Reverse Discrimination Case

[jillio]

I'm sorry but the case at State Supreme Court isn't finished. The decision will be out this summer. Where in the world did you get the info about Supreme Court dismissing the case? Did you have an insider information at Supreme Court or something?

furthermore.......

U.S. Supreme Court Hears New Haven Firefighters' Reverse Discrimination Arguments


simple - this case is nothing but a farce. just one of your typical Al Sharpton-like rage.

From your own post, Jiro. A case is not heard at the U.S. Supreme Court Level unless it has already been heard at the local level, and then the state Supreme Court Level.

No, this is simply a case of your speaking out with attempted authority about things on which your information is less than adequate.

 

[jillio]

You guys do realize that the "rule of three" means that they can choose from the top 3 scorers and promote them to the top positions? So it's not JUST eligibility, but they also take the top scorers to promote. Unfortunately, even with multiple positions, the black people can't even make the cut.

I just put the test results in an excel spreadsheet. 50-50 would STILL not get any blacks promoted. If the oral is 60% and the written is 40% instead, they MIGHT get promoted for lieutenant only . Fair or not? It's weird how arbitrary weights on a test can make such a difference.

Of course I realize it, and that just goes back to what I have been saying all along. Due to weighting, etc. the test is not cross culturally valid. BTW, I am in the process of running stats on the scores you provided.

 

[Jiro]

From your own post, Jiro. A case is not heard at the U.S. Supreme Court Level unless it has already been heard at the local level, and then the state Supreme Court Level.

then you need to re-read my post carefully and slowly. I CLEARLY said "I'm looking forward to hearing the decision from higher court..."

Where in the world did I ever said that this case is being heard at U.S. Supreme Court?

 

[jillio]

you seem to think that it's my first work experience and it also seems that you have a rather unhealthy fascination about my job position. Are you just jealous that my salary is higher than yours? To answer your silly question.... for over 10 years, I've worked for small businesses including start-up dot.com, major pharmaceutical company, major hospital, universities, and couple of freelancer jobs.... and it's not always a webmaster job.

And in all those jobs, you were no doubt at the bottom of the heirarchy. I doubt seriously that your salary tops mine. I know what IT average salary is at university. And I might remind you, you are the one that brought the topic up. If you don't want your job discussed, don't bring the topic up in regard to me. Hmm...5 jobs in 10 years? Not much for longevity, are you? Guess you didn't make it very far up the heirarchy in that amount of time.*hint - my "webmaster" job is a senior position. yes I serve & speak & function as adviser/consultant for this university.* Coincidentally, this university is known for business/emergency management and counseling service so yes I think I know a thing or two about this a bit better than you (except counseling service ).* so... are you telling me that a rape counselor like you know this better than me?

I am not a "rape counselor", Jiro. You are again assuming. Shame on you. And given that I have had to take numerous research methodology, assessment and testing, and statistics courses to obtain an advanced degree in my field, am licensed to do both testing and assessment, and to compile data and statistically interpret it, yeah, I'd say my expertise far outweighs yours when it comes to this topic.


and that's why I said it's very apparent in your inability to understand this "bureaucracy" politic... simply because of your simplistic comment "hierarchy is hierarchy"

Heirarchy is heirarchy, Jiro. That is why it is called "heriarchy" independent of the arrangement. Duh. Otherwise, it would be called something other than "heirarchy". And again, that has absolutely nothing to do with the topic of this thread.

 

[jillio]

I was an engineering major at Virginia Tech for a while. You know what's funny? A engineering class or club with 1-2 pretty girl tends to fare better (better attendance record, better testing score, better mood) than a class with no girls.

What happened? Poor test scores?

 

[Jiro]

And in all those jobs, you were no doubt at the bottom of the heirarchy. I doubt seriously that your salary tops mine. I know what IT average salary is at university. And I might remind you, you are the one that brought the topic up. If you don't want your job discussed, don't bring the topic up in regard to me. Hmm...5 jobs in 10 years? Not much for longevity, are you? Guess you didn't make it very far up the heirarchy in that amount of time.

Actually - you asked for my work experience. and 5 jobs? Who said I had 5 jobs? You are again assuming as well. tsk tsk tsk

I am not a "rape counselor", Jiro. You are again assuming. Shame on you. And given that I have had to take numerous research methodology, assessment and testing, and statistics courses to obtain an advanced degree in my field, am licensed to do both testing and assessment, and to compile data and statistically interpret it, yeah, I'd say my expertise far outweighs yours when it comes to this topic.

hmmmm.... work experience... or lab experience... This reminds me of the case in appointing Federal Reserve Chairman- an over-experienced person or an over-educated person. I'm sorry but I didn't realize that this fireman job regarding appointing captain requires an advanced degree, statistics courses, research methodology, and such.

Heirarch is heirarchy, Jiro. That is why it is called "heriarchy" independent of the arrangement. Duh. Otherwise, it would be called something other than "heirarchy". And again, that has absolutely nothing to do with the topic of this thread.

actually it does. Your lack of understanding on how this bureaucracy works is quite apparent in your inability to understand this issue. Again - the test score is not the deciding factor in choosing the captain - something that your advanced degree in statistic course and such does not cover.

 

[jillio]

that's why the test is not a deciding factor. On the book, it may be so but in office politic - it's not. If the world revolves around the idea where the nominee is picked based on test score.... then GWB wouldn't be our President, eh? McCain wouldn't be a US Navy Captain either due to his underachieving status at Naval Academy

Those are totally fallicious examples, and only indicate your lack of understanding. And if test takers were told that the deciding factor was test scores, which is obviously what they were told given the lawsuit, and test scores were not the deciding factor, this fire dept is open to many more lawsuits than this one. You don't even grasp the implication of what you are trying to intimate.

 

[Jiro]

What happened? Poor test scores?

probably. and a low attendance record as well. just my little observation.

funny thing is... I have recalled reading articles about in K-12 setting, a single-sex class fares better than co-ed class.

 

[darkdog]

A student t is a converted score. The t score is determined by the z score. The z score is determined by the raw score. T scores and Z scores are simply formulas for raw scores that allows us to see where a score falls on a normal distribution. If all the scores have been plotted, it is obvious whether a distribution falls on a normal curve, or whether the distribution is skewed either positively or negatively. Raw scores are basically useless for determining distribution. That is why they are converted to t-scores and z-scores.

What if it does have 2 modes? That doesn't affect distribution. A normal distribution or a skewed distribution can easily have 2 modes.

A score doesn't need to fall at zero in order to skew a distribution. Few scores fall at exact zero. The far end positively or negatively of any distribution (those scores falling outside 3 SDs) will represent only .5% of any population. A majority of scores falling +/- 2 SDs from the mean will skew a distribution.

The Gaussian curve simply refers to scores plotted on a histogram that will approximate a normal distritubtion. This theory is based on the Quincunx Board developed by Galton, and is based on natural laws of the universe and explained through mathematical laws of probability. It is not a different distribution, but simply another way of demonstrating a normal distribution, and also of explaining why a skewed distibution means one needs to investigate the instrument to determine the reason for the skewness. In any normal distribution, the specific area +1SD to-1SD will include 68% of any population. +2SD to -2SD will include 95% of any population. 99.5% of the population will fall withing 3 SDs of the mean in any normal curve. These percentiles do not change on any normal distribution no matter if it is plotted using a histogram, a scatter plot, or a bell curve.

Normal curves and skewd curves are frequency distributions. Both the normal, the positively skewed, and the negatively skewed distribution identifies where an individual's score falls relative to the rest of the group. The mean is the most commonly used measure of central tendency and is determined by dividing the sum of X by N. Standard deviation, along with range, and interquartile range, is a measure of variability, and it is used to describe how scores vary around the mean in a standard fashion. The formulas for determining the SD is as follows:the square root of the sum of X-M squared divided by N.

Thus concludes your introductory statistics lesson for the day.

So you get the z-scores and the t-scores for the various sup-populations and then what? Is there a specific formula that you use to say it's valid or invalid? Because the odds of it perfectly fitting a normal curve are pretty low, especially with a small population. It sounds to me like you're saying it will involve comparing the cdf of a perfectly normal curve versus the cdf of the different groups and making sure the percentages within 1 SD, 2 SD, 3 SD are closely aligned. Is that right? If so, how close?

By the way, I'm aware that a Gaussian distribution and a normal distribution are the exact same thing. You lost me where you said that a Gaussian distribution "...will approximate a normal distritubtion". If they're the exact same thing, how can one be an approximation of the other?

A normal distribution cannot "easily have 2 modes" as you put it. If it does have two modes, it ceases to be a normal distribution by definition. You could have a distribution which is the superposition of 2 normal distributions with different means, but the distribution itself will not be normal.

Let's get our terms straight here. Skew, like mean, variance, etc. is a parameter of a distribution, not a distribution itself. There are a lot of different distributions: normal, Poisson, binomial, student-t, chi-square, Rayleigh, Laplacian, uniform, gamma, and on and on. They all have different shapes and different probability density functions. Some have skew and others don't. I'm guessing you mean the skew normal distribution, which is basically a normal distribution with some extra terms to throw in skew. It is not an actual normal distribution, but it's similar.

 

[Jiro]

Those are totally fallicious examples, and only indicate your lack of understanding. And if test takers were told that the deciding factor was test scores, which is obviously what they were told given the lawsuit, and test scores were not the deciding factor, this fire dept is open to many more lawsuits than this one. You don't even grasp the implication of what you are trying to intimate.

just because ADA law requires disabled people to have an equal opportunity as able-people doesn't mean the employment is required to provide the accommodation that they want. Same concept.

the real world just doesn't work the way as you'd think. oh well. it's ok. <pat on your head>

 

[lsfoster]

Of course I realize it, and that just goes back to what I have been saying all along. Due to weighting, etc. the test is not cross culturally valid. BTW, I am in the process of running stats on the scores you provided.

And yet you still haven't provided any reason that race should be the only "cultural" issue considered when comparing the scores. You have also not explained what types of cultural differences there would be between a group of men who most likely grew up and were taught in the same area and schools that would make this test so unfair only to the minorities taking it. You have also not explained how all of your "experience" with determining test validity and reliability does not make you question the idea of judging the validity of something based solely on the race of the people in question, ignoring any and all other factors. You also have yet to admit that without seeing the test itself, or having any knowledge of firefighting, the important qualities required for promotion, or what was tested on this particular exam, you are doing nothing more than agreeing with the opinion of some group somewhere, despite the fact that their decision could be just as easily biased as anything else in this case.

You are also still ignoring the fact that the timeline does imply discrimination against the white firefighters, since the results were thrown out based simply on race, and not that the city thought the test was invalid.

 

[jillio]

So you get the z-scores and the t-scores for the various sup-populations and then what? Is there a specific formula that you use to say it's valid or invalid? Because the odds of it perfectly fitting a normal curve are pretty low, especially with a small population. It sounds to me like you're saying it will involve comparing the cdf of a perfectly normal curve versus the cdf of the different groups and making sure the percentages within 1 SD, 2 SD, 3 SD are closely aligned. Is that right? If so, how close?

There are several specific formulas to determine such. And you are incorrect. Even a small population can fit a normal curve, if that is the population used for norming. What you are talking about it generalizability, and that is a different concept altogether. I have already explained the percentages of populations that are represented in the various SDs. That gives you the information you are requesting.
\
By the way, I'm aware that a Gaussian distribution and a normal distribution are the exact same thing. You lost me where you said that a Gaussian distribution "...will approximate a normal distritubtion". If they're the exact same thing, how can one be an approximation of the other?

The graph of a Gaussian is a characteristic symmetric "bell curve" shape that quickly falls off towards plus/minus infinity. A normal distribution of test scores does not fall off quickly toward zero. I have already explained this to you. Therefore, it approximates a normal distribution.
A normal distribution cannot "easily have 2 modes" as you put it. If it does have two modes, it ceases to be a normal distribution by definition. You could have a distribution which is the superposition of 2 normal distributions with different means, but the distribution itself will not be normal.

That would depend on the number of the population that is to be considered as falling within the "normal range".

Let's get our terms straight here. Skew, like mean, variance, etc. is a parameter of a distribution, not a distribution itself. There are a lot of different distributions: normal, Poisson, binomial, student-t, chi-square, Rayleigh, Laplacian, uniform, gamma, and on and on. They all have different shapes and different probability density functions. Some have skew and others don't. I'm guessing you mean the skew normal distribution, which is basically a normal distribution with some extra terms to throw in skew. It is not an actual normal distribution, but it's similar.

Sweetie, I have my terms straight. And the skew is not a parameter. Statistical parameters are values, usually unknown, and therefore has to be estimated, used to represent a certain population characteristic. Student-t is not a distribution. It is a converted raw score. Chi-Square is a statistical hypothesis test and can only be used when the null pyhpothesis have found to have been true in previous statistical analysis. We are not even concerned with hypothesis testing in this situation. There is no hypothesis to be tested. We are determining the validity of an assessment. Evidently, your understanding of the rest of the terms you have thrown out is fairly superficial as well. A skewed distribution is not a normal distribution. You need to return to your statistics text book. If a distribution is skewed, it is not normal. That is a very basic concept.

 

[naisho]

So you get the z-scores and the t-scores for the various sup-populations and then what? Is there a specific formula that you use to say it's valid or invalid? Because the odds of it perfectly fitting a normal curve are pretty low, especially with a small population.

This is what I have been attempting to say a few posts back.. It's valid to do statistical analysis of the data to be interpreted depending on what is trying to be observed, but I felt that there is a drawback to conduct this due to various reasons given why.

That's why I believed there has to be a calculation in the error measurement of the data provided to indicate to the observer that there may be inconsistencies in deciphering the overall graph, it gives the impression that the analysis provided might be "skewed" off the path somewhere.

Each time back while researching a specifics to a case on a sociological (macro) scale we always had to give a "margin of error" result. Unless times have changed..

I really have not brushed on this area of expertise for a few years, now I regret it!

 

[jillio]

And yet you still haven't provided any reason that race should be the only "cultural" issue considered when comparing the scores. You have also not explained what types of cultural differences there would be between a group of men who most likely grew up and were taught in the same area and schools that would make this test so unfair only to the minorities taking it. You have also not explained how all of your "experience" with determining test validity and reliability does not make you question the idea of judging the validity of something based solely on the race of the people in question, ignoring any and all other factors. You also have yet to admit that without seeing the test itself, or having any knowledge of firefighting, the important qualities required for promotion, or what was tested on this particular exam, you are doing nothing more than agreeing with the opinion of some group somewhere, despite the fact that their decision could be just as easily biased as anything else in this case.

You are also still ignoring the fact that the timeline does imply discrimination against the white firefighters, since the results were thrown out based simply on race, and not that the city thought the test was invalid.

I never said race should be the only consideration. You are making silly arguments again. As soon as I complete the extensive statistical analysis I am performing on the scores provided, I'll post them. If you plan to refute my point, I suggest you begin doing your own statistical analysis.

 

[Jiro]

I never said race should be the only consideration. You are making silly arguments again.

now you are making silly distortion again it is VERY apparent that the race is your MAJOR consideration.

 

[jillio]

now you are making silly distortion again it is VERY apparent that the race is your MAJOR consideration.

When it comes to cross cultural validity it is a major concern. However, major in no way implies ONLY. And in this case, gender is not an issue, as there were no gender differences in the population taking the test. Duh!

 

[jillio]

This is what I have been attempting to say a few posts back.. It's valid to do statistical analysis of the data to be interpreted depending on what is trying to be observed, but I felt that there is a drawback to conduct this due to various reasons given why.

That's why I believed there has to be a calculation in the error measurement of the data provided to indicate to the observer that there may be inconsistencies in deciphering the overall graph, it gives the impression that the analysis provided might be "skewed" off the path somewhere.

Each time back while researching a specifics to a case on a sociological (macro) scale we always had to give a "margin of error" result. Unless times have changed..

I really have not brushed on this area of expertise for a few years, now I regret it!

Again, the two of you are talking about hypothesis testing and generalizability. Neither one of those concepts apply here. We are determining the cross cultural validity of an insrtument, not hypothesis testing nor generalizing results to an entire population based on a sample.

I think I see where the confusion lies. You are applying incorrect analysis.

 

[lsfoster]

I never said race should be the only consideration. You are making silly arguments again. As soon as I complete the extensive statistical analysis I am performing on the scores provided, I'll post them. If you plan to refute my point, I suggest you begin doing your own statistical analysis.

When it comes to cross cultural validity it is a major concern. However, major in no way implies ONLY. And in this case, gender is not an issue, as there were no gender differences in the population taking the test. Duh!

Who said anything about gender? I'm not sure who you think you're showing up with your clever "Duh!" about something nobody mentioned...
Also, maybe you'd like to share how your "extensive analysis" incorporates anything other than race, since you have no other data available. If it's the only thing you can consider, that would in face make it your "only" consideration.

 

[Jiro]

When it comes to cross cultural validity it is a major concern. However, major in no way implies ONLY. And in this case, gender is not an issue, as there were no gender differences in the population taking the test. Duh!

right.... and this test has abided by federal civil law and the firm was contracted to create a culturally-unbiased test. what's left is the State Supreme Court to decide if the test is a violation of federal law or not....

I guess we better sit tight and let the real experts do the math and studies who actually have tests & data at their disposal. no need for you to do the computations since you've never seen the test... Duh!

 

[jillio]

Who said anything about gender? I'm not sure who you think you're showing up with your clever "Duh!" about something nobody mentioned...
Also, maybe you'd like to share how your "extensive analysis" incorporates anything other than race, since you have no other data available. If it's the only thing you can consider, that would in face make it your "only" consideration.

Nor was I talking to you. I was posting a response to Jiro's comment.

And gender is one of the cultural considerations one needs to consider in determining cross cultural validity. In this case, however, gender is not a consideration, as all the test takers were male.

I will share my analysis as soon as it is finished. Maybe you'd better get to work on your own. And since the white firefighters filed their lawsuit based on race, they have made it a primary issue in this case.