Asymmetrical Information

Excellent post, I guess you've got to read Asymmetrical Information to find good examples of, well, Asymmetrical Information.

I've often wondered about this topic. To use safely non-loaded terms, if early 20th century Irish really were a bunch of drunks (with plenty of exceptions), what would be wrong with putting up a "No Irish Need Apply" sign?

Winterspeak: your link's not quite right.

I took one of the tests. It hasn't changed much in seven years. I noticed two things that, if not controlled for, could be big problems. First, if you have someone categorize things one way 50-100 times, it's no surprise that they're slower and make more mistakes if you immediately ask them to do it the opposite way. If the orders of categories aren't varied, that's a fatal flaw. Second, is this thing controlled for handedness? My right hand's faster on the keyboard than my left.

Winterspeak,

Isn't "statistical differentiation" still something that people should be aware of and concerned about? That is, given the limited information available on a resume and interview, isn't the "rational" choice (at least in close cases) to hire the white guy based purely on the odds of criminality?

And isn't that a fairly serious problem? I have to think that it is.

None of which is to say that psychologists with an axe to grind are able to design a decent experiment, which they aren't.

Job interviews, which should be all about the individual's identification with those of his own race and sex and age, in order to clarify the individual's likely behavior, are forbidden from even addressing these topics. So it's no wonder that hirers are thrown back on their assumptions.
I suggest a new law requiring employers to ask specifically about the implications of an applicant's race, sex and age, so as to bring out any factors that may differentiate him from the collective mass whose outward characteristics he shares.

And when was the last time you hired someone based solely on a resume and interview? Does nobody check references anymore?

Great Post, Jane.
I took the test and noted that it is indeed assuming several things that cannot be simply granted. One major flaw is the assumption that a faster response is more likely representative of someone's "true feelings".

Hogwash. While a kneejerk response might indicate a set of assumptions that are less socially acceptable, it might not. The test does not seem to differentiate among several competing possiblities. Handedness, confusion ("what is the test asking me to do again?", competing impulses (is gay "bad" or is gay "homosexual"?). The test doesn't tell the difference. Moreover, it is unclear whether a quick reaction is "an unfiltered true feeling modified by social acceptability" or if it is instead an overlearned response that the person does not in fact harbor, or *more importantly* ever act on. (unless Harvard wants thought police to keep us from thinking unwanted thoughts).

The test is in my view rather pointless. I learned to cheat on the test right away, and got a "good" result. It couldn't tell my views if it tried.

Many employers have a "speak no evil" rule about references. And if there is something wrong, don't they want to get rid of the bad apple without going through a tortuous and risky firing procedure?

Unless I personally know the references, I just ignore them. If I check a reference who I don't know, then I have no reason to believe that they will tell me anything other than what a wonderful person and hard worker the person is. I don't even know if they actually worked with them, they could just be a childhood friend or relative using a fake name.

If I personally know the reference, then it is a good idea, otherwise it's useless.

The issue of statistical differentiation discrimination aside, I think the tests do a good job of detecting bias/predilection. They just confirmed my own introspective assessments - I was told I have a strong automatic preference for:

beautiful people over rich people
talent over effort
meg ryan over julia roberts

On the other hand, I first did the race one a year or so ago, and was told I (a white male) had a moderate automatic preference for whites. I was just told the converse, that I had a moderate automatic preference for blacks. I don't think I've changed, but who knows?

Where they do fall down is in the accompanying survey questions. Eg:

"Air passengers arriving in the United States must pass through a checkpoint where Customs officers may examine contents of baggage in search of contraband such as illegal drugs. Should Customs officers be more ready to examine contents of baggage for a Black passenger than a White passenger?"

My answer to that would be very different if the second sentence were prefaced by "Conditional on the country of origin...".

Regarding checking references:

Given the bias against saying bad things, it may be worth checking references because the occasional bad one is relatively credible and may protect you from a costly mistake.

I once had someone work for me who had many good qualities but was unbelievably difficult overall (I won't bore you with the details). When she left, I gave her a recommendation letter and didn't want to bar her from all future jobs by saying everything I thought. If I had refused to give her a letter, or had given her a bad one, she would have had to go home to the Philippines, which has limited economic opportunities.

So, I mentioned her many positive qualities but didn't describe how and why she drove me crazy. It wasn't a glowing letter but was one that would allow her to get a new job, if she did well in the interview. Nevertheless, if someone had called and asked, I would have been open and honest. A letter is selective. Ask questions and dig a bit, and sometimes you'll learn more.

The worst error about these fruitless endeavors is that it shows nothing about outcomes. It assumes that reflexive bias will translate into, well, something. But that is a huge assumption, and on that is most certainly wrong.

But is sure sells well. It suggests to me that a "Conservative-Liberal IAT" would be fun ... if the researchers took it themselves.

I think, barring those with truly crazy ideologies, statistical differentiation with respect to race is the leading cause of racism in practice.

If, when searching for an English tutor, I automatically exclude anybody with a last name of Chang, on the grounds that people named Chang are statistically less likely to have good English, is that racism?

What about if, after asking for writing samples, I still decide to exclude people named Chang, regardless of the quality of the writing sample, on the grounds that, all other things being equal, somebody named Chang is still less likely to be good at English than somebody with an equally good writing sample who's named Johnson?

Under your logic, these are all defensible decisions, given that time is limited (and time is always limited in the real world), and all-encompassing metrics are difficult to devise. But they would be recognized as racism nonetheless.

"Project Implicit" may be another reminder of Harvard University’s declining importance. This school is so vastly overrated. It is now something of an intellectual whorehouse. Harvard’s culture seems dominated by political correctness. Only a short time ago, Larry Summers was rebuked for merely suggesting that studies be performed to learn about the differences between men and women. “Project Implicit” is apparently more of the same nonsense.

For damning criticism of the implicit racism study, try Tetlock and Arkes, Attributions of Implicit Prejudice, or Would Jesse Jackson"Fail" the Implicit Association Test? [pdf]

I think there are many forms of predjudice, of which racism is one, and that instances of prejudice occur primarily in the sub-rational areas of the brain - that is, in the subconscious. I trust people who are like me more than I trust people who are different.

But then again, I also trust people I know to be trustworthy more than I trust people I know to be untrustworthy, or than people I don't know at all. So prejudice is also partially conscious.

So prejudice, including racism, is partially subconscious and partially conscious. It seems to me that the real question is, to whom can we assign responsibility for the negative effects of racism? And it seems to me that the answer has to take into account the degree to which the racism is conscious.

Gary Becker's definition of racism, as you describe it, is full of holes. In the old South, an employer who refused to hire blacks, as sales people, for example, (or any job defined by the local mores as "white"), would be both racist and smart. Hiring blacks in a manner that would offend his racist customers would lose him business. The fact that a smart black employee might perform the job "better" than a dumb white one is irrelevant. Racism certainly reduced the economic potential of the South, but within that racist system, businesses that complied with racist values were doing the smart thing, if not the right one.

What's interesting to note about this study was a topic that was covered in Malcolm Gladwell's book, "Blink." Most people fail miserably when subconsciously associating "bad" things with black people, but when they were primed with images of prominent African-Americans (they were shown images of Nelson Mandela, given biographies of Colin Powell, listened to Martin Luther King, Jr), they showed a significant improvement in the test.
In other words, on some level, our "racist" feelings can be eased by exposing ourselves to examples that contradict the stereotype. Think about the ratio of black males on television getting arrested to the black males on television, say, earning a PhD.

It's not rational, it IS racist.

I couldn't the implici link to work. I don't know if it's me or the link.

Best of Luck!

WRT Jesse Jackson, he tried to repair the damage done by his comment (that he'd be more worried about being followed by a black man than a white man) by claiming that what he *really* meant to say was that a predominately white neighborhood would have better police protection.

Employers are less likely to pay attention to references because most companies will only confirm past employment, due to lawsuit issues.

Is it also not unlikely that some of the "anxiety" associated with black people might be ... anxiety over past racism?

Especially in a university setting?

The most useful information to come out of references is what's not said. People will rarely give glowing references ("best worker I've ever had") unless they mean it. A reference full of lukewarm praise tells you a lot.

Remember the qualifier:

If you do not have detailed information on an individual, it is rational to evaluate them on group averages.

Once you've interviewed someone for a job, you (presumably) have detailed information on that person, at least to the degree relevant to the job being selected for. If -- despite that information -- you continue to evaluate the candidate on the basis of (say) his race, then you are being racist.

I am not convinced that refusing to hire blacks as salespeople in a racist culture is necessarily "the smart thing", either; particularly if they were willing to work for less than whites. One would probably lose some sales to people whose racism was truly deep-seated. On the other hand, offering comparable goods for a lower price is a difficult temptation for even hardcore racists to resist (as many lower-class whites in the old South found out when freedmen became more numerous). And there might even be some of those hardcore racists who genuinely prefer being served by blacks.

Regarding references, keep in mind that many employers have policies against providing them at all. I work at a Fortune 100 firm, and we only provide employment history, e.g. date hired, dates of promotion, etc. Apparently, the "no reference" policy is due to liability concerns.

Suppose the recommender is European and actually gives an honest recommendation letter ;>? Whenever I've been involved in hiring, we normally know the recommenders, something of the work of the potential hire, and so it's the interview that helps discriminate between people.

Regarding race, I think it's important to challenge one's "instincts" and look at the person in front of you instead of just seeing the stereotype.

Gary Becker's definition of racism, as you describe it, is full of holes. In the old South, an employer who refused to hire blacks, as sales people, for example, (or any job defined by the local mores as "white"), would be both racist and smart. Hiring blacks in a manner that would offend his racist customers would lose him business. The fact that a smart black employee might perform the job "better" than a dumb white one is irrelevant. Racism certainly reduced the economic potential of the South, but within that racist system, businesses that complied with racist values were doing the smart thing, if not the right one.

Jane Galt mentioned only a part of Becker's theory. Becker also went on to formalize that racial discrimination can only persist if consumers are racist. A racist seller or producer will be eventually be driven out of business if he continually discriminates against productive employees.

I former co-worker and I were looking for work, (he was actually fired, and I was laid off 3 months later), and during the process of looking for work, he wondered if they were giving him a bad reference. So I called, from my new job, inquiring about him. All they would tell me is that he worked there.
I should have asked about myself to compare, but I've heard a lot like this before (reference Biff above, among others).

As far as the implicit test, it's rooted in subconcious notions. They tell you to go as fast as you can, and if you do that, you are likely to repeat patterns that you went through previously (ie, see 'joy' and hit the 'e' key like last time). Hitting the 'e' key, or taking more time to hit the 'i' key doesn't indicate bigotry.

And, as a war-supporting, affirmative action opposing conservative, I had no differentiation between dark and light skin. But I did try to take time to think . . .

Great Post, Jane.

Jane Galt mentioned only a part of Becker's theory.

Winterspeak. Winterspeak! And along those lines, how did ya guys navigate the written test instructions?

This is a subtle question with many interesting folds.

Being racist or not should be decided on how you treat individuals within a group, how you update your beliefs based on that new information, and also how strongly you seek out that individual information. Your starting beliefs about group averages can be utterly legitimately as biased as whatever data you have. Care is required in discussing all these issues to not undermine the very foundations of knowledge itself. The same populational ideas and facts which let us know black folks have historically been treated worse lie behind the starting distributional knowledge. Without some trusted means to come to such broad knowledge, well the anti-racism project is over before it's begun.

Bigotry and biases are about how you apply knowledge, and ultimately quantitative rather than qualitative. Model-dependent attention is even a big problem in establishing the objectivity of science. Sure there will be patterns that suggest various hiring strategies. While I haven't examined any such data personally, my intuition strongly suggests that in most scenarios (yes, I could construct counter examples), race is by far a weaker predictor of performance for some specific tasks than other traits such as experience, education, intelligence, attitudes, socialization, state of birth, etc. Those more significant variables are all a little harder to suss out. except maybe state of birth or some such which is probably on par with race as a predictor. :-) People are motivated to actually lie, exaggerate or otherwise obscure the most predictive variables.

Ah, so there's another issue. More predictively potent variables are often harder to identify or measure due to deliberate misrepresentation toward the advantage of potential selectees, but what we always have in life is the joint measurement-prediction problem. So, part of the selection question comes down to how much we all need to control our inner monkeys. The ape within tries to ascribe more predictive significance to more easily identifiable features. We all feel that tug, don't we? We don't think "Oh, I should have known that variable was more predictive!" Our reaction is "Damn, I should have properly classified that person as one of *those* based on some easily identifiable trait". This is also why earnest data-driven, prediction-motivated folks tend to value honesty more highly than people who make decisions emotionally, as the latter aren't really going to use the information anyway. It's also why people tend to have "zero tolerance" reactions to various forms of dishonesty which is arguably a knee-jerk totalism not so well-justified, rationally, as we're all more or less revealing at various times and in various circumstances. (I'm not advocating or condoning deliberate deception or anything, but only tying together some points about, well, how commonplace asymmetrical information is and how much we all participate in that).

Anyway, the joint problem can be overwhelming or at least impractical. In response, people concoct (and propagate) overly simplistic heuristics to get by. Though it is slowly getting better, modern medicine, for all its grounding and aid from biological sciences, is in practice riddled with such artifacts. This is one route to an argument that though homo sapens sure seems very tribalistic, it may be more related to the structure of knowledge and finite resources than genetics. So, where's the balance and boundary between identification of easily applyable rules and hard-core racism (or whatever you would call a bigotry to replace hte race category that would be easily agreed upon as equally simplistic)?

In many sensible contexts where the question arises, selectors are strongly motivated to discover, by whatever means available, more individual information about potential selectees. The original entry asserted its absence, but the very *desire* for it is what discriminates (no pun intended) the racist from the rational in practice. With an ansatz of no access to more data, I would say there's no way to even answer the question. And so in that sense I agree with the original point completely and am just loquaciously elaborating on the thesis that it's racist to not desire or seek the data because of an early racial revelation, but not racist merely to believe in populational trends.

Another even more subtle issue here is that potential selectors are more focused on the "tail" rather than the meat of given probability distributions. Their risk preferences are just such that they want to manage their downside more than optimize their average. Then one gets into not group averages, but various percentiles of one form or another where racial differences can be more stark, and in the context of decision making one gets into trying to second guess or tell someone what their risk preferences are. That's a slipperly slope, too. Yet, unchecked by vociferous notification or at least discussion, our inner apes generally do seem to have far too much sway.

In any case, the overall gist of my reply here is that racism, prejudice, and all forms of "bias" are actually an unquantified but nonetheless quantitative spectrum, that the general effects are very hard to eliminate in their entirety, and that all this is really tied to foundational issues in the theory of knowledge. I don't know if my examples motivate these thoughts so well. In any event, I actually think it's naive to have a *precise* litmus test for "racist or not", though knowing what is clearly out of bounds is probably necessary and not that hard. The very enterprise of such a precise litmus test falls perilously close to the same trap as what is being castigated or promoted, anyway. "What group is the person in? "Black haters" Can we use this group to predict and identify their detailed ideas and actions? Friend or foe!??!"

Maybe it's better to just say, we all have our prejudices. We all sadly even develop more with time. We can all strive to be better, more rational, more evaluative. Those are just good ways to fullfilling whatever other wants we have. Race is probably not a good predictor as other categories most of the time. So, sure paying undo attention is both foolish and mean. But, also, as Jane Galt says, take care not to undermine the basis of knowledge by way of your attack of misjudgement. It likely only requires slight care in how you state things.

As to the broader problem of all bias,training or re-training any detailed thought certainly requires effort. What's a "fair" equilibrium? Who knows? Certainly not I. I do think it's a hard and interesting question, though, at the root of answering how to allocate one's time with almost any project uncovering the truth of anything. And now, this little box annoys me! Adieu!

prominent African-Americans ...images of Nelson Mandela. Ahem!

Irrational racism is simply not an issue in preventing qualified black people from getting jobs in the United States, and the proof of that is the world of professional sports.

There is no way that blacks would comprise the majority of NFL and NBA players if racism was any sort of actual issue of importance. The owners of sports franchises in the 1950s were not, I'm sure, among the most enlightened and liberal types around. But they wanted to make money and to make money they needed to hire the best players, and the best players tended to be black. And if whites were as racist as so many people want to make them out to be, pro sports would have collapsed by now because whites wouldn't watch black players outperform white players.

The fact is that there are real differences between the races, and not just trivial skin tone differences. Blacks excel in football, basketball, and boxing because of their genetic advantages in certain physical attributes. The brain is also a physical organ and there is no reason to think that traits like intelligence, time preference, and other aspects of personality are somehow divorced from genetics. This, in my opinion, goes a long way to explaining differences in performance in different fields by different races. As more and more is learned about our genes, undoubtedly these personality/intelligence/genetic links will become clearer, as already the medical differences between the races are becoming clear. (Certain drugs are less effective for blacks than for whites, etc.)

But whatever differences are found, the fact that blacks dominate professional sports, and to a large extent, the music industry, demonstrates that racism is no obstacle to people with an ability to prove their actual aptitude in a field.

I wish people would stop falling over themselves to make arguments showing how un-racist they are. It's such baloney and there is no reason to be ashamed of thinking there are real, significant differences between the races. THERE ARE. It is irrational to think that there aren't. Just because there are differences doesn't justify unfair treatment though, any more than the fact that men and women are different justifies unfair treatment. But please, let's stop pretending that there aren't any differences. That is such a lie.

Only a short time ago, Larry Summers was rebuked for merely suggesting that studies be performed to learn about the differences between men and women.

This is nonsense. Some of Summers's biggest critics at Harvard actually perform studies to learn about the differences between men and women (see, for example, Elizabeth Spelke). Without getting into the question of whether the criticism of Summers was fair or not, it is simply wrong to think that this was a debate between people who just want to do some studies in the name of truth and others who don't want any studies to be done.

Regarding Project Implicit, I just think it is good for people to be consciously aware of cultural stereotypes. Some of the complaints above (asymmetry of handedness, order of presentation) are obviously considered in the analysis of the results. There's definitely a gap between stereotypes affecting a split-second reaction task versus affecting job hiring decisions. And while it's not the case that eveyone is the same, and it's possible to argue that using group characteristics in the absence of more personal information is actually quite rational, let's not forget phenomena like "mascotting" in which very different standards are applied to members of different groups, for no real rational reason. There's a dark side to stereotyping and the basic way around it is simple awareness.

Blacks excel in football, basketball, and boxing because of their genetic advantages in certain physical attributes.

Why not hockey? Do they get cold?

What specific attributes of black vs. white vs. Latino "physical attributes" explain why blacks are disproportionately successful in the sports you mention -- now -- whereas baseball is increasingly Spanish-speaking, to the extent that the current GM of the Mets has been accused of trying to build an all-Latino team?

How is that 50 years ago, basketball was considered obviously a game suited to Jewish attributes -- because otherwise, why were there so many of them playing it? -- and now it's not?

Surely the question of which people devote their youth to trying to become a basketball/baseball/hockey player greatly affects the relative proportions of groups in those sports. And surely the comparative opportunities in other, non-sports fields must affect that as well.

In other words, no, just because there are plently of blacks in certain sports doesn't prove that there is no significant racism in the hiring that the other 99.5% of the population is involved in.

Winterspeak: two comments.

First, the fact that it may be justifiable doesn't mean it isn't, technically, racist. If you discriminate based on racial characteristics, I think there's a good case to be made that there's racism there, even if the racial characteristics are valid. I suspect, in fact, that this "valid discrimination" combined with subconscious bias is the cause of most lingering racism that activists target.

Second, since this bias is unconscious, it seems like awareness campaigns might be less effective than one would suppose. Convincing people that men and women are equal won't help, since they already believe that consciously. We would need to target unconsious biases; this might justify a (private and voluntary) affirmative action system (see the benefits orchestras got when they began blind auditions).

More details at my blog

MarkJ,

How do you separate culture from genetics? For example, imagine that your parents and compeers stress excelling in sports over academics. Let's take your example. Perhaps, they see many images of blacks doing well in sports, but not so many black engineers. Given these inputs, many more would attempt to excel in athletics over academics. If there was another group which stressed academics over athletics, then more of that group would excel in academics. Now, examine the two populations. There will be a difference, but it would not be genetic, but rather cultural.

In terms of hiring, there have been academics who have tried submitting resumes to companies (to be fair, I have only seen the newspaper reports, not the actual studies, so cannot comment on their methodology) with "black sounding" and "white" sounding names, with the ones with the "black sounding" name more qualified, and those with the "white sounding" names with a criminal record and found that those with the "white sounding" names were more likely to be given interviews. This would suggest that there is an implicit bias in the initial screening processes. There are also issues of networks--but given that there are also studies which seem to show that when normalized for experience and educational levels, the earning power of blacks and whites is much closer, then I would suspect that the aforementioned biases can be overcome and the greater difficulty may lie in educational attainments. There, a component is likely due to cultural differences in value assigned to education, as well as institutional differences (for example, greater assignments of blacks to lower tracked courses, special ed., etc. based one expectations).

One problem that I've noticed is that a lot of people in the US these days seem to put a lot more stock in "ability" over hard work. For example, when I tried to work with school teachers to improve math/science education, they seemed very much possessed of the belief that some kids would do better than others due to ability. While of course there are outliers, hard work matters. This idea that "he's just good at sports" doesn't respect the time and training that go into making a professional athlete. The idea that "she's just good" at math, doesn't respect the time that she may be putting in to do well. (Note that I am not claiming that working hard is going to make a successful scientist out of a mentally impaired kid--but an average kid who applies herself can do very well).

Regarding Larry Summers,

As I understood it, his argument was that the tails for intelligence for men were wider than for women. That is, there are more very dumb men and more very smart men than there are very dumb women and very smart women. Since scientists granted tenure are at the tails of the distribution, this would explain the discrepancy. It is a "appealing" argument in that it doesn't require any action. However, another argument may be related to differences in priorities. If one adjusts for single women without children, compared to single men without children then what are the discrepancies? A possibility might be that in the particular case of science, if a woman marries (especially if she marries another scientist), then she will often face the challenge of solving the two body problem. As the number of faculty positions is rather small compared to the number of applicants (say 200:1 at a good institution), then geographic flexibility is a must for all but the very best (and very best is VERY few). For the case of married women if her husband is more established, then this may not be a viable option. If she has children, then this could be another factor. These issues would not be related to bias in the field, but rather to extrinsic factors. The geographic factor may be difficult to overcome, but issues related to children might be reduced by affordable daycare. Other things to consider may be pipeline issues. More women are now attending college than men, but which fields do they choose to study? If there are less going into mathematically intense fields, then this might be something that can be addressed in schools. I could go on, but my point is that if one simply observes, "there are more women in education than men, so obviously there can be no bias against them in any field. Thus obviously, the reason for the difference must just be that women aren't as good as science as men" then one fails to consider what the real problems may be and whether those may be addressed.

"How is that 50 years ago, basketball was considered obviously a game suited to Jewish attributes -- because otherwise, why were there so many of them playing it? -- and now it's not?"

But the modern game of basketball is very different now than it was 50 years ago - so it could well be reasonable to say that once it was "suited to Jewish attributes" (whatever those might be in this case) and no longer is.

What does crossing the street have to do with more easily associating white with "peace" and black with "sadness" than the other way around? By using a wide range of positive and negative terms, the test reduces the likelihood of very specific associations throwing off the results. Even if you are very quick to connect "white" and "right" because they rhyme, or "black" and "criminal" because you watch Cops, your test results will only reflect implicit bias if there is a general trend that extends to the other words as well.

Kevin F,

Moreover, it is unclear whether a quick reaction is "an unfiltered true feeling [un-?]modified by social acceptability" or if it is instead an overlearned response that the person does not in fact harbor, or *more importantly* ever act on.

...

The worst error about these fruitless endeavors is that it shows nothing about outcomes. It assumes that reflexive bias will translate into, well, something.

I think the issues of where these biases originate and whether they are "acted on" deliberately fall outside the scope of the experiment, as do questions about their consequences. These are highly subjective: some people might take the knowledge that they harbor implicit biases against old people and try to consciously account for that when interacting with others; other people will see their implicit bias in favor of men and assure themselves that this is a valid reflection of the state of the world and so nothing to worry about.

The survey questions try to pin down how implicit biases may or may not be reflected in people's actual behaviors and beliefs, but the bias measuring tests themselves don't even attempt to.

biff wrote --
the "no [letters of] reference" policy is due to liability concerns

Isn't it also 'liability concerns' that fuels today's corporate Affirmative Discrimination?

"Why not hockey? Do they get cold?"

Thin ankles.

Judging people based on the group average with respect to race is, by any reasonable definition, racist, just without the pejorative conotation.

Judging people based on the group average with respect to race is, by any reasonable definition, racist

Well, the two definitions of "racist" at dictionary.com are:

(1): The belief that race accounts for differences in human character or ability and that a particular race is superior to others.
(2): Discrimination or prejudice based on race.

There are many circumstances in which judging a person based on the group average for their race meets neither definition.

I just comleted a few of these tests. They're kinda fun. But I have some serious reservations about their accuracy.

I took the native vs. "white" American test. Being both, I was curious. But it seemed that the options were very skewed. I associate "native american" with high cheekbones, a certain level of melancholy, and an inability to grow thick facial hair. In the pictures I was given of Native Americans (all of which had to be over 100 years old), native americans were poorly dressed people, with leathery skin and feathers. "White" americans had nice dress clothes and thinner skin.

It turns out I have a slight preference for fine stitching and working indoors. I couldn't tell you what my racial preferences are.

barbar-
"Without getting into the question of whether the criticism of Summers was fair or not,"

Seems to me that that is the point: was the criticism of Summer's remarks fair? Did they address his point and argue against it rationally, or did they demonize him personally and never approach his query? From everything I've read about the affair, it's proetty obviously the latter.

resigned-

Nice hypothesis; too bad that Haaavaaad isn't interested in testing it. It might come out negative.

email is human readable - aloud

"The Implicit Project implicitly assumes that any differentiation between blacks and whites is racist, and does not consider that case of statistical differentiation. This means they do not consider that an individual may make decisions based on factors other than race, of which race is merely a statistical marker."

I wasn't able to make it to the page for the test, but in the article at USNews.com, its pretty clear they are talking about stereotypes and not racism.

Thats not to say that this wasn't implied by the article, but theres lots of racism too. I've heard this at least three times in the last year "I'm not racist, but..." and two explicitly racist jokes.

Its a pretty important point that stereotypes and racism are not the same thing. Not at all. I think thats the point Winterspeak made. However, I bet lots of blacks find this verbal distinction to be one without a difference in results to them for the following reason:

From the article:
"I don't think that I investigated a hate crime at a school that did not begin with the lower level of slurs and stereotypes and then escalated"

Use this same logic as Winterspeak, but from the perspective of a black person facing some minor discrimination, one that may even be completely based on statistical differentiation without any racism. Anyone can use statistical discrimination, ya know - and it would serve to make a person hyperaware of situations that could turn dangerous to them. Like, for example, having some very minor discrimination thrown in their faces.

Barber et all,

WRT to Blacks in sports. The idea that one "race" has physical attributes that make them "excel" in sports complete tripe. Sports have always been the avenue to success for the lower classes. Look at any professional team sport over the last 150 years and what you will find is that the ranks of professional athletes has historically been drawn from the lower classes, particularly in the era before the advent of multi-million dollar athletes. The reason for this is clear; it is one of the few venues where �talent� is the only metric. Success in business has a huge social component, sure ability and ambition can compensate and often does, but as you move up the ranks in business from grass roots entrepreneurs to the boardrooms of the largest companies the demographics become more and more white, male and Protestant as you move up the stack.

In things like sports and (and to a lesser extent entertainment), hard work and ambition will only get you so far. You can squander your talent if you�re a complete jerk, but an exceptionally talented athlete, has to be close to a psychopath before a team will reject them out of hand. Jews were a force in Basketball because they were among the lower classes and the early NBA was a struggling league. Baseball rosters were filled with poor uneducated whites because no one with an education would want to take the pay cut to play in the big leagues. (this also explains why there were, at one time, are so many floor specialists on the NYSE, they were poor kids hired to write quotes on chalk boards before the advent of technology, they figured out how the system worked and took over).

Of course overt racism infected pre WWII sports as blacks were barred from Baseball since the late 19th century on the theory that white fans would not pay to see blacks on the same field (Football and Basketball didn�t have the prestige that Baseball had and tended to get talent where they could find it). Even the NBA in the 60�s had an unspoken quota system embodied in the old joke that you could have no more then two blacks on the floor at one time when playing at home, three on the road and five when you�re behind.

Present day sports have a cultural aspect; in the Dominican Republic and South American cultures, Baseball is to kids in those countries what it used to be to white kids in the 50�s. Blacks embrace basketball (understandable as it is a game uniquely suited to urban living), and Canadians play hockey. If any kind of material ability were involved then you would see blacks who missed the cut in the NBA dominating things like beach volleyball, which is a game that requires same kind of skills. Golf is dominated by WASPs (Tiger Woods not withstanding but culturally he is a WASPish as Phil Mickelson). I could go on, but you get the point.

Rick, it's quite true that cultural differences make up a lot of the racial disparities in sports. You didn't get close to exhausting the examples — can you think of any champion swimmers that weren't white? Skiers?

But there are physical differences that do matter, too. Not to be overly hackneyed here, though it's frankly unavoidable . . . Kenyan distance runners, West African (or African-American) sprinters, height being (um) something of an advantage in basketball, and some races being on average much taller than others, and likely having different degrees of variation to boot.

Something I'm curious about -- given that American blacks are, with few exceptions, descended entirely or in part from slaves, why *don't* they have above-average strength and endurance? Were slaves not bred for those traits, or did the breeding just not work?

Also, to Michelle's list of racial advantages, I'd like to add Samoan sumo wrestlers -- Samoans definitely have an inherent genetic advantage over the Japanese when it comes to height and weight.

(2): Discrimination or prejudice based on race.

There are many circumstances in which judging a person based on the group average for their race meets neither definition.

Sorry, Dan, "based on the group average for their race" does not change that you are basing it on race. So it always meets definition no. 2.

Sorry, Dan, "based on the group average for their race" does not change that you are basing it on race. So it always meets definition no. 2

It only meets definition #2 if you discriminate. "Discriminate" and "judge" are not synonyms.

Michelle,

I would disagree with the example of Kenyan distance runners and say that the culture here is most likely the reason that east Africans, Kenyans in particular, tend to be distance runners. In that, culture distance running is to them what Hockey is to Canadians, it is ingrained in the fabric of the culture. Kip Kano was their �Babe Ruth� a national hero that every kid wanted to be like.

And distance running is unique among sports in that there are hard coded genetic dispositions that can determine whether or not you can be trained to be an elite runner. There are weight to lung capacity ratios that are measurable and quantifiable that can determine if you have the genetic ability to be an elite long distance runner and I know of no study that has demonstrated that these ratios show up more in East Africans the West Africans.

While height is important in basketball it�s not a crucial metric, contrast the play of 7�7� Manute Bol verses guys like Spud Webb, Mugsy Bowles, Mark Price and Vinnie Johnson. And while the league average of 6�7�, is a full 9� taller then the national average, the fact is that shorter players experience a ton of pressure to �give it up� very early on in their sports career so there is a �cultural� bias (basketball culture) that coaches have that leads them to dismiss the talent of those who don�t �fit the mold�.

It used to be thought that lifting weights would actually diminish the ability of players to hit a baseball which has been pretty well disproved in this era of steroid enhanced power hitters.

"Discriminate" and "judge" are not synonyms.

A meaningless distinction. Just assuming you know what the group average for a race is, is racism.

I gave it an admittedly quick try and didn't get it at all. I was supposed to associate names with either 'Asians' or 'Whites'. So far so good.

But as soon as I picked 'Lee' as white, a red X appeared on the screen and blocked the test until I switched my pick to the - correct, I presume ? - asian category.

What's implicit about forcing people to fit to a pre-programmed stereotype ? How does that even constitute a 'test' ?

It looks like this may have been a warm-up or demo test to get acquainted with the overall interface. But how could a practice run which insists on prejudiced answers not have an impact on the actual test that follows ?

I'm mystified as to the applicability or relevance of the results thus gathered for this specific test.

A meaningless distinction. Just assuming you know what the group average for a race is, is racism.

So if you're walking down a street at night, and two large men approach you from opposite directions -- irrespective of ALL other factors (and assuming you're not fleeing from the FBI for some reason) do you prepare to make a wide arc around the guy in a full business suit, or the one wearing the costume of a gangsta rapper?

Keep in mind that to make that decision -- you have about five seconds before the three of you converge -- you might be assuming you know the group average for a particular dress type.

Anony-mouse, that's dressism, not racism. If a large man of any race is approaching me late at night, I have no reason to suspect he is out to do me harm, unless of course I assume the average member of his race is a criminal; and that would be racism.

How many people even know what the statistical differences are, even to having a decent guess? The absence of statistical knowledge makes that theory incredibly implausible. Much more likely that it's based on racist stereotypes.

Comments are Closed.