Racism in America: Black woman pretends to be white, job offers come rolling in...

[Ryan]

Agreed, but do think there may be a simple explanation since this wasn't a true double-blind experiment. Since the website likely has a tool where recruiters return only the "new" profiles, it may be completely understandable that she got calls on her "new" white profile and few on her "old" black profile. She should have created two new profiles (one black, one white) and observed any differences in callback rates. I'd also want to confirm there weren't any obvious non-racial factors which may have made a difference as well; for example if the new profile had a professional email address like "bianca.white@email.com" and her old profile was listed under "black_power@kill-whitey.com"

She also created a new account without race specified...and got nothing. That kind of throws a wrench into your "new profiles" theory.

 

[xBiffx]

Are you kidding? The evidence is pretty compelling - there were no differences between all three profiles she used EXCEPT for race.

Also, she didn't apply for jobs - employers were seeking her out after finding the online profile she posted. Basically, any employer would have found all three of her profiles when they did an applicant search - and the only difference between all three was race.

So you are saying that all resumes posted on job search sites get equal attention? An employer can't just pick the one at the top of a search?

Everything you posted is assumption. Everything.

As glenn1 pointed out, timing is key here.

 

[woolfe9998]

She went looking for different treatment based on here answers on race. Her and others seem to think that is exactly what happened here. How is that not confirmation bias?

Are you trying to say that she wasn't trying to show racism with this stunt?

She may have been trying to show just that, but when you submit identical CV's except for ethnicity and get different results that isn't a confirmation bias. Confirmation bias would operate to make her methodology flawed so as to confirm her viewpoint. For example, if she consciously or unconsciously created slight differences in the CV's to make the "white" candidate seem more qualified. She may have done that, but it isn't in evidence here.

If you think her bias is causing her to misinterpret the result, then let's approach it this way: if she really did submit identical CV's but for ethnicity and got different results, how would you explain the different results? What perfectly valid explanation is she overlooking because of her bias?

 

[glenn1]

She also created a new account without race specified...and got nothing. That kind of throws a wrench into your "new profiles" theory.

She didn't create a third "no race specified" profile, she just alternated between selecting black and "decline to identify" on her old profile.

You are correct, however, that an even better blind test would be to compare response rates between 3 new profiles: white, black, race not specfied.

Two years ago, I noticed that Monster.com had added a “diversity questionnaire” to the site. This gives an applicant the opportunity to identify their sex and race to potential employers. Monster.com guarantees that this “option” will not jeopardize your chances of gaining employment. You must answer this questionnaire in order to apply to a posted position—it cannot be skipped. At times, I would mark off that I was a Black female, but then I thought, this might be hurting my chances of getting employed, so I started selecting the “decline to identify” option instead. That still had no effect on my getting a job. So I decided to try an experiment: I created a fake job applicant and called her Bianca White.

 

[nehalem256]

They sound 'ghetto' to you because you equate ethnically black names with being ghetto, which is the whole point. If someone is discriminating against class and people assume ethnically black names equate to low class...well...that's racism. Furthermore, the study controlled for this effect.

This isn't rocket science.

From the medieval French name Yolande, which was possibly a form of the name Violante, which was itself a derivative of Latin viola "violet". Alternatively it could be of Germanic origin.
This name was borne by a 12th-century empress of the Latin Empire in Constantinople, who was originally from Flanders. It was also used by her descendents in the royal families of Hungary (spelled Jolánta) and Spain (sometimes spelled Violante). Another famous bearer was a 13th-century countess of Vianden in Luxembourg who joined a convent against her parents' wishes, later becoming the subject of medieval legend.

http://www.behindthename.com/name/yolanda

Or maybe because it is a name associated with low-class blacks (commonly referred to as ghetto).

Furthermore, including other groups that you think might also be discriminated against would in fact be meaningless. The presence of other discrimination does not affect whether or not blacks are discriminated against.

We have been over this before, you didn't learn anything last time and you won't this time either.

It works to establish whether there is racism. If the discrimination is because of class or just the name being "strange" to those hiring then the solution to the problem would be different don't you agree?

 

[nehalem256]

She didn't create a third "no race specified" profile, she just alternated between selecting black and "decline to identify" on her old profile.

You are correct, however, that an even better blind test would be to compare response rates between 3 new profiles: white, black, race not specfied.

It would also have been a better blind test if she used the same name (at least first name) for all applicants.

You cannot change multiple variable and then decide which changed variable you think was the reason between the different results.

 

[xBiffx]

What perfectly valid explanation is she overlooking because of her bias?

Right when I was about to answer, someone did it better.

You cannot change multiple variable and then decide which changed variable you think was the reason between the different results.

Thanks nehalem256 for that comment.

 

[glenn1]

It would also have been a better blind test if she used the same name (at least first name) for all applicants.

You cannot change multiple variable and then decide which changed variable you think was the reason between the different results.

That doesn't change my main point - this isn't about multivariate testing. My hypothesis was and remains that the key variable is the profile age instead of race; newer profiles get flagged in the system and are thus more visible to recruiters. Perhaps there's also a recruiter bias towards newer profiles being considered more valuable prospects because their skill sets are fresher.

Regardless, the author of the article didn't see fit to perform an experiment to test my theory, so it's somewhat moot.

 

[fskimospy]

It works to establish whether there is racism. If the discrimination is because of class or just the name being "strange" to those hiring then the solution to the problem would be different don't you agree?

The solution could be different but it wouldn't change the fact that it is racism.

Black name -> no hire
and
Black name -> poor -> no hire

are no different insofar as racism is concerned as you are attributing a negative attribute to someone based solely on having an ethnically black name. Thinking people with black sounding names are all poor is racist too.

This particular lady's experience doesn't really matter as it is anecdotal evidence at best. Luckily we don't need to rely on anecdotal evidence to show that racial bias exists in hiring.

 

[sunzt]

she should change Bianca White's race to black and see what happens.

 

[glenn1]

The solution could be different but it wouldn't change the fact that it is racism.

Black name -> no hire
and
Black name -> poor -> no hire

are no different insofar as racism is concerned as you are attributing a negative attribute to someone based solely on having an ethnically black name. Thinking people with black sounding names are all poor is racist too.

This particular lady's experience doesn't really matter as it is anecdotal evidence at best. Luckily we don't need to rely on anecdotal evidence to show that racial bias exists in hiring.

I have read about repeated studies demonstrating this effect; I linked one below for example. Still doesn't change that the "experiment" conducted by the article author lacks any scientific rigor and is thus is useful only for political debate fodder.

http://www.nber.org/papers/w9873

 

[nehalem256]

The solution could be different but it wouldn't change the fact that it is racism.

Black name -> no hire
and
Black name -> poor -> no hire

are no different insofar as racism is concerned as you are attributing a negative attribute to someone based solely on having an ethnically black name. Thinking people with black sounding names are all poor is racist too.

This particular lady's experience doesn't really matter as it is anecdotal evidence at best. Luckily we don't need to rely on anecdotal evidence to show that racial bias exists in hiring.

But it would be different if the following is also true

White name -> poor -> no hire
Black name -> poor -> no hire

Then no racism would exist.

Sorry that certain names, both white and black ones, imply a person comes from a certain background that might be undesirable.

But for some reason no one ever does a study using the low-class white names. Maybe they are afraid of what they might find?

Also note that her anecdotal evidence is about receiving phone calls not being hired.

 

[fskimospy]

But it would be different if the following is also true

White name -> poor -> no hire
Black name -> poor -> no hire

Then no racism would exist.

Sorry that certain names, both white and black ones, imply a person comes from a certain background that might be undesirable.

But for some reason no one ever does a study using the low-class white names. Maybe they are afraid of what they might find?

Also note that her anecdotal evidence is about receiving phone calls not being hired.

This is not actually true of course.

Even granting your point (which is highly suspect) the issue is that mainstream black names used by huge percentages of the black community are subject to that while a few outlying white names might also be subject to the same discrimination. If all black names are subject to this discrimination and 5% of white names are... that's racism. Period.

This should have been obvious to you, but you're trying desperately to find reasons why racism isn't there as opposed to just objectively evaluating the evidence. Speaking of confirmation bias...

 

[MagickMan]

"Caucasian-sounding woman"

Its funny cuz if a white person says the opposite then thats fucking racist.

Yep. Racism isn't just one direction, not by a long shot.

 

[Pray To Jesus]

Confirmation Bias possible.
Methodology was flawed.
Low power.

Results derived by study are not trustworthy.

 

[nehalem256]

This is not actually true of course.

Even granting your point (which is highly suspect) the issue is that mainstream black names used by huge percentages of the black community are subject to that while a few outlying white names might also be subject to the same discrimination. If all black names are subject to this discrimination and 5% of white names are... that's racism. Period.

This should have been obvious to you, but you're trying desperately to find reasons why racism isn't there as opposed to just objectively evaluating the evidence. Speaking of confirmation bias...

Its not racism to discriminate on names by class. That should be obvious to you. And apparently it is obvious to the people that do these studies which is why the leave low-class white names off the list, because it would break their attempts to show racism.

 

[fskimospy]

Its not racism to discriminate on names by class. That should be obvious to you. And apparently it is obvious to the people that do these studies which is why the leave low-class white names off the list, because it would break their attempts to show racism.

When people decide that a huge majority of names used by one race are low class and a small minority of names in another race are low class, that's racism.

And with that, I'm done discussing this with you. If you wish to educate yourself further on this issue you can go look at the last time we talked about it. My guess is that you're too frightened of losing your sense of persecution to do that though.

 

[nehalem256]

When people decide that a huge majority of names used by one race are low class and a small minority of names in another race are low class, that's racism.

Unless of course a huge majority of names used by one race are used by low class members of that race.

In fact I recall the study you used before showing that. Every black name chosen was provably from a lower class background than every white name chosen.

 

[SheHateMe]

I never know the name "Yolanda" was considered a "low class black name".

 

[xBiffx]

I never know the name "Yolanda" was considered a "low class black name".

Me neither. This is the only Yolanda I can remember.

http://www.youtube.com/watch?v=hXxIX3ZVvFs

 

[fskimospy]

Unless of course a huge majority of names used by one race are used by low class members of that race.

In fact I recall the study you used before showing that. Every black name chosen was provably from a lower class background than every white name chosen.

Lol. Did you not read the study? By the very measure you are citing there was not only no statistically significant positive correlation between the class measure they used, there was actually a negative one within all four groups. ie: the higher the class measure that they used the LESS LIKELY one was to be called back. This held true for more contemporary measures of class such as areas of town a person lived in as well.

I love it when you try to cite evidence that explicitly shows you're wrong.

 

[nehalem256]

Lol. Did you not read the study? By the very measure you are citing there was not only no statistically significant positive correlation between the class measure they used, there was actually a negative one within all four groups. ie: the higher the class measure that they used the LESS LIKELY one was to be called back. This held true for more contemporary measures of class such as areas of town a person lived in as well.

I love it when you try to cite evidence that explicitly shows you're wrong.

I seem to recall you saying I wasn't allowed to compare within a group because such comparison were not statistically relevant

It is a fact that people with the black names were from a lower class background than those with white names. And more important. This is known by essentially everyone.

If you ask who comes from a better off background Yolanda or Madison. I bet 99 out of 100 will pick Madison. And I bet they would be right.

If you ask who comes from a better off background Billy Bob or Madison. I bet 99 out of 100 will pick Madison. And I bet they would be right.

If you ask who comes from a better off background Abby or Madison. I have no clue who will come out on top. Because no one knows.

 

[fskimospy]

I seem to recall you saying I wasn't allowed to compare within a group because such comparison were not statistically relevant

This is why I keep telling you that you don't understand statistics. First, nowhere have I ever told you that comparison within groups is not statistically relevant, at least not for that reason alone.

Your argument is that it was class factors rather than racism. You then cited as support for this the evidence from the study. In order for the study to show support for this you should find that there is a statistically significant positive correlation between names of differing class background. There was not. Therefore, the evidence you cited did not show what you claimed it did. In short, you lied. (or just didn't understand what you were looking at). It is just the icing on the cake that the results actually showed a negative correlation in all cases.

It is a fact that people with the black names were from a lower class background than those with white names. And more important. This is known by essentially everyone.

If you ask who comes from a better off background Yolanda or Madison. I bet 99 out of 100 will pick Madison. And I bet they would be right.

If you ask who comes from a better off background Billy Bob or Madison. I bet 99 out of 100 will pick Madison. And I bet they would be right.

If you ask who comes from a better off background Abby or Madison. I have no clue who will come out on top. Because no one knows.

So your argument is that everyone knows that black people are poorer than white people, therefore when people don't hire blacks instead of whites it's because of classism, not racism.

Thinking all black people are poorer than whites IS racism, you nitwit.

Also, I'll take either of those bets with you if you're interested. How about $100?

 

[nehalem256]

So your argument is that everyone knows that black people are poorer than white people, therefore when people don't hire blacks instead of whites it's because of classism, not racism.

Everyone knows that people with certain names are more likely to be from a low class background. Are you saying this isn't true?

Thinking all black people are poorer than whites IS racism, you nitwit.

All black people are not poorer than whites.

In essence the study changes 3 things with the name:

(1) The perceived race
(2) The perceived class
(3) How "mainstream" the name is

It would be fairly easy to design a test to determine which was the cause. Add name cohorts that consit of

(1) "white" names that are perceived to be low-class (ask white-trash names)
(2) Use foreign sounding names (Pierre, Vladimir, etc)
(3) Completely screwy names like Tigget that no one has any association with at all.

Now if you added these cohorts and "black" names were still the obvious losers I would say you have a strong case that the perceived race of the candidate was the reason behind them not getting call backs.

But for some reason although it was no trouble at all for me to come up with this expanded study it was not done.

 

[fskimospy]

Everyone knows that people with certain names are more likely to be from a low class background. Are you saying this isn't true?

I'm saying that there is no statistical evidence in this case that names with lower class backgrounds were associated with lower callback rates.

You tried to claim this, it was false. It was either due to your ignorance or duplicity.

All black people are not poorer than whites.

In essence the study changes 3 things with the name:

(1) The perceived race
(2) The perceived class
(3) How "mainstream" the name is

It would be fairly easy to design a test to determine which was the cause. Add name cohorts that consit of

(1) "white" names that are perceived to be low-class (ask white-trash names)
(2) Use foreign sounding names (Pierre, Vladimir, etc)
(3) Completely screwy names like Tigget that no one has any association with at all.

Now if you added these cohorts and "black" names were still the obvious losers I would say you have a strong case that the perceived race of the candidate was the reason behind them not getting call backs.

But for some reason although it was no trouble at all for me to come up with this expanded study it was not done.

Haven't we already amply covered that your knowledge about how to conduct studies is incredibly poor? This study already controlled for class background but in your opinion in these cases 'everyone didn't know' that certain names were lower class than others. (LOL) Please connect me to the database of what 'everyone knows', btw.

You just don't like what the results tell you so you're desperately searching for a way to convince yourself that it's invalid. This is pretty standard procedure for you. I really did enjoy the part where you cited part of the study that explicitly disproved your argument though. I mean that's really stupid, even for someone as dimwitted as you.