I remember back in the ’70’s when this subject was much in the news, someone made up a short “sample” test skewed to knowledge and terminology common in the black community, as a challenge to white folks trying to define, e.g. “deuce-and-a-quarter.”
It meant a Buick (Electra?) 225.
That’s the only one I can remember from the test – possibly because I knew the answer. I also knew some black folks. ; ) Highly unlikely my mother or her neighbors of any age could have passed that test.
Culturally biased, indeed, but it gave one a sense of what was really meant by culturally biased exams.
This, of course, is why the College Board no longer describes the SAT’s as “aptitude tests.”

There is just no way that someone raised in a house w/o books and parents who don’t even realize the child’s school is doing a bad job, or who don’t see education as worth the effort, is going to do well on those tests, compared to someone whose parents began reading to them in the crib, talking constantly to them, and sent them to private/Catholic/suburban schools.

As a person who did pretty well on standardized tests myself, I needed to be shown why they didn’t work for everyone. but once that happened, it wasn’t hard to see.

We know that there is no rational, scientific basis for using a concept as nebulous and unscientific as race to predict intelligence, technical ability, judgement, or any other quality. So, given an objective measure of the qualities required of a fireman, we’d expect to see a random correlation between races and scores on the test. The proportion of the races represented in the group that scored highest would be esentially the same as that in the group that scored lowest.

If we don’t have such a distribution– if, as in this case, the “objective” measure selects for one racial group –then something is wrong somewhere. The measuring standard has to be biased somehow.

That’s logic. Does logic now have any place in the Reago-Bush court?

From what year (in the past) has it been thought that there were more than sufficient numbers of qualified firefighters from every group (however defined)?

In that time has the makeup of those promoted been equitable or skewed?

Justice Kennedy looked at the record and decided there was NO evidence of the process being so flawed as to invite liability. Justice Ginsburg decided the process was flawed, but she really did not give clear and convincing reasons for saying it was flawed.

Another point they disagreed on strongly was the motivation behind the decision to not certify the tests results. Justice Kennedy delved into the political influences on the decision that seemed to have predetermined the outcome of the process. Were the City officials using fear of litigation as a pretext for not certifying ? This was a very sticky point. It had been ignored by the lower courts . It did not matter what the Civil Service Board decided, the Mayor had already announced that he would overrule the Board if they certified the results. In that case, why did they bother to spend $100,000 and put the applicants through such an ordeal if the outcome had been predetermined? Judge Ginsburg did not address this question. The USA did address it in the amicus brief and that one one of the reasons they recommended vacate and remand for a trail on the facts.

The biggest problem I had was the part where the City claimed that by not certifying, they were treating everyone equally and not engaging in an illegal hiring practice….that it was a race neutral decision and not disparate treatment. This was the axis around which the case spun. The decision not to promote anyone was making a choice between two set of litigants . They chose one and lost. Now they get to face the other set.

I certainly agree that any kind of evaluation is to some degree subjective, including performance evaluations and references. Just about anything can be skewed intentionally or by the incompetence or laziness of an evaluator. Not to mention maliciousness.
I sure don’t claim to know the answer to this knotty problem.
But I’d like to hear the talking heads acknowledge the thorniness, not pretend it’s a simple matter of “common sense,” as so many are doing.

I could swear that I read somewhere, while the case was pending, that there was some sort of “practical” type of test as wel in New Haven, on which the black candidates did very well, and that that was part of the consideration in tossing out the written test.

The test had two parts. The applicants had have a certain score on the written in order to be eligible to be interviewed by a 3-person panel made up of fire department officers from across the country with each panel consisting of 1 black, 1 white, and 1 Hispanic.

The union contract had set out the weighting to give 60% weight to the written and 40% to the oral. (Some other districts give a percentage for senority.)

Justices Kennedy and Ginsburg took very different views of the record all the way through and this is one of the issues they treated very differently. Kennedy thought it was reasonable and rational. There was no evidence of it being arbitrary.

At the Civil Service Board hearings, a member of the Firebirds (black firefighters group) from Bridgeport told the board that Bridgeport used a different weighting giving more to the oral and less to the written and that this gave them a better outcome or less disparate impact. Justice Ginsburg latched on to this as evidence of the New Haven test being flawed . Justice Ginsburg did not mention that there are at least two lawsuits by Bridgeport firefighters based on this ‘better’ method. The CS director had changed the weights to less the disparate impact and was challenged for race norming which the courts have ruled is illegal. It’s worth looking into the problems in Bridgeport before declaring them a model for other departments to follow.

The CSB also heard from a competitor of IOS (the consultants who were paid $100,000 to design the New Haven promotion exams) who evaluated the exams. He thought he could have come up with a better exam (and later was hired ) and talked about the use of assessment centers which some departments used instead of written and oral exams that are more performance based. They did not seemed to have been an option at the time in 2003 but he recommended them for the next promotion phase. He recommended that results of the 2003 exam be certified. Could these assessment centers be considered as a better option under the 2-step burder of proof shift in a disparate impact suit brought by the minority firefighters? Is this one of many questions that should have been presented to a jury with witnesses under or and cross-examined? This seemed to be the one thing all nine justices seemed to agree on. The majority put themselves in place of the jury and decided what a reasonable jury would find. They decided the New Haven 20 would prevail and issued a summary judgement. Justice Ginsburg and the minority favored remand as well but balked at the summary judgement and went the other way. (????)

The problem with tests and evaluations is a subjective/objective tension. Standardized tests have been used because they are more objective and if done well can measure knowledge and skills required for a job. The further an employer goes away from standardized tests, the more subjective they become and less tranparent the process becomes. Employers might like that since it’s hard to prove anything one way or another. Employees, however, like to know that there are measurements and objective criteria so that they can look at the playing field and the rules before hand when they are competing for promotions. Civil service exams were instituted to counter nepotism, cronyism and all the other -isms that corrupt the process. No one claims they are perfect and there will always be flaws. They can always be improved. However, every move towards subjective evaluations introduces opportunities for mischief.

References and job performance evaluations has been discarded from the New Haven promotion process because of their subjective nature. Senority is always controversial, but at least it is measurable and tracks with experience related to the job. But it’s not easy to measure things like leadership ability, interpersonal skills, and other personal qualities. It ’s because of these tensions, there will always be a lot of work for labor lawyers and many more cases like this making their way up the ladder.

I think they would have saved themselves a lot of problems here by sending it back for a trial on the facts with the new standard that was introduced and some clear guidelines on how to apply it.

Honestly, you can’t just answer that with a pat response without having facts and particulars and history. Would it be the same for a city in the north or in the south? What about one with a long history of Jim Crow law issues even after the civil rights era was long over and done with? Or with a city that has a substantial ethnic population and make-up on the force, but almost all white leadership ranks for years traditionally? Is that fair? What if the whites are better managers and for safety reasons that’s been the better course?

Honestly, it’s impossible to answer a flat out response on something like that. The fact pattern and history is integral in my mind of whatever individual case you look at. It’s not fair to discriminate against someone based on their race or gender any more than it’s fair to discriminate against them because they aren’t ethnic? How do you achieve a fair balance? Wish I knew. *g* But having some minorities and women in leadership roles is important, just as not passing over qualified applicants because they aren’t women or minorities is.

These are really tough questions that municipalities have to answer for themselves all the time. City government is one of the tougher things to manage, and I say that having been a city attorney for one several years ago. All of the personnel issue get magnified substantially when the public eye is on you — it’s not easy making anyone happy, let alone everyone.