Is the racism against blacks warranted?

[RampantAndroid]

Shens.



Proof?

Did you not follow the supreme court case (that, iirc, encompassed TWO cases) dealing with schools in michigan using race as a qualifier?

 

[fskimospy]

I really don't get how you can say that the companies who had absolutely nothing to say about discrimination are relevant in determining how much discrimination is in the job market. Its ridiculous.

What I am saying is that we have absolutely no evidence that they even saw the resumes or that they weren't mishandled or misplaced or the HR director suddenly died or aliens stole them. In any case IF they saw them they made no judgement whatsoever on the subject of racial discrimination therefore to use them as "beef" in this bullshit broth you're cooking is hilariously ridiculous. That is the beginning and the end of it.
This is why I call you a pompous prick. Who the fuck are you? Why should I accept your definition of "reality" simply because you came down from your throne of bullshit and decided to post in a forum?
Your position is ridiculous.
Of course both sides don't have equal merit, your position is absurd. The companies who had nothing to say about racial discrimination one way or the other should not be considered in any way when determining the level of significance of the data that is supposedly showing racial discrimination.
How can you say its significant (based on this study) when there are only 450 or so resumes that were responded to? Or that only 232 companies made any responses at all? These 232 companies are the only businesses out of 1300+ who had any data on the central question of discrimination. The 1098 who didn't like the "white" OR the "black" are absolutely irrelevant to the question of racial discrimination.

Period.
End of the story.
C'est tout.

The data is about response rates, from which inferences are made. That's the whole point. The study is about discrimination, but the actual data collection is not. It is purely response rates. You want to exclude people from response rate measurement because they didn't respond...which is certainly a novel way of doing things.

You want to exclude people from a study on response rates if they didn't respond to either. This is beyond retarded. You want to do so because you think the Chicago and Boston job markets have been afflicted by mass HR incompetence. That's stupid.

You want to do this because you don't understand what you're talking about and because you have become emotionally invested in not being wrong.

 

[buckshot24]

The data is about response rates, from which inferences are made. That's the whole point. The study is about discrimination, but the actual data collection is not. It is purely response rates. You want to exclude people from response rate measurement because they didn't respond...which is certainly a novel way of doing things.

No, I don't care about call back rates. I care about the amount of discrimination and this study had only 232 companies who made any comment on that subject.

You want to exclude people from a study on response rates if they didn't respond to either.

You just said the study was about discrimination, now its about call back rates?

This is beyond retarded. You want to do so because you think the Chicago and Boston job markets have been afflicted by mass HR incompetence. That's stupid.

They had no input on the level of discrimination in those markets because they didn't call the "white" resumes or the "black" resumes, that is all we know about them. They made no comment about which resumes they preferred. Therefore they shouldn't be used in determining what level of confidence we can have in the level of discrimination in these markets.

 

[fskimospy]

Its a study about discrimination that uses call back rates as its data point, you insane idiot. The level of significance is about the difference in callback rates, not the difference between companies that called back one or the other explicitly. You want to eliminate the businesses that didn't call people back from data about call back rates. This is mind bogglingly stupid.

How do you not know or understand this.

 

[SheHateMe]

Did you not follow the supreme court case (that, iirc, encompassed TWO cases) dealing with schools in michigan using race as a qualifier?

1) That wasn't the argument that was put up. He said that most colleges and universities use the quota system to keep "statistically smart" students out of college.

That is false.

2) Oh wow, two cases in Michigan! That totally proves that most of the Colleges and Universities in this country admit off of a racial quota system (to avoid lawsuits) and that "smart kids can't get into college because we let minorities in first".

 

[buckshot24]

Its a study about discrimination that uses call back rates as its data point, you insane idiot.

You could include letters to Santa in the study and they'd have just as much significance as the companies who made no calls as far as discrimination is concerned.

The level of significance is about the difference in callback rates, not the difference between companies that called back one or the other explicitly.

I've conceded this point multiple times, in fact its one of my criticisms of the study. By going for call back rates instead of actual discrimination they can (in my view) artificially beef up the level of significance of the study as far as discrimination is concerned by including data points that had absolutely nothing to say about discrimination.

You want to eliminate the businesses that didn't call people back from data about call back rates. This is mind bogglingly stupid.

I've said just the opposite. When rates are being employed you need to include the no calls. However when you look at the discriminating companies the data is simply sparse and unconvincingly small.

How do you not know or understand this.

Exactly, how can you not understand this? I want more data where the company implicitly makes their preferences known and those are the companies who made one call or more. The companies who called nobody had nothing whatsoever to say about discrimination therefore they MUST be excluded in ANY calculation that has anything to do with discrimination. The rates angle is just a ruse to make tiny data points into some sort of iron clad statistically significant finding.

The argument being pushed is as if we can tell the significance of the proportion of shots an NBA player makes between two pointers and three pointers by including his total number of shots taken. His misses have nothing to say about the proportion of his makes. They are two totally separate questions.

 

[fskimospy]

Actually this study would be the equivalent of studying what percentage of two pointers he makes and what percentage of 3 pointers he makes and them determining if one is a more difficult shot. Nice work on self ownage.

The rates are what they are. Statistics tells us that the gap in callback rates is extremely unlikely to have happened by chance. Considering the confines of the study, only the difference in names makes sense. This is basic logic. You have successfully made me throw my hands up in frustration at your epic ignorance on this one. You're just going to have to live with being this way I guess.

 

[TheVrolok]

How the fuck is this thread, with the premise in the op, still alive? Christ.

 

[buckshot24]

Actually this study would be the equivalent of studying what percentage of two pointers he makes and what percentage of 3 pointers he makes and them determining if one is a more difficult shot. Nice work on self ownage.

I guess I'll have to take your word for it because I have no idea how that would be equivalent.

The rates are what they are. Statistics tells us that the gap in callback rates is extremely unlikely to have happened by chance. Considering the confines of the study, only the difference in names makes sense.

Which is fine and I DO agree with this but I just don't think the sample of "voters" is large enough to discount a large amount of susceptibility to random variation. I'm sorry if I have frustrated you on this. I'm just not convinced in the way that you and the study are looking at the data is a valid one. It's definitely possible that I am wrong, I just haven't seen anything coming from your posts that would convince me of this possibility.

This is basic logic. You have successfully made me throw my hands up in frustration at your epic ignorance on this one. You're just going to have to live with being this way I guess.

I don't need your help but thanks for trying to meddle in my life. Typical of most liberals I suppose to stick their noses in where it doesn't belong.

 

[fskimospy]

I guess I'll have to take your word for it because I have no idea how that would be equivalent.

Two separate and distinct attempts at accomplishing the same goal with differing rates of success. It's exactly the same.

Which is fine and I DO agree with this but I just don't think the sample of "voters" is large enough to discount a large amount of susceptibility to random variation.

Okay but math says you're wrong.

I don't need your help but thanks for trying to meddle in my life. Typical of most liberals I suppose to stick their noses in where it doesn't belong.

Funny, I would consider someone with zero expertise trying to critique research design 'sticking their nose where it doesn't belong'. I'm guessing this level of hostility to education is what got you to where you are now though.

 

[buckshot24]

Two separate and distinct attempts at accomplishing the same goal with differing rates of success. It's exactly the same.

I was talking about the proportion of 2 pointers vs 3 pointers made. The number of overall attempts has zero effect on that proportion.

Okay but math says you're wrong.

Oh, you're "math" now. You really are a pompous prick aren't you?

Funny, I would consider someone with zero expertise trying to critique research design 'sticking their nose where it doesn't belong'. I'm guessing this level of hostility to education is what got you to where you are now though.

You're the one who brought it to my attention. I didn't go out looking for it. I'm not sticking my nose in their research. You're sticking your nose into my business by telling me to interpret it in only your way. As it stands they seemed to be weighting the very few data points that show racial discrimination with data that shows nothing of the sort, in fact it shows us NOTHING at all.

 

[buckshot24]

How the fuck is this thread, with the premise in the op, still alive? Christ.

Because there are two stubborn fuckers who won't stop posting.()

 

[fskimospy]

I was talking about the proportion of 2 pointers vs 3 pointers made. The number of overall attempts has zero effect on that proportion.

Oh, you're "math" now. You really are a pompous prick aren't you?

You're the one who brought it to my attention. I didn't go out looking for it. I'm not sticking my nose in their research. You're sticking your nose into my business by telling me to interpret it in only your way. As it stands they seemed to be weighting the very few data points that show racial discrimination with data that shows nothing of the sort, in fact it shows us NOTHING at all.

Your NBA analogy just showed that you don't understand the study. Amazing, after all this.

My NBA analogy was an accurate description of it. Hopefully that aids your understanding.

 

[buckshot24]

Your NBA analogy just showed that you don't understand the study. Amazing, after all this.

LOL. I used it to show that the misses (no calls) has no effect on the proportion of the two different kinds of makes (calls to "black" or "white" resumes).

My NBA analogy was an accurate description of it. Hopefully that aids your understanding.

Thanks.

 

[werepossum]

You're the one arguing for legally codified racism in the 21st century.

Not letting the statistically smartest people go to college is really going to help with that.

It helps as they can draw the best thousands from a talent pool of millions/billions, something they are ironically not able to legally do when selecting candidates from within the united states.

Affirmative action is codified discrimination, but not necessarily codified racism. Racism implies that blacks as a race are unable to compete with other races; affirmative action merely recognizes that statistically speaking, blacks are at a disadvantage both in the resources they can command to support their education and in the failed educational systems in which a disproportionate number of black kids are trapped. Affirmative action is neutral as to inherent racial ability, which is why it also benefits Hispanics and other races such as Pacific Islanders who did not necessarily suffer from slavery and/or segregation in significant numbers, but who for whatever reason are statistically under-represented in the particular area be it higher education or basket weavers.

You are correct that denying university (or more properly, denying a particular university in most cases) is a problem both for the individuals and for society at large. Having a permanent underclass, especially one based on race or ethnicity or religion, is also a problem both for the individuals and for society at large. This is why we have affirmative action with significant limitations, to try to best fix the latter with the least societal damage from the former.

It's amazing you can write a post and not stop to see the inconsistency in what you are saying. Why is Joe and Tom more common? Is it more common all over the world? If we were in Ghana, Kweku or Kofi would be very common and thus normal. Here not so much. Because there are less Ghanaians and more Caucasians. That is the definition of white privilege that you still are taking for granted. By virtue of living in a country dominated by Caucasians your views, customs and even names are seen to be in the norm. So, all the inherent human biases come to play in every aspect of life with those who are not in the norm. You get it?

If we were in Ghana, Kweku or Kofi your point would be valid because the views, customs and names of Ghana, Kweku or Kofi would be the norm. There is no possible way to have a functional society without the views, customs and names of the majority being the norm; that's part of the definition of a culture. If we were in Ghana, Kweku or Kofi a resume with a name like Joe or Tom would be outside of the norm and the employer's reaction would be based on his own biases plus his expectations of what kind of person a Joe or a Tom would likely be. As with any prejudgment, some Joes and Toms would benefit from his expectations and some would be penalized, and for some employers the personal biases held about the group to which a Joe or a Tom most likely belongs would overshadow the expectations of what kind of person a Joe or a Tom would likely be. (E.g. if I hate Hawaiians that will likely prevent me from hiring them even if I believe they are the smartest and hardest working people in the world, the only likely exception being if I literally cannot afford not to hire them.)

I like your definition of white privilege though, for in a decade or so white privilege will be replaced with brown privilege. I'll be too old to benefit, but perhaps my grandson can get some affirmative action to offset that brown privilege.

EDIT: Note that in many areas of the nation whites are either already a minority or are a sufficiently small majority that Jose and Tomas are just as much the norm as are Joe or Tom.

 

[werepossum]

Its a study about discrimination that uses call back rates as its data point, you insane idiot. The level of significance is about the difference in callback rates, not the difference between companies that called back one or the other explicitly. You want to eliminate the businesses that didn't call people back from data about call back rates. This is mind bogglingly stupid.

How do you not know or understand this.

Expand it and perhaps you'll see his point. Suppose my test resumes are so bad that I send out 1,000,000 resumes and get only six responses. Should my results' significance then be based on a sample of 1,000,000? If there is only one call back and it's for a "black name", could I really say my results statistically prove white people are unemployable because of my sample size of one million? I'm not saying this is a definitive argument, but it's worth considering.

1) That wasn't the argument that was put up. He said that most colleges and universities use the quota system to keep "statistically smart" students out of college.

That is false.

2) Oh wow, two cases in Michigan! That totally proves that most of the Colleges and Universities in this country admit off of a racial quota system (to avoid lawsuits) and that "smart kids can't get into college because we let minorities in first".

Well, he has a ghost of a point. Most universities definitely use some metrics other than scholastic performance which are specifically designed to get black students at or near a target percentage as a form of affirmative action. Most universities have a limited number of students they can or will accept. Therefore some non-black students are denied who would have been accepted were the admission process colorblind. Sucks to be them. You may justify this by citing white privilege, but if the kid is a Vietnamese or Cambodian refugee who came here with nothing, no possessions and unable to even speak the language, sure white privilege wears a bit thin.

I wouldn't characterize these kids as necessarily smarter because a kid who scores 700 on the SAT coming from a typical war zone inner city school is almost certainly smarter and/or harder working than a kid who scores 700 on the SAT coming from a typical private prep school. It's also worth pointing out that since many of these metrics are based solely on race, a doctor's kid receives exactly the same advantage from them as does a crack head welfare mother's kid. Affirmative action works much the same in contracting, where a black multimillionaire contractor can often get a contract at 10% above a low bid from a non-millionaire white contractor because of "white privilege".

None of this particularly bothers me because any human endeavor is inherently unfair, but it would be nice if affirmative action were limited to those who actually are poor and/or disadvantaged rather than open to anyone who has the complexion for the connection.

 

[fskimospy]

Expand it and perhaps you'll see his point. Suppose my test resumes are so bad that I send out 1,000,000 resumes and get only six responses. Should my results' significance then be based on a sample of 1,000,000? If there is only one call back and it's for a "black name", could I really say my results statistically prove white people are unemployable because of my sample size of one million? I'm not saying this is a definitive argument, but it's worth considering.

That result would not actually be statistically significant, which is of course the point. You would have a T score of about 1. With that in mind your argument doesn't make much sense. (also, there are basic statistical rules against using data with such extremely small returns like 1 out if a million) you have been very reasonable in your questions here, which I appreciate, and there's nothing wrong with raising them.

If you are asking if significance should be based on your sample of one million, it depends on what you were measuring. If you were measuring response rate as this study was, you should absolutely be using the 1 million number. Think about it. Say you got 6 responses with that horrible resume. 5 were white and 1 was black. Not only would that still not be significant at the 95% level, but think of all the other bad conclusions you could make. You might say that 17% of the people who you knew about preferred black people, when really the response was one in a million. Is that a good conclusion to make?

The question fundamentally being asked here was if white sounding people got called back more often than black sounding ones. The answer to that was yes, and the answer was statistically significant. Now the only question is why. Considering the lack of differences outside the names, it's really hard for me to think of an alternate explanation. If someone can come up with one that doesn't involve mass telephone outages across several cities I'm down to hear it.

 

[werepossum]

That result would not actually be statistically significant, which is of course the point. You would have a T score of about 1. With that in mind your argument doesn't make much sense. (also, there are basic statistical rules against using data with such extremely small returns like 1 out if a million) you have been very reasonable in your questions here, which I appreciate, and there's nothing wrong with raising them.

If you are asking if significance should be based on your sample of one million, it depends on what you were measuring. If you were measuring response rate as this study was, you should absolutely be using the 1 million number. Think about it. Say you got 6 responses with that horrible resume. 5 were white and 1 was black. Not only would that still not be significant at the 95% level, but think of all the other bad conclusions you could make. You might say that 17% of the people who you knew about preferred black people, when really the response was one in a million. Is that a good conclusion to make?

The question fundamentally being asked here was if white sounding people got called back more often than black sounding ones. The answer to that was yes, and the answer was statistically significant. Now the only question is why. Considering the lack of differences outside the names, it's really hard for me to think of an alternate explanation. If someone can come up with one that doesn't involve mass telephone outages across several cities I'm down to hear it.

Sounds reasonable. My statistics courses were 30+ years ago and while I still have at least one of the books, I don't have time to relearn it.

As far as whether black-sounding names get called back at a lower rate than do non-black sounding names, I'd count that as one of those things everyone knows.

 

[buckshot24]

If you are asking if significance should be based on your sample of one million, it depends on what you were measuring. If you were measuring response rate as this study was, you should absolutely be using the 1 million number. Think about it. Say you got 6 responses with that horrible resume. 5 were white and 1 was black. Not only would that still not be significant at the 95% level, but think of all the other bad conclusions you could make. You might say that 17% of the people who you knew about preferred black people, when really the response was one in a million. Is that a good conclusion to make?

See, you've basically just restated what I've been saying. Since they were measuring call back rates they can use the 5000 (or 2500 each) resumes for significance because they asked the question in a way where less actual discrimination could be presented as more significant.

I'd like more data points where possible racial discrimination could be determined. The study had only 232 companies where this was possible or 450 resumes (only six in werepossum's 1 million example).

The question fundamentally being asked here was if white sounding people got called back more often than black sounding ones. The answer to that was yes, and the answer was statistically significant. Now the only question is why. Considering the lack of differences outside the names, it's really hard for me to think of an alternate explanation. If someone can come up with one that doesn't involve mass telephone outages across several cities I'm down to hear it.

I think this is a more interesting question than whats been going on for the last few pages so I'll give a few causes for the different rates of call backs.

None of these are being condoned by me.

1. Racist employers
2. Non racist employers who subconsciously prefer people like themselves.
3. The possible belief that firing black employees will bring the charges of racism so as long as there are enough black employees already employed they prefer the less (in their minds) risky employee. This is racism but in a more subtle form.

The real issue is what do we do about it if anything? Is there anything? I don't know if there is anything that can be done that wouldn't make it worse.



The idiot companies who choose the poor quality resume over the good quality resume will get a shittier employee no matter the race. They lose by racially picking one over the other. So their discrimination costs them which is exactly the best way to get rid of it in the long run.


In the meantime as a prospective black employee with a black sounding name, simply send out more resumes. It isn't fair but life has never been fair nor will it ever be fair.

 

[fskimospy]

No buckshot, I was showing why what you were saying was dumb, the only difference was that I showed how subsampling for no reason could lead to bad conclusions as opposed to in your case where pointless subsampling was leading to bad rejection of conclusions. (for illustration's sake it is easier to show the former)

Also, all three possible alternative explanations to racial discrimination in hiring were in fact cases of racial discrimination in hiring.

 

[buckshot24]

No buckshot, I was showing why what you were saying was dumb

Well you failed dipshit.

the only difference was that I showed how subsampling for no reason could lead to bad conclusions as opposed to in your case where pointless subsampling was leading to bad rejection of conclusions. (for illustration's sake it is easier to show the former)

I'm not rejecting anything you miserable git. I'm fairly certain that it is harder for black people to get a job. How much harder? I don't think this study really tells us.

Also, all three possible alternative explanations to racial discrimination in hiring were in fact cases of racial discrimination in hiring.

Then what should be done about it? Do you have any solutions or are you just going to bitch about it?

 

[fskimospy]

The study was never meant to tell you exactly how much harder. You also thought that me telling werepossum why your idea was dumb was me saying the same thing you have been all along.

I have no desire to discuss possible solutions with you after how you've behaved in this thread, sorry.

 

[buckshot24]

So you deprive the forum of your genius because you're upset about my behavior.

I'll say that the problem isn't oppressive therefore we shouldn't do anything about it. There are already laws against discrimination and if a company breaks them then the laws should be enforced.

I think the best way to combat discrimination is for companies to lose money by doing it and not through fines.