As mentioned:
IMHO he deserves to be compensated for the cost, and I agree the timing sucks, but it is not anti-white discrimination, once again you can only reach that conclusion by denying that there is or was no anti-black and anti-Hispanic problem that the law tried to take care of. That the law should be changed is something I could agree, but I have to see more evidence than just this.
Chief Pedant I too would like to see an analytical breakdown of those who passed. One of the problems with these ideas is that we don’t recognize all the truly unexceptional white people who exist in the world. There simply are more white people than black people so even if the disparate advantage white people have is a small amount, the filtering processes will ultimately lead to more white people than black people, simply because white people outnumber black people by a significant margin.
I don’t think Barack Obama, Colin Powell or Condoleeza Rice have a problem taking tests. They are probably smarter than the vast majority of white people. If in this particular sample of the population two whites and one hispanic happen to be the most qualified, then so be it.
GIGOBuster Well the law might help in some cases, but it seems to me that assuming that blacks are going to fail standardized tests leaves us without a very common tool that we use to establish merit.
The issue in these testing cases often boils down to whether the testing is rationally related to the job duties. A defense to allegations of employment discrimination is the “Bona Fide Occupational Qualification” or BFOQ. It is specifically NOT a violation of civil rights law to screen for abilities that are inherent to the job.
So, for example, if a company says that to work as a secretary you must be no more than 120 lbs, that will screen out most men with no logical connection to the performance of duties as a secretary. Being 120lbs is not a BFOQ of being a secretary, the occupational qualifications of a secretary are things like high typing speed/accuracy, knowledge of MS Word and similar, etc. . This is likely to be illegal gender based discrimination in hiring, directed towards men.
On the other hand, if to be a firefighter you must be able to perform a strength test (lets say, lifting 120 lbs) that would not be discriminating against women, even if it screens out most women, if it is completely connected with the job duties, thus being a BFOQ.
In one of the greatest tragedies in employment law, these waitresses suing Hooters settled, so we never found out if having big boobs can be a BFOQ as claimed by the employer 
(Note that only under very rare circumstances can customer preference be a BFOQ. A BFOQ normally has to be an inherent aspect of the job.)
So, if these exams test something that is inherent to the job, let’s say, I don’t know, fire code rules or life saving procedures, they would be okay. If they test for general scholastic aptitude, I would think not.
So, the takeaway - a test that screens out all or most minority applicants while testing for something unrelated to the job duties, generally is unacceptable. This is what “needlessly screens out minority applicants” means in the context.
Hello Again I seriously hope that a firefighter has to lift an amount of weight that is far beyond the capability of most women. Lifting 120lbs wouldn’t even come close to dragging me from a burning building. I seem to remember there being a case about that issue specifically where the lifting requirement was 300lbs and they said it discriminated against women. It was a long time ago and I don’t know any specifics so I might not be remembering it all that well.
Well, yes and no, in the case of the firefighters I think that having good muscles, street smarts and to also be able to communicate properly with the community helps a lot in an emergency.
AFAIK this is not the only test the firefighters had to take.The city still can not just select people that are not competent for the job.
Incidentally, I agree on the issue of women firefighters, because that is related to the overwhelming physical demands of the job. Physical standards should not be changed IMHO.
However, once you take into account what was the issue of New Haven, then one can notice that many would not want to see a firefighter that is an intellectual giant with a weak body attempting to pull a fat guy from a burning building. 
Of course. I can’t tell what Frank Ricci’s stature is by his pic: Frank Ricci's Page - Fire Engineering Training Community
The thing about it though is that this test was for promotions within the Fire Department, and not about him being a grunt on the ground. You’d want your firefighters who are pushing pencils to have some intellectual capacity.
A test that is multiple choice can easily be biased. I will give a brief example, though the bias won’t necessarily be “anti-black.”
I am putting together a math question for the SAT. The math question, rather than just saying, "33% of 525 is ____ " asks the question using a baseball hypothetical, in which a player has 525 at bats, and hits .330, resulting in how many hits? Now, for those who are at all aware of baseball and baseball statistics, this question is not difficult to answer. But if you are not aware of baseball, if you pay little attention to the game, and the statistics that accompany it, you will be at a disadvantage in trying to determine what exactly the question is asking.
This is a very simple example of what is a very complex science: reducing or removing test bias from standardized tests. One way to tell if a test is biased is to look at the results of the test over time and see if they produce unexpectedly low scores for certain groups of test takers, when other potential score-lowering factors are accounted for. Indeed, one of the ways of proving that discrimination is occurring against a particular group is to use statistics showing otherwise unexplainable low results for a protected class, such as “blacks.”
So, the issue in the New Haven case is this: the city offers its firefighters a test, the results of which will be used to create a promotion list. The list will follow the “rule of three,” that is, that promotions have to be offered first to candidates who score in the top three rankings of the results. When the test is scored, it turns out that no blacks, and IIRC only one Hispanic, fall into the top three ranks. In addition, overall, the scores of “black” and Hispanic candidates are significantly lower as a group than those of the non-minority candidates; relatively few of the minority groups “passed” the test (which makes them eligible for promotion at all, once the top three ranks are promoted or decline promotion). The City of New Haven finds these results alarming; they worry that this is evidence that the test unintentionally discriminated against the “black”, and to a lesser degree Hispanic, candidates. Fearful that a lawsuit under Title VII would follow, they decided to annul the results and try again with a different test. As a result, certain non-minority candidates sue under Title VII, asserting that the city has engaged in intentional discrimination against the non-minority candidates.
As for whether or not it is acceptable to discriminate in favor of non-“white” people as a government, the answer is yes, it is acceptable, but how it can be done is not exactly settled law. Starting with the Bakke case in 1978, up through the University of Michigan cases of a couple years ago, the Supreme Court has said that race may be a factor used to determine the admittance of candidates to schools and the like. In addition, other cases have established that affirmative action programs are an acceptable method of remedying the impact upon society of past discrimination against “blacks.” The legal analysis for this follows the following algorithm: remedying the effects of anti-former slave discrimination is an issue of vital importance to government, and as long as the method is very closely tailored to help accomplish that goal, it will pass the otherwise strict scrutiny to which we subject such discriminatory acts.
The law in this area is VERY fluid. Stay tuned for developments. 
Has anyone published the test itself, so it can be evaluated for bias or for the relevancy of the questions? That would seem to be an important thing to look at here.
Excellent post and test question example, DSY.
Brenda Berkman sued the City of New York (the physical test was 120lbs of deadweight cement.) At trial it was proven that female firefighters who could easily carry up to 300lbs live weight could not lift the 120lbs dead weight,and thus, is was not a good measure of job capability. 100% of the women failed the deadweight test, and 50% of the men. It was a new test that had never before been administered. Brenda Berkman was an attorney and a marathon runner at the time.
Berkman went on to become a firefighter and on 9/11 was among those on the scene who ran in full gear into the collapsing buildings to save people. I believe her engine company, based in Park Slope, Brooklyn was a first responder. She survived the experience and retired after 24 years in 2006 with the rank of Captain. The FDNY remains 99% male.
I would say that even when promoted firefighters still see action and have to deal with the community in an emergency.
As mentioned before, I agree that rules need to be changed to avoid situations like this, but it is clear that attempting to make this a reason to be against Sotomayor is silly.
Yeah, I don’t even care as much about Sotomayor’s involvement as I am intrigued by the case itself.
Sure, but I was not born yesterday. 
Mentioning her but not the other judges shows to me that there was another agenda in the OP.
Umm, those other judges are not being nominated for the Supreme Court and so I don’t know who they are because their names are not on Television 24/7? Could be that, or you could have found some deeply rooted bias.
You lawyers get the engineer in me all riled up. You can’t ignore a test because it gives you a result you don’t like. If there is something demonstrably wrong with the test, as in your first example, then I can see discarding the test. However, if the test accurately measures knowledge (i.e. What fire extinguisher should be used on an electrical fire) and minorities do worse, it isn’t the tests fault. It is the minorities fault for being less skilled at their profession than their white counterparts.
^^This.
Ok, so then I get to say again that blaming anything regarding this case on Sotomayor is silly. Hey, I’m easy.
Sotomayor is not racist at all as many Republican leaders are saying even though they are troubled by her remarks.
It is only former extreme conservative Republicans and gasbags who are trying to say otherwise.
It doesn’t seem like she’s a racist to me either, but the way this has played out is pretty anti-white racism as I can tell. Not her fault, but racism as a trick of procedure.
But that itself begs the question-it assumes that the test accurately measures knowledge. It’s completely different, as you should very well know, if, as in the firefighter case, we don’t know whether the test is accurate.
Let me give you an engineering example. You design a one-foot ruler. Do you know if it accurately measures length? No. You have to test it to see if the markings are correct.
If you test it against another ruler known to be correct, and they match, then you are right that the simple fact that you don’t like the results of measurements of unknown objects isn’t enough to justify rejecting them.
It’s different if you don’t have an object of known length. You take your ruler, and measure a fish you caught (legal minimum, seven inches).Your ruler says six inches–and you don’t like that much. But does that mean the fish is too short, or your ruler wasn’t designed correctly? You can’t tell on the facts as I’ve given you.
And further, if you do have an object of known length, and measure it against the ruler, you can find out if the ruler is accurate. If you don’t have any ruler to test your new ruler against, but you do have a screwdriver that is less than six inches long (it fits in a toolbox of known size), and measure it, and your ruler says eleven inches, the ruler is wrong. It’s wrong even though you don’t know precisely what the length of the screwdriver is.
It may be hard to test an exam while designing the test–there is no “other ruler” to use–but it is possible to check the results of the test when it’s first applied.
We have an easy way to test if a ruler is accurate at determining length. Not so much for a test for fire-fighting knowledge (as in the baseball example earlier, a test that seems neutral on its face can in fact be biased in unwanted ways).
Given that, if a test suggests there is such a bias, we have to ask if that is evidence that there is a disparity, or if there is a bias inherent in the test. At the very least, there’s no way to tell the two apart without knowing how the test was developed. Further, if (as I think is the case), we have no reason to believe that one racial group is any better than another in knowing how to fight fires, seeing a disparity in the test results suggests the disparity is coming from the method of testing, not the thing being measured. And that’s a good reason to reject the test.