You know, the funny thing about AAVE (didn't even knew that was a term) is that I, as a southern black guy, was raised around people who used it and can quite easily slip into it myself, so reading the wiki article is sorta weird. A lot of the stuff it's explaining is just shit that's normal for me to say around other black people. That whole Remote Phase verb thing never really occurred to me until I read the explanation and realized that I'd been doing that most of my life.
Welcome to language. If you learn a language in a non-formal setting, you don't learn the technical aspects of it. The only reason I know about AAVE is that I took a class on Creole Languages and we studied it for a couple of weeks and there's actually a lot of very interesting stuff about it. It's definitely not a dialect and definitely falls into the classification of a Creole language, because it has it's own hybrid grammar and is a whole lot more than just "dat is used instead of that".
You say its "definitely not a dialect" because you learned it from a biased source. When all you have is a hammer, everything looks like a nail. When you have an expertise in Creole, you see it everywhere.
A biased test is not a good indicator of who is the best.
There is no evidence that the test was biased.
Except whites passed the test at twice the rate of blacks and no nonwhite made the top 30. Statisticly this would not have happend in a non-biased test between equaly quailified test subjects.
AND BEFORE YOU SPEAK: There is no evidence that Blacks are less quailified then Whites, or less determined to succede, or less willing to study.
Ricci is the case every one mentions, with all the things he did to pass the test. Nobody mentions what the other 19 did to pass, or what the black firefighters did to pass. Cherrypicking Ricci himself is not an good argument.
As one Hispanic quoted anonymously by the New Haven Independent put it, the test favored "fire buffs"—guys who read fire-suppression manuals on their downtime and paid test-manual writers to come to New Haven to speak.
So it would seem that there is in fact evidence that the white firefighters went to greater lengths to prepare for the test.
Right, but that's not actually evidence that they are any better at their jobs, just that they are better at passing the test. What the test is testing for is another matter, like in Britain in the past, grammar school entrance exams tested how "clever" the students were by asking things like naming classical music composers, which had the result of discriminating in favour of middle class people. So in this fire dept case, your black guy and your white guy might be equally "good" at the job itself, but just when they got to a test some of the white guys like Ricci would have been better able to answer questions about naming every part of a fire hose, or saying when the New Haven fire dept was founded or something.
What the test is testing for is another matter, like in Britain in the past, grammar school entrance exams tested how "clever" the students were by asking things like naming classical music composers, which had the result of discriminating in favour of middle class people.
Again, the fact that biased tests exist is not evidence that this test was biased.
But if I'm not mistaken, New Haven's testing was the same used for New York, and featured a number of questions about sky scrapers and helicopter rescue, which they don't have in New Haven, and you wouldn't know from on the job training, or particularly need to know. And who did that benefit? The fire buffs.
A biased test is not a good indicator of who is the best.
There is no evidence that the test was biased.
Except whites passed the test at twice the rate of blacks and no nonwhite made the top 30. Statisticly this would not have happend in a non-biased test between equaly quailified test subjects.
AND BEFORE YOU SPEAK: There is no evidence that Blacks are less quailified then Whites, or less determined to succede, or less willing to study.
Ricci is the case every one mentions, with all the things he did to pass the test. Nobody mentions what the other 19 did to pass, or what the black firefighters did to pass. Cherrypicking Ricci himself is not an good argument.
As one Hispanic quoted anonymously by the New Haven Independent put it, the test favored "fire buffs"—guys who read fire-suppression manuals on their downtime and paid test-manual writers to come to New Haven to speak.
So it would seem that there is in fact evidence that the white firefighters went to greater lengths to prepare for the test.
If that quote is true, then the test is, in fact, probably heavily biased. One common way of biasing a test towards a particular group is by testing trivia which is technically commonly available, but would only be reasonably known by insiders. Minutiae about fire-suppression manuals would be a good example of that.
Now, I'm not saying the test is intentionally biased or biased against non-white test takers, but it is biased against people who do not have a history with firefighting. Given that the test was probably written by one of the firefighters, my bet is that he was just writing questions based on what he knew, which included a lot of minutiae from fire-suppression manuals, so it probably wasn't intentional. However, it is still biased, which is a problem.
But if I'm not mistaken, New Haven's testing was the same used for New York, and featured a number of questions about sky scrapers and helicopter rescue, which they don't have in New Haven, and you wouldn't know from on the job training, or particularly need to know. And who did that benefit? The fire buffs.
Actually it was specificaly made for New Haven by a third party independet contractor. To prevent such bias. The test was not vetted by anyone in the New Haven FD to prevent leaks, it was only vetted by one FD guy from Atlanta.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Right, but that's not actually evidence that they are any better at their jobs, just that they are better at passing the test.
And the test, by definition, is measuring who is more qualified for the position. No one had a problem with the proposed test before hand. Not the city, not the union, no one. So clearly everyone was in agreement before hand that this was an unbiased way to determine who would be promoted. And then they didn't like the results.
You can argue that the test is not the best way to decide promotions because it isn't really testing a lot of things that are important for the job, however I don't see you proposing an alternative method of evaluation. If you try and measure soft attributes like "leadership qualities" or "communication skills" the results will just be even more biased.
Exept there was not one magic lecture that made a difference, it might have been a massive systemic bias.
If one lecture was the difference between passing and failing, it would have been proof positive of a flawed and biased test, unsuited to determine promtions.
Nice try, but no cigar.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Exept there was not one magic lecture that made a difference, it might have been a massive systemic bias.
If one lecture was the difference between passing and failing, it would have been proof positive of a flawed and biased test, unsuited to determine promtions.
Nice try, but no cigar.
The black figherfighters didn't fail. They just got somewhat lower scores and because there was a limited number of openings it didn't get to them.
If all of these high-scoring white firefighters were following the same study regimen and pooling their resources to set up extra lectures and whatever else, that could easily explain the results.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
Exept there was not one magic lecture that made a difference, it might have been a massive systemic bias.
If one lecture was the difference between passing and failing, it would have been proof positive of a flawed and biased test, unsuited to determine promtions.
Nice try, but no cigar.
The black figherfighters didn't fail. They just got somewhat lower scores and because there was a limited number of openings it didn't get to them.
If all of these high-scoring white firefighters were following the same study regimen and pooling their resources to set up extra lectures and whatever else, that could easily explain the results.
Once again, Whites passed at twice the rate of blacks, as in proportionaly twice the percentage of whites passed as blacks.
Plus there has been no mention of white firefighters pooling their resources and since it would have helped their case a lot I find your suggestion strange. "We got together and studied, we invited the blacks to join but they said no" would have stoped this in its tracks
If on the other hand the white Firefighters got extra lectures given to them by senior staff(all white), while the blacks did not. It would have been a clear cut case of discrimination and abuse of department resources. By the way I don't claim that, just pointing out the flaws in your case.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
For the 118 firemen who took the exams, the pass rate for black candidates was approximately half that of the corresponding rate for white candidates:[5]
The passage rate for the Captain exam was: 16 (64%) of the 25 whites; 3 (38%) of the 8 blacks; 3 (38%) of the 8 Hispanics[6]. The top 9 scorers included 7 whites and 2 Hispanics; given that there were 7 Captain vacancies when the tests were administered, and that the "Rule of Three" in the City Charter mandates that a civil service position be filled from among the three individuals with the highest scores on the exam, it appeared that no blacks and at most two Hispanics would be eligible for promotion.
The passage rate for the Lieutenant exam was: 25 (58%) of the 43 whites; 6 (32%) of the 19 blacks; 3 (20%) of the 15 Hispanics. All the top 10 scorers were white; given that there were 8 vacancies, under the "Rule of Three" it appeared that no blacks or Hispanics would be eligible for promotion.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Unless you can show the mechanism by which the test was biased, statistics doesn't prove anything.
Statisics are the mechanism with wich one can prove a test was biased.
Anecdotes and claims of cultural superiority are what is lacking in proof.
The odds of not a single Black person reaching the top 20 scorers in an unbiased test are probably more remote then the odds of the test being biased.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
0
Options
AegisFear My DanceOvershot Toronto, Landed in OttawaRegistered Userregular
edited July 2009
I'm wary of using statistics to indicate whether or not this test was biased when we only have one data point to look at. Sure, the anomaly provides incentive to look into the test itself to see if there's a more systematic effect, but the result by itself doesn't say anything either way in terms of conclusiveness.
I'm wary of using statistics to indicate whether or not this test was biased when we only have one data point to look at. Sure, the anomaly provides incentive to look into the test itself to see if there's a more systematic effect, but the result by itself doesn't say anything either way in terms of conclusiveness.
Oh, I agree completly, The test result is not conclusive and the town should have look into it. I said as much in my post up there. I for one do not suspect malicous intent on any party.
However people are claiming Anecdotes trump Statistics and that to look further is robbing people who succeded at the test. Good faith effort by the white firefighters do not allow them to benefit from a suposed bias. Removing promotions in case of bias is not bad faith.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Unless you can show the mechanism by which the test was biased, statistics doesn't prove anything.
Statisics are the mechanism with wich one can prove a test was biased.
No, because you have done nothing to show causality. Unless they did better on the test for no other reason that they are white (and not that they live in the suburbs, or because they got a better education in high school, or any one of dozens of possible explanations) it's not racial bias.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
Unless you can show the mechanism by which the test was biased, statistics doesn't prove anything.
Statisics are the mechanism with wich one can prove a test was biased.
No, because you have done nothing to show causality. Unless they did better on the test for no other reason that they are white (and not that they live in the suburbs, or because they got a better education in high school, or any one of dozens of possible explanations) it's not racial bias.
OK, STOP IT RIGHT THERE. You have been cheerypicking statments all night and it stops right here. You seem to skip right over what I said about odds. Using the statisics given to us by the test, combining it with the personel files of the GODAMN NEW HAVEN FIRE DEPARTMENT, you can calculate the odds of any set of circumstanes you want.
Want to calculate the odds of a White suburan multigenerational Firefighter passing compared to a Black Urban first generation Firefighter? Take their background, compare it to the test and calculate their chances of passing. The answer is in the statisics, NOT IN YOUR FUCKING ANECDOTES. T
STATISICS CAN PROVE IF THERE IS CAUSALITY YOUR ANECDOTES CANNOT.
Oh and before you answer; I don't know if the test was biased, neither do you, no matter what you claim. I am open to the fact that it might be, because of the strange result. You seem to close of any thought bias based on your weird-ass anecdotes.
In fact all your anecdotes seem to revolve around one subject: "white firefighters are culturaly superior to black ones" wich in itself is a significant sign of bias.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Unless you can show the mechanism by which the test was biased, statistics doesn't prove anything.
Statisics are the mechanism with wich one can prove a test was biased.
Anecdotes and claims of cultural superiority are what is lacking in proof.
The odds of not a single Black person reaching the top 20 scorers in an unbiased test are probably more remote then the odds of the test being biased.
Actually, I mostly agree with you, but I want to be highly technical on one point because statistics is a very tricky business. You can never prove anything with statistics; you can only fail to reject the hypothesis. In this case, the hypothesis is that the test is biased. Statistics can either allow you to reject the hypothesis or fail to reject the hypothesis. There is a subtle, but very important distinction between failing to reject the hypothesis and accepting the hypothesis.
The statistical anomalies that you mention don't prove anything, but rather indicate a problem. In order to prove the hypothesis, you need to do a case study.
Oh and before you answer; I don't know if the test was biased, neither do you, no matter what you claim. I am open to the fact that it might be, because of the strange result. You seem to close of any thought bias based on your weird-ass anecdotes.
I think that from a legal standpoint you should assume the test isn't biased unless you can point to some feature of the test that would result in discrimination based on race. And proof of discrimination based on race is the only thing that could justify throwing out the results of the test.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
Kakaos, you are completly right, statistics have flaws and a case study would have been needed. Wich is what I have argued for most of the thread. Of course if they had signed off on the Test because "Wah, wah, whites deserve promotions" without doing so they would have been just as wrong as if they threw it out.
Legal all you need do is show impact of the test to warant an investigation, wich could have uncovered bias or proved it unfounded.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
0
Options
AegisFear My DanceOvershot Toronto, Landed in OttawaRegistered Userregular
edited July 2009
I'm confused as to the continued debate on this beyond discussing the actual ruling itself. We've already agreed upon the fact that no statistical analysis is going to prove whether or not the test is biased. As such, some people are using falling on the assumption the test is biased unless shown otherwise, while others are rested on the assumption that the test is unbiased unless shown otherwise. Neither side is inherently wrong, considering both are using the initial assumptions to then form an opinion to probe whether or not they're correct based upon the case at hand (to which we don't have an answer since it's sent back to the lower courts).
Considering lack of conclusive proof either way, why is there a heated debate back and forth?
Kakaos, you are completly right, statistics have flaws and a case study would have been needed. Wich is what I have argued for most of the thread. Of course if they had signed off on the Test because "Wah, wah, whites deserve promotions" without doing so they would have been just as wrong as if they threw it out.
Legal all you need do is show impact of the test to warant an investigation, wich could have uncovered bias or proved it unfounded.
Which is the standard this ruling establishes, unlike previously where no investigation was actually necessary. So... we agree that SCOTUS ruled correctly on this one?
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
Kakaos, you are completly right, statistics have flaws and a case study would have been needed. Wich is what I have argued for most of the thread. Of course if they had signed off on the Test because "Wah, wah, whites deserve promotions" without doing so they would have been just as wrong as if they threw it out.
Legal all you need do is show impact of the test to warant an investigation, wich could have uncovered bias or proved it unfounded.
Which is the standard this ruling establishes, unlike previously where no investigation was actually necessary. So... we agree that SCOTUS ruled correctly on this one?
No dont agree, you have argued that the test shouldn't have been thrown out because the white firefighters took the test in good faith.
I have argued that i shouldn't have been thrown out because no qualified investigation was carried out.
Same action, different reasons.
I honestly can't tell if the ruling mandates that the city should have had an investigation.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Kakaos, you are completly right, statistics have flaws and a case study would have been needed. Wich is what I have argued for most of the thread. Of course if they had signed off on the Test because "Wah, wah, whites deserve promotions" without doing so they would have been just as wrong as if they threw it out.
Legal all you need do is show impact of the test to warant an investigation, wich could have uncovered bias or proved it unfounded.
Which is the standard this ruling establishes, unlike previously where no investigation was actually necessary. So... we agree that SCOTUS ruled correctly on this one?
No dont agree, you have argued that the test shouldn't have been thrown out because the white firefighters took the test in good faith.
I have argued that i shouldn't have been thrown out because no qualified investigation was carried out.
Same action, different reasons.
I honestly can't tell if the ruling mandates that the city should have had an investigation.
I have been arguing that there should have been a proper investigation since the beginning of this thread.
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
My solution would be to fix it for next time. The easiest and cheapest solution would be the one presented in the Slate article: take everyone who gets above a certain score and just randomly select however many people you need. A more expensive solution would be to have the department offer a free preparation class for the test, so that everyone gets the same amount of resources going into it.
That first solution is exactly what they did.
Not according to what I've read. They had like 13 spots open which went to the top 13 scores on the test. What I am saying is that you take everyone who got like a 90 or more (which from what I understand included plenty of black candidates) and then lottery it.
Read the ruling. They used something referred to as the "Rule of Three" to look at the top performers. They considered banding, where you basically round the test results. (This is why I asked what is wrong with letter grades, essentially.) Using this method, they would have still looked at the "top 13" scores, but would have had more people to look at, and this would have included some blacks. However, this would have been a violation, as well.
A biased test is not a good indicator of who is the best.
There is no evidence that the test was biased.
Except whites passed the test at twice the rate of blacks and no nonwhite made the top 30. Statisticly this would not have happend in a non-biased test between equaly quailified test subjects.
AND BEFORE YOU SPEAK: There is no evidence that Blacks are less quailified then Whites, or less determined to succede, or less willing to study.
Ricci is the case every one mentions, with all the things he did to pass the test. Nobody mentions what the other 19 did to pass, or what the black firefighters did to pass. Cherrypicking Ricci himself is not an good argument.
Actually, statistically this could happen in a non-biased test. It might be statistically improbable, but it could happen.
And, I am amused that nobody has harped on what I mentioned earlier. What this is a much stronger indicator of, to me, is that that city does a lousy job of making sure that the minorities are trained well enough that they can pass the tests. Failing the tests at this level can be indicative of bias at levels other than on the test.
ChanusHarbinger of the Spicy Rooster ApocalypseThe Flames of a Thousand Collapsed StarsRegistered Userregular
edited July 2009
Oh, for fuck's sake.
"Ricci is invariably painted as a reluctant standard-bearer; a hardworking man driven to litigation only when his dreams of promotion were shattered by a system that persecutes white men."
Right. Because clearly not getting hired because you have a disability is okay. And something about whistle-blowing.
If anything, this just proves that he's smart. We have a system in the US that is based around litigation. Not using that system would be retarded.
Although, I don't get why they're calling him to testify at Sotomayor's confirmation hearings. What the hell kind of relevance does this have to anything?
HamHamJ on
While racing light mechs, your Urbanmech comes in second place, but only because it ran out of ammo.
0
Options
ChanusHarbinger of the Spicy Rooster ApocalypseThe Flames of a Thousand Collapsed StarsRegistered Userregular
Right. Because clearly not getting hired because you have a disability is okay. And something about whistle-blowing.
If anything, this just proves that he's smart. We have a system in the US that is based around litigation. Not using that system would be retarded.
Although, I don't get why they're calling him to testify at Sotomayor's confirmation hearings. What the hell kind of relevance does this have to anything?
I actually understand the motivation for calling him to testify. The point is to show that she puts race politics above the rule of law (not saying I agree with that point, though), using his case as an example.
The whole Joe-the-Plumber-ism thing they're trying to do with him is exceptionally dumb, however.
Right. Because clearly not getting hired because you have a disability is okay. And something about whistle-blowing.
If anything, this just proves that he's smart. We have a system in the US that is based around litigation. Not using that system would be retarded.
Although, I don't get why they're calling him to testify at Sotomayor's confirmation hearings. What the hell kind of relevance does this have to anything?
Or he could be the kind of fucktard that sue for whiplash after a 5 mph fenderbender. And the Whistleblower case seem to have been bullshit.
Kipling217 on
The sky was full of stars, every star an exploding ship. One of ours.
Posts
QEDMF xbl: PantsB G+
It is not so ludicrous to suggest that none of them were.
Right, but that's not actually evidence that they are any better at their jobs, just that they are better at passing the test. What the test is testing for is another matter, like in Britain in the past, grammar school entrance exams tested how "clever" the students were by asking things like naming classical music composers, which had the result of discriminating in favour of middle class people. So in this fire dept case, your black guy and your white guy might be equally "good" at the job itself, but just when they got to a test some of the white guys like Ricci would have been better able to answer questions about naming every part of a fire hose, or saying when the New Haven fire dept was founded or something.
Though not as ludicrous to suggest that all 30 white firefighters where "Fire buffs" and not a single black firefighter was.
If that quote is true, then the test is, in fact, probably heavily biased. One common way of biasing a test towards a particular group is by testing trivia which is technically commonly available, but would only be reasonably known by insiders. Minutiae about fire-suppression manuals would be a good example of that.
Now, I'm not saying the test is intentionally biased or biased against non-white test takers, but it is biased against people who do not have a history with firefighting. Given that the test was probably written by one of the firefighters, my bet is that he was just writing questions based on what he knew, which included a lot of minutiae from fire-suppression manuals, so it probably wasn't intentional. However, it is still biased, which is a problem.
Actually it was specificaly made for New Haven by a third party independet contractor. To prevent such bias. The test was not vetted by anyone in the New Haven FD to prevent leaks, it was only vetted by one FD guy from Atlanta.
And the test, by definition, is measuring who is more qualified for the position. No one had a problem with the proposed test before hand. Not the city, not the union, no one. So clearly everyone was in agreement before hand that this was an unbiased way to determine who would be promoted. And then they didn't like the results.
You can argue that the test is not the best way to decide promotions because it isn't really testing a lot of things that are important for the job, however I don't see you proposing an alternative method of evaluation. If you try and measure soft attributes like "leadership qualities" or "communication skills" the results will just be even more biased.
If all 30 of them attended that same lecture, it's not even unlikely.
If one lecture was the difference between passing and failing, it would have been proof positive of a flawed and biased test, unsuited to determine promtions.
Nice try, but no cigar.
The black figherfighters didn't fail. They just got somewhat lower scores and because there was a limited number of openings it didn't get to them.
If all of these high-scoring white firefighters were following the same study regimen and pooling their resources to set up extra lectures and whatever else, that could easily explain the results.
Once again, Whites passed at twice the rate of blacks, as in proportionaly twice the percentage of whites passed as blacks.
Plus there has been no mention of white firefighters pooling their resources and since it would have helped their case a lot I find your suggestion strange. "We got together and studied, we invited the blacks to join but they said no" would have stoped this in its tracks
If on the other hand the white Firefighters got extra lectures given to them by senior staff(all white), while the blacks did not. It would have been a clear cut case of discrimination and abuse of department resources. By the way I don't claim that, just pointing out the flaws in your case.
For the 118 firemen who took the exams, the pass rate for black candidates was approximately half that of the corresponding rate for white candidates:[5]
The passage rate for the Captain exam was: 16 (64%) of the 25 whites; 3 (38%) of the 8 blacks; 3 (38%) of the 8 Hispanics[6]. The top 9 scorers included 7 whites and 2 Hispanics; given that there were 7 Captain vacancies when the tests were administered, and that the "Rule of Three" in the City Charter mandates that a civil service position be filled from among the three individuals with the highest scores on the exam, it appeared that no blacks and at most two Hispanics would be eligible for promotion.
The passage rate for the Lieutenant exam was: 25 (58%) of the 43 whites; 6 (32%) of the 19 blacks; 3 (20%) of the 15 Hispanics. All the top 10 scorers were white; given that there were 8 vacancies, under the "Rule of Three" it appeared that no blacks or Hispanics would be eligible for promotion.
Statisics are the mechanism with wich one can prove a test was biased.
Anecdotes and claims of cultural superiority are what is lacking in proof.
The odds of not a single Black person reaching the top 20 scorers in an unbiased test are probably more remote then the odds of the test being biased.
Currently DMing: None
Characters
[5e] Dural Melairkyn - AC 18 | HP 40 | Melee +5/1d8+3 | Spell +4/DC 12
Oh, I agree completly, The test result is not conclusive and the town should have look into it. I said as much in my post up there. I for one do not suspect malicous intent on any party.
However people are claiming Anecdotes trump Statistics and that to look further is robbing people who succeded at the test. Good faith effort by the white firefighters do not allow them to benefit from a suposed bias. Removing promotions in case of bias is not bad faith.
No, because you have done nothing to show causality. Unless they did better on the test for no other reason that they are white (and not that they live in the suburbs, or because they got a better education in high school, or any one of dozens of possible explanations) it's not racial bias.
OK, STOP IT RIGHT THERE. You have been cheerypicking statments all night and it stops right here. You seem to skip right over what I said about odds. Using the statisics given to us by the test, combining it with the personel files of the GODAMN NEW HAVEN FIRE DEPARTMENT, you can calculate the odds of any set of circumstanes you want.
Want to calculate the odds of a White suburan multigenerational Firefighter passing compared to a Black Urban first generation Firefighter? Take their background, compare it to the test and calculate their chances of passing. The answer is in the statisics, NOT IN YOUR FUCKING ANECDOTES. T
STATISICS CAN PROVE IF THERE IS CAUSALITY YOUR ANECDOTES CANNOT.
Oh and before you answer; I don't know if the test was biased, neither do you, no matter what you claim. I am open to the fact that it might be, because of the strange result. You seem to close of any thought bias based on your weird-ass anecdotes.
In fact all your anecdotes seem to revolve around one subject: "white firefighters are culturaly superior to black ones" wich in itself is a significant sign of bias.
Actually, I mostly agree with you, but I want to be highly technical on one point because statistics is a very tricky business. You can never prove anything with statistics; you can only fail to reject the hypothesis. In this case, the hypothesis is that the test is biased. Statistics can either allow you to reject the hypothesis or fail to reject the hypothesis. There is a subtle, but very important distinction between failing to reject the hypothesis and accepting the hypothesis.
The statistical anomalies that you mention don't prove anything, but rather indicate a problem. In order to prove the hypothesis, you need to do a case study.
I think that from a legal standpoint you should assume the test isn't biased unless you can point to some feature of the test that would result in discrimination based on race. And proof of discrimination based on race is the only thing that could justify throwing out the results of the test.
Legal all you need do is show impact of the test to warant an investigation, wich could have uncovered bias or proved it unfounded.
Considering lack of conclusive proof either way, why is there a heated debate back and forth?
Currently DMing: None
Characters
[5e] Dural Melairkyn - AC 18 | HP 40 | Melee +5/1d8+3 | Spell +4/DC 12
Which is the standard this ruling establishes, unlike previously where no investigation was actually necessary. So... we agree that SCOTUS ruled correctly on this one?
And just for the records the town acted badly, but legaly according to the common understanding of the law at the time.
SCOTUS declared the common understanding flawed and did so according to political lines. 5-4 with the conservatives in the majority.
I don't think its going to last though, Kennedy wrote the majority opinion and he is the middle man on the court.
The town can delay certefication by opening a new investigation into the test, but they probably wont.
No dont agree, you have argued that the test shouldn't have been thrown out because the white firefighters took the test in good faith.
I have argued that i shouldn't have been thrown out because no qualified investigation was carried out.
Same action, different reasons.
I honestly can't tell if the ruling mandates that the city should have had an investigation.
I have been arguing that there should have been a proper investigation since the beginning of this thread.
Read the ruling. They used something referred to as the "Rule of Three" to look at the top performers. They considered banding, where you basically round the test results. (This is why I asked what is wrong with letter grades, essentially.) Using this method, they would have still looked at the "top 13" scores, but would have had more people to look at, and this would have included some blacks. However, this would have been a violation, as well.
Actually, statistically this could happen in a non-biased test. It might be statistically improbable, but it could happen.
And, I am amused that nobody has harped on what I mentioned earlier. What this is a much stronger indicator of, to me, is that that city does a lousy job of making sure that the minorities are trained well enough that they can pass the tests. Failing the tests at this level can be indicative of bias at levels other than on the test.
"Ricci is invariably painted as a reluctant standard-bearer; a hardworking man driven to litigation only when his dreams of promotion were shattered by a system that persecutes white men."
Statistics show correlation; they never prove causality.
Ouch, yes your right. sorry everyone.
Right. Because clearly not getting hired because you have a disability is okay. And something about whistle-blowing.
If anything, this just proves that he's smart. We have a system in the US that is based around litigation. Not using that system would be retarded.
Although, I don't get why they're calling him to testify at Sotomayor's confirmation hearings. What the hell kind of relevance does this have to anything?
I actually understand the motivation for calling him to testify. The point is to show that she puts race politics above the rule of law (not saying I agree with that point, though), using his case as an example.
The whole Joe-the-Plumber-ism thing they're trying to do with him is exceptionally dumb, however.
Or he could be the kind of fucktard that sue for whiplash after a 5 mph fenderbender. And the Whistleblower case seem to have been bullshit.