. Weightlifting is physical and thus trivial to defend, but I’d bet that spatial reasoning and verbal skills are equally vital for firemen. But if these tests ever generate inequality of outcome, they will be hard to defend.
In fact more occupations are linked to intelligence than you’d think. Aptitude testing was once widespread among hiring agencies, who poured significant resources into developing or licensing them, presumably because it let them pick the best applicants. It took a Supreme Court decision, Griggs v. Duke Power, to force them to stop in the interests of equal outcomes. (One laggard is the NFL, which never pretended to produce statistical representation, and uses the Wonderlic IQ test.) It is rather plausible that this led to the boom in college degrees and subsequent tuition inflation: if you can’t directly test for intellectual aptitude for your job, then the next best thing you can do is use an indirect signal - a diploma - for their intelligence.
That depends upon how the test is set up. In my example, one could as easily say that a failure to answer the question accurately by a larger percentage of females would simply show that females can’t do math. Indeed, this was the presumption for decades, until someone came along and said, “hey, maybe we should re-examine that concept.” Amazing how smart girls became after they did. :smack:
Which is NOT to say that it cannot be the case that, in general, minorities are less qualified in general than “white” candidates, or that these specific minority candidates were less qualified than the “white” candidates. But frankly, anytime you point to something as simple as skin-color and try to explain disparate results solely on the basis of that factor, I think the engineer in you should raise your eyebrows in skepticism. 
Sam, I don’t know what the evidence on the issue was, but I do know that the City called in a company to examine the test (different from the company that produced the test), and there was testimony from that examining company that supported the determination that the test had aspects that could produce disparate results racially. This was a part of what the City relied upon in deciding to invalidate the results of the test.
Also, as someone else just above has pointed out, the results, though seemingly statistically less-than-overwhelmingly-significant, violated the EEOC’s rules on disparate impact. Had the City proceeded with the results, the were fairly certain to face a Title VII challenge that would have been supported by the EEOC rules. I’m not conversant enough on the law in the area to know if this sets up a conclusive presumption of disparate impact, or only a rebuttable one.
One of the important points to keep in mind about the case is that it only tangentially involves concepts of reverse discrimination, although the firefighters who sued certainly wanted to make the case out as one of reverse discrimination. The City was faced with a problem. They had evidence from the results that raised at a minimum a rebuttable presumption that their test was racially discriminatory. Under the rules of the EEOC, they would have faced near certain challenge from the minority test takers had they accepted the results and used them to establish the top three ranks for promotion. They were advised that they would likely lose such a challenge. So they did what they were supposed to do: reject the results of a biased test. The effect of this created an adverse impact upon the essentially all “white” firefighters who would have been eligible for promotion under the “rule of three.” But the district court judge, in a ruling eventually adopted by the Circuit Court of Appeals by per curiam en banc order (IIRC), rejected the claim that this was a Title VII violation itself, since Title VII mandated the action taken by the City.
This is not to say that the Supreme Court will have the same view of the case when the decision comes down in the next few weeks. My guess is that it will be remanded to the Circuit Court for further decision after clarification by the Supreme Court as to how to handle the seeming impasse that the Title VII regulations and statutes have created here. Or, they might send the matter back down with a determination that, whatever the minority candidates may have in the way of remedies for disparate impact, the City cannot avoid that through discriminatory action against the non-minority candidates who presumably demonstrated the knowledge sufficient to support promotion. I didn’t pay attention at the time of orals to find out what Justice Kennedy appeared to think (I was busy with Geometry students!).
The term, ‘reverse discrimination’ is an unholy racist epithet. It’s beyond disgusting. There is no such thing as ‘reverse discrimination’. There is discrimination, period.
I don’t think skin color has any bearing on the inherent ability of someone to perform on this test. It isn’t as though putting the white firefighters in a tanning booth for a few hours will diminish their performance. Skin color does, however, have a large impact on the environment a person is raised in. Black people have been the subject of systematic discrimination. As a result they, on average, go to worse schools, have worse family situations, grow up poorer, and in general grow up in a worse environment for education. Because of this, I expect them to be on average less skilled than their white counterparts.
I don’t really have a problem with affirmative action type programs. I do have a problem with people blaming the thing that reveals the truth instead of blaming the underlying truth.
Yeah listening to Hutchinson and McConnell today, it certainly seemed like they would be opposed to such shenanigans.
What if I’m measuring reflectivity?
My understanding is that the district court sealed the tests from public view. Moreover, the district court’s decision (warning: PDF) turned solely upon the disparate outcome ratio. The court did not attempt to analyze the test itself.
No argument is being made that the test itself is biased in some way; indeed, the test creator, Industrial/Organizational Solutions would be at liability risk themselves for their business if they did not take great pains to purge their tests of bias.
The body of material covered on the test was available to all participants on a equal basis. No argument is being made that preparation opportunity was unequal. New Haven's Racial Test - WSJ No argument is being made that those who performed poorly somehow were disadvantaged in any way.
The only argument being advanced is that the act of testing itself had a disparate impact and that because there was a disparate performance, there must have been something wrong with the test. This is nonsense. The same disparate results occur in thousands upon thousands of tests across the nation and across the world, and the rank-order is universally the same: blacks do poorly compared with whites and asians on standardized tests for cognitive ability and retention. All the New Haven test does is confirm what every school in the country already experiences.
The conclusion is as clear as it would be if the test were for skillsets for the NBA: there are inherent, innate differences between populations of people, and testing confirms it. It confirms it over and over again; it confirms it when any mitigating factor is corrected for, and it will continue to confirm it until we stop testing, stop categorizing populations or become so intermarried that genetic differences among populations smooth out.
SCOTUS will reverse Judge Sotomayor’s decision to allow New Haven to disregard the test simply because it did not lead to racial diversity. Testing will never lead to diversity because populations have different average abilities. There is no merit to the argument that this test discriminated in any way other than to distinguish those who were able to prove their ability to master the material from those less able to do so. There may be better ways to find the best firefighter captains and lieutenants; there may be qualities beyond the ability to “read and retain” (to quote a black firefighter); there may be good reason to have a racially diverse firefighter leadership. But one thing is obvious: there is a disparate ability among populations that does not go away simply because we want it to or wish it did not exist. The New Haven test results are not some sort of oddball aberration resulting in unfair discrimination. They are no more surprising than a test which finds women cannot hit a golf ball as far as men. What else is new?
To the contrary, the Court specifically found there had been evidence that the test contained material that was not job related, that working firefighters judged the written exam as a poor indicator of future performance, and evidence that when a different test was given 1996 and 1999, there was no disparate impact.
In other words court found that the FD had a good faith belief in the illegality of the test when they decided not to certify the results.
The court found the plaintiffs established their prima facie case of facts which give rise to an inference of discrimination, but also the FD had proferred a legitimate non-discrimatory reason for their action, which was not a pretext (that they did not certify in good faith belief that the test was contrary to Federal law).
“Defendants’ motivation to avoid making promotions based on a test with a racially disparate impact, even in a political context, does not, as a matter of law, constitute discriminatory intent, and therefore such evidence is insufficient for plaintiffs to prevail on their Title VII claim.”
Since minorities overall tend to underperform on multiple choice tests, it could be argued that the choice to base promotion primarily on a multiple choice test–as opposed to, say, an oral exam or a practical exam–created a disparate racial impact.
I, personally, would not argue that. But they definitely had reason to think they’d run afoul of Title VII if they used the results of the test as they were. And while throwing the results out and using other criteria was a shitty thing to do, it’s not clear to me that they were doing anything illegal by doing so. So faced with a choice between an option that probably leads to legal trouble and one that probably doesn’t, they didn’t really have much choice.
Does the law need to be updated? Maybe. I think we have probably come far enough to hard-wire some protections for those who are harmed by overly agressive attempts to promote diversity. But from my lay understanding of the relevant laws, the city picked the better of two bad options.
No. I see nothing in the decision where the Court made such a finding. The only “evidence” of problems with the test itself is an assortment of allegations to that effect from those who did not do well. The AIR (Adverse Impact Ratio) was no different from that of prior exams; the pass rate for blacks was, and is about half of the pass rate for whites. And here is the remarkable thing: that underperformance, by itself, is considered by the Court to be sufficient reason to throw it out, because under EEOC guidelines it will obviously have an adverse impact. Duh. By this logic the PGA Tour’s Q-School is illegal because it has an adverse impact upon women.
There is an obvious problem here screaming to be addressed: If we insist upon categorizing populations for the sake of “diversity” we are going to have to come to grips with the fact that those populations are differently enabled. If we want to have standards that enable promotions, we are going to have to find ways to skew the tests with some sort of subjective evaluation that avoids absolutes because populations with less ability to “read and retain” a body of information are not going to be able to compete. (The decision here mentions that the City of Bridgeport simply expanded the weight of the oral exam as a means to this end.)
Folks, as a society interested in finding ways to create opportunity for all, we have got to stop dancing around this problem. I see nothing but farce upon farce as the traditional explanations for racially disparate are peeled away one by one.
From the decision: http://www.cir-usa.org/legal_docs/ricci_dc_op.pdf
“It appears that the reasons for testing disparities remain
elusive. Dr. Helms testified that many theories
exist, but experts on standardized testing nationwide
have been unable to satisfactorily fully explain the
reasons for the disparity in performance observed on
many tests.”
The explanation is simple: different populations have different gene pools. The only reason this explanation remains “elusive” is that when we are staring at it, we pretend it is not there.