Reverse Discrimination Case

[jillio]

ah - the post #248. I see that you made a lengthy last-minute edit. it's about time. let's see if my mind can grasp it.

My edit relates only to the way that Naisho entered and identified data in the SPSS program, and the statistics he asked it to run.

 

[Jiro]

My edit relates only to the way that Naisho entered and identified data in the SPSS program, and the statistics he asked it to run.

so where's the result? perhaps i'm not seeing it. perhaps I need to grab some shuteye.

 

[jillio]

so where's the result? perhaps i'm not seeing it. perhaps I need to grab some shuteye.

Yep,you are just not seeing it. I have given you the skews that indicate problems in reliability based on z-scores. Z-scores are used to indcate relative position, and to interpret the meaning of a raw score.

Let me ask you this: Have you ever had a professor grade an exam on a curve?

And I know what you mean by needing shut eye.:P

 

[Jiro]

Yep,you are just not seeing it. I have given you the skews that indicate problems in reliability based on z-scores. Z-scores are used to indcate relative position, and to interpret the meaning of a raw score.

so it's in post #348? if not - what post #?

Let me ask you this: Have you ever had a professor grade an exam on a curve?

that I understand.... and the answer's yes.

And I know what you mean by needing shut eye.:P

yep.... toodle!

oh btw - I have made an error in my post. You were right that it is being heard by US Supreme Court. I'm aware of the procedures involved but I'm not exactly sure how I make such an error. perhaps I really should get some

 

[jillio]

so it's in post #348? if not - what post #?


that I understand.... and the answer's yes.


yep.... toodle!

oh btw - I have made an error in my post. You were right that it is being heard by US Supreme Court. I'm not exactly sure how I make such an error. perhaps I really should get some

You can answer this in the morning, but when an exam is graded on a curve, what exactly is the professor doing? How do they arrive at the number of points used in the curve? How do they know that a curve is even needed? I can promise you this: they don't just add arbitrary points because they like the students or because they feel bad that a large percentage got low grades.

What happens when a test is graded on a curve is that a majority of the scores fall toward the negative end of a normal curve. That creates a positive skew. That indicates that there was a problem with the test format for that particular population, otherwise, according to the mathematical laws of probability, the scores would have fallen on a normal curve. In order to correct the positive skew in scores, and correct for any problems in test format that was used, they statistically determine the number of points that must be used to bring the distribution into a normal curve. Professors will grade on the curve because most tests given in a classroom have not been checked for validity. In fact, correcting for skews and bringing the scores into a normal distribution is what the term "grading on the curve" means. Populations of classrooms differ. Tests are changed every semester or quarter so as to prevent them being passed on. Therefore, when a professor gets a postive skew, they correct for it by grading the instrument on a curve. They don't just look at the raw scores, but they interpret the meaning of those scores in relative position.

That is very similar to the situation we are discussing.

 

[ncff07]

im not reading this many pages of bs and bickering but ive read enough to know this thread is a damn joke and im offended by some of the posts about firefighters. guess were just dumbasses.

 

[jillio]

im not reading this many pages of bs and bickering but ive read enough to know this thread is a damn joke and im offended by some of the posts about firefighters. guess were just dumbasses.

And that is exactly what I am trying to demonstrate is untrue. The problem with the scores lies within the test, not the test takers.

 

[lsfoster]

Sorry. You are using raw scores, and you are using only the combined score. Raw scores are meaningless; you need z-scores, and then, if you wish to compare further, you will need t-scores. These must be computed for all participants in all domains and then plotted. If you don't compute across all domains, you don't see the influence of weight given to subsets.

I personally don't think that a lot of what you post is necessary in civil, adult discussion...but it all goes back to roomfor interpretation, now, doesn't it?

Funny, because they are all that was considered by the city when it threw out the results. If you want to discuss whether or not they made a poor decision (which was the original point of this thread), then that would be all you could consider.

If you check the raw scores and the way they are distributed, you will discover that you have a postive skew in the raw scores. When converted to z-scores, the scores are still distributed in a postive skew. When you sort by race, and domain, you will find a positive skew in all domains for the minority races, with the greatest postive skew for blacks in the written subset, and the greatest positive skew for Hispanics in the oral subset. However, because more weight was given to the written scores, the positive skew disparately shows with blacks moreso than Hispanics in the combined scores. This indicates a problem with validity for the minority populations.

It doesn't definitively indicate a problem, since you still have absolutely no other information. You have no indication of how well qualified any of these men were, and are ignoring any other factors that might affect their scores. Without any information about the exam, or any knowledge of the men and their qualifications, you can only say that it might be a problem with the validity, just as much as it might be the case that the individual minority applicants in this specific situation weren't as qualified. You have absolutely no more evidence than anyone else.

im not reading this many pages of bs and bickering but ive read enough to know this thread is a damn joke and im offended by some of the posts about firefighters. guess were just dumbasses.

I don't know where anyone said anything even remotely close to that. If you would like to point out some of these comments, I would happy to either explain what I meant, or apologize if you found something offensive.

And that is exactly what I am trying to demonstrate is untrue. The problem with the scores lies within the test, not the test takers.

And again, you have no evidence which would allow you to say that.

 

[Reba]

Ahhh...those things that can't be tested with pen and paper.

That's why the pool of test takers had to be deemed "eligible" (time in service, experience, etc.) before taking the test, and they also took an oral exam covering possible fire fighting scenarios.

 

[Reba]

.. They could've opted to take another exam that was a little bit more fair as well more accessible for the minority firefighters.

Was that option available?

 

[jillio]

Funny, because they are all that was considered by the city when it threw out the results. If you want to discuss whether or not they made a poor decision (which was the original point of this thread), then that would be all you could consider.

How, in fact do you know that raw scores were all that was consider? Where exactly does it say that anywhere? Especially since all the courts that have previously heard the white firefighters' case (it is in the U.S. Supreme Court at this point. The final stop.) have upheld the decision to throw out the test scores for the purpose of promotion. And I'm certain that the city did not undertake to throw out the test it paid $100,000 for without asking someone "What do these scores mean?" In order to answer that question, z-scores must be obtained. It is obvious now that you are continuing to argue for no other reason than to simply be argumentative. You are also assuming that the the city officials that threw the test out were operating on as little information and comprehension as you ,yourself, are demonstrating. And raw scores are not all to be considered in evaluating their response to the problem, because the raw scores, in and of themselves, do not tell us, nor the city, nor the courts, what they need to know. I suppose you can believe, if you choose, that raw scores are all that was considered prior to scapping a test that had cost the city $100,000, and that raw scores were all the courts considered in upholding the city's action by dismissing the white firefighters' lawsuit, but that is not evidence of wrongdoing on the city's behalf. It is simply evidence of the fact that you will believe in fairy tales.

It doesn't definitively indicate a problem, since you still have absolutely no other information. You have no indication of how well qualified any of these men were, and are ignoring any other factors that might affect their scores. Without any information about the exam, or any knowledge of the men and their qualifications, you can only say that it might be a problem with the validity, just as much as it might be the case that the individual minority applicants in this specific situation weren't as qualified. You have absolutely no more evidence than anyone else.

Yes, dear, it does definitively indicate a problem. When you achieve a skew such as the one produced in the results of this particular testing instrument, there is definately a problem somewhere in the test. And again, I will remind you that the courts have had sufficiewnt evidence presented to dismiss the lawsuit. You don't know from the test what the qualifications of the applicants were...it has already been stated by the professionals that the test failed to assess that properly. Which simply give more credence to the fact that the instrument itself was flawed. I have more evidence than you, because I did the statistical analysis on the raw scores. If you want to discuss evidence of bias, I will suggest that you do the same. Unless, of course, the reason that you have failed to do that in order to attempt support for your argument is that it is completely out of your realm of ability. Which, in and of itself, removes all weight from your argument.

I don't know where anyone said anything even remotely close to that. If you would like to point out some of these comments, I would happy to either explain what I meant, or apologize if you found something offensive.

I understood exactly what ncff07 was talking about.

And again, you have no evidence which would allow you to say that.

Yes, I do. Just because you refuse to see the elephant standing in the middle of the room does not mean that it doesn't exist. This is not a case of the evidence of problems with validity in the testing instrument doesn't exist, it is simply evidence of your refusal to accept that it does exist in order to continue arguing for argument's sake. Are you related to the firefighter who filed the lawsuit and continues to argue a dead point all the way to the Supreme Court?

 

[jillio]

That's why the pool of test takers had to be deemed "eligible" (time in service, experience, etc.) before taking the test, and they also took an oral exam covering possible fire fighting scenarios.

And the oral portion of the exam was weighted inproperly to assess for true ability and knowledge. That is one of the problems with the instrument that was cited by the professionals who examined the instrument. And I described, using Z-scores, how the improper weighting of the subsections actually caused a greater skew in specific populations when figured into the overall scores. In fact, the imporper weighting, when looked at in terms of each individual score, also had a disparate effect on several of the individual white firefighters. I would imagine those are the ones that didn't file a lawsuit.

 

[jillio]

Was that option available?

The city cannot issue a new instrument and resume the promotions until this has been decided in the courts. Since the white firefighter responsible for filing the lawsuit has decided to appeal the decisions of the lower courts that heard the case all the way to the Supreme Court, they will not be able to take another exam until that decision is handed down. This lawsuit is what has prevented alternative solutions to the problems found in the original testing instrument.

 

[Jiro]

And that is exactly what I am trying to demonstrate is untrue. The problem with the scores lies within the test, not the test takers.

just so we all understand this issue clearly which is on U.S. Supreme Court -

BREAKING: U.S. Supreme Court Agrees to Consider New Haven Firefighters Case

The court accepted the case to decide the following question:
Whether municipalities may decline to certify results of an exam that would make disproportionately more white applicants eligible for promotion than minority applicants, due to fears that certifying the results would lead to charges of racial discrimination.

New Haven, Ct. Firefighters Claim Reverse Discrimination

But in the city of New Haven, Vargas, who is Hispanic, and 19 white firefighters say that is where the equal treatment ends, and discrimination begins. They allege that they were denied promotions because the city gave preferential treatment to blacks.

Matt Marcarelli, who is white, got the top score on a promotion exam in 2003 and was first in line for captain. But when the city reviewed all the test results, it found that the pass rate for black candidates was about half the corresponding rate for white candidates. None of the black firefighters scored well enough for an immediate promotion. As a result, the city threw out the test results.

In New Haven, city officials knew they were headed for a catch 22 when the test results came back. If the city certified the test results, it was confident it could expect a lawsuit from the black firefighters. But when it threw out the test results, it instead got a lawsuit from mostly white firefighters.
.....
"If we lose this," New Haven firefighter Octavius Dawson said, "the implication is catastrophic. I mean, where does it end. Not just with the fire department. Police department, education, who knows where it could end?"

But aggrieved firefighters say they want to take race out of the equation.

"We want to be treated just as firefighters, whether we are men, women, white, African American, Hispanic. We want to be treated as firefighters, period."

in short - this case is about or related to Affirmative Action... not the test itself. So it's the political agenda within the city, not the test.

 

[jillio]

just so we all understand this issue clearly which is on U.S. Supreme Court -

BREAKING: U.S. Supreme Court Agrees to Consider New Haven Firefighters Case


New Haven, Ct. Firefighters Claim Reverse Discrimination


in short - this case is about or related to Affirmative Action... not the test itself. So it's the political agenda within the city, not the test.

How can it not be about the test when the question is whether the city can opt not to certify test scores due to disparity in the scores? It clearly states that the case questions the city's right to not certify test scores that favor white applicants. If the test favors white applicants, one must question why white applicants are being favored. In this case, it was found that the test itself created at least some of the disparity in scores, meaning that more white applicants were eligible for promotion based on test scores that came from a test that was found to be cross culturally invalid. Without the test, there would be no lawsuit, because no scores would have been thrown out, and the promotions would not have been revoked. No matter how you look at it, the test is the foundation of the lawsuit, because without it, there would be no lawsuit. The courts are being asked to decide, in this specific case, if the city was justified in its action. That action was the throwing out test scores based on disparities that indicated problems in validity.

Using said test scores, because it was found that the test indeed contained problems that created bias for white candidates, would have been in violation of Title VII. It all goes back to the test scores and whether the results of the test were valid. If the test had been found to produce valid scores and did not contain bias, then it is not in violation of Title VII to promote white candidates and not promote Black candidates. The intent of the law is to insure that the qualifications of minority and majority candidates are assessed equally. In this case, the test is what was used to assess that. And the test was found not to assess ability equally for all candidates. That does not mean that black candidates were less qualified. We cannot know that until all the candidates are given a test that does not contain racial bias.

From Jiro's first link:

The U.S. Supreme Court today agreed to consider the case of New Haven, Connecticut firefighers who alleged reverse discrimination when the city denied them promotions despite high test scores, ostensibly out of fears that the test may have discriminated against minority applicants.

In order to address the above, the test itself must be an issue.

 

[Jiro]

How can it not be about the test when the question is whether the city can opt not to certify test scores due to disparity in the scores? It clearly states that the case questions the city's right to not certify test scores that favor white applicants. If the test favors white applicants, one must question why white applicants are being favored. In this case, it was found that the test itself created at least some of the disparity in scores, meaning that more white applicants were eligible for promotion based on test scores that came from a test that was found to be cross culturally invalid. Without the test, there would be no lawsuit, because no scores would have been thrown out, and the promotions would not have been revoked. No matter how you look at it, the test is the foundation of the lawsuit, because without it, there would be no lawsuit. The courts are being asked to decide, in this specific case, if the city was justified in its action. That action was the throwing out test scores based on disparities that indicated problems in validity.

Using said test scores, because it was found that the test indeed contained problems that created bias for white candidates, would have been in violation of Title VII. It all goes back to the test scores and whether the results of the test were valid. If the test had been found to produce valid scores and did not contain bias, then it is not in violation of Title VII to promote white candidates and not promote Black candidates. The intent of the law is to insure that the qualifications of minority and majority candidates are assessed equally. In this case, the test is what was used to assess that. And the test was found not to assess ability equally for all candidates. That does not mean that black candidates were less qualified. We cannot know that until all the candidates are given a test that does not contain racial bias.

which is impossible. What's next? create the test that does not contain disability bias (such as dyslexic)?

"In New Haven, city officials knew they were headed for a catch 22 when the test results came back. If the city certified the test results, it was confident it could expect a lawsuit from the black firefighters. But when it threw out the test results, it instead got a lawsuit from mostly white firefighters."

 

[jillio]

which is impossible. What's next? create the test that does not contain disability bias (such as dyslexic)?

"In New Haven, city officials knew they were headed for a catch 22 when the test results came back. If the city certified the test results, it was confident it could expect a lawsuit from the black firefighters. But when it threw out the test results, it instead got a lawsuit from mostly white firefighters."

No, it is not impossible. And dyslexia is a disability covered under the ADA, so refusing to provide accommodation for a dyslexic employee when they are taking an employment exam is a definate case of discrimination under the ADA.

And the catch 22 was created because of the disparities in test scores created by the instrument. Both sides had a basis for a lawwuit because the test scores were inherently biased either for one side or against the other. Had bias not been evident, no one would have had a basis for a lawsuit. You can't have bias against without also creating bias for.

 

[Jiro]

No, it is not impossible. And dyslexia is a disability covered under the ADA, so refusing to provide accommodation for a dyslexic employee when they are taking an employment exam is a definate case of discrimination under the ADA.

And the catch 22 was created because of the disparities in test scores created by the instrument. Both sides had a basis for a lawwuit because the test scores were inherently biased either for one side or against the other. Had bias not been evident, no one would have had a basis for a lawsuit. You can't have bias against without also creating bias for.

again - it's impossible to create tests with no racial bias.

Supreme Court to Hear New Haven Race Case

New Haven firefighter Frank Ricci did what he was supposed to do. He bought the recommended books and studied for a promotion exam. Despite his dyslexia, Ricci scored high enough to qualify for a promotion, but the department threw out all test results. No blacks and only two Hispanics scored high enough to be promoted.

Over a dozen white firefighters and one Hispanic filed suit against the city in 2004, claiming it violated their constitutional rights and Title VII of the Civil Rights Act by discriminating against them based on race. Had the fire department certified the test results, however, the lower scorers likely would have sued the city under Title VII's "disparate impact" provision. The fire department was damned if it did and damned if it didn't.

The district court granted summary judgment in favor of the city and dismissed the case. A three-judge panel of a federal appeals court affirmed the dismissal. Bush-appointed judges sought to have the case re-heard, but the court declined by a vote of 7 to 6. Judge Jose Cabranes, a Clinton appointee who dissented, defined the issue this way:

"May a municipal employer disregard the results of a qualifying examination, which was carefully constructed to ensure race-neutrality, on the ground that the results of that examination yielded too many qualified applicants of one race and not enough of another?"

In denying the rehearing, the court contended that a government employer "faced with a prima facie case of disparate-impact liability under Title VII" does not violate the title or the Equal Protection clause by taking facially neutral, but race conscious, measures to avoid liability. With a collective straight face, the court claimed that throwing out the test because too many whites passed it was "facially race-neutral."

Earlier this month, the Supreme Court agreed to hear the case. The disparate impact theory of liability will be front and center. Thirty-eight years ago, the Supreme Court held in Griggs v. Duke Power Co. that for purposes of hiring, an employer's use of a high school diploma requirement and two standardized written tests violated the Civil Rights Act. Black applicants disproportionately lacked diplomas or failed the tests. Griggs laid out the disparate impact analysis for employment. Absence of discriminatory intent is not the end of the discussion. Even if an employment practice is "facially neutral," (a scored test, for instance) it is suspect if it has a disparate impact on members of a protected class. To avoid liability, businesses would have to demonstrate that such tests are a justified "business necessity" or related to job performance.

The court in Ricci also will take up the issue of skin deep-only diversity. In 2003, the Supreme Court held in Grutter v. Bollinger that racial diversity is a "compelling state interest" that justifies race preferences in college admissions. (Justice Sandra Day O'Connor said the practice won't be necessary in 25 years.) The presence of a so-called critical mass of minority students provides educational benefits for other students.

Granting preferences to and discriminating against individuals on the basis of race are two practices that should have been tossed out of the government decades ago. Is ensuring racial diversity in a fire department a compelling state interest that justifies tossing out test scores because black firefighters failed to make the grade? They need more training and better study habits, not lower standard hand-holding.

 

[Daredevel7]

One difference between Jillio and me in this case: She has faith that a cultural bias-free test can be created whereas I do not. Oh well!

One thing that I've noticed that no one pointed out.... the fire dept DID try for a cultural bias-free test by hiring a consultant to generate the test. Is it really their fault that the test didn't do what they hoped for? They, unfortunately, got themselves into a Catch 22 by "trying to do the right thing".

 

[Jiro]

One difference between Jillio and me in this case: She has faith that a cultural bias-free test can be created whereas I do not. Oh well!

One thing that I've noticed that no one pointed out.... the fire dept DID try for a cultural bias-free test by hiring a consultant to generate the test. Is it really their fault that the test didn't do what they hoped for? They, unfortunately, got themselves into a Catch 22 by "trying to do the right thing".

and also sexism! do you think your engineering program/exams are perhaps.... gender-biased?

better sue sue sue!