I couldn’t open the housing one. In terms of job performance you could say the hirers were using bayesian probability, so applying stricter standards based on the population results. Which I agree is unfair. http://faculty.washington.edu/mdj3/MGMT580/Readings/Week%201/Roth.pdf
What is your explanation for the abysmal performance of black students on MCATs as compared with whites and asians four years on from the same educational opportunity?
What is your explanation for the same gaps persisting on the United States Medical Licensing Exam four years on from that?
It’s all well and good to paint with some sort of broad brush that there is an ongoing institutionalized conspiracy to keep blacks from performing equivalently with equivalent education opportunity. I find no evidence for such a position.
In nearly any secondary school, immigrant asians will outperform native blacks regardless of income. In colleges and universities there is a tremendous amount of concern and effort directed at getting black students up to par on the assumption that they have been matriculated with disadvantaged backgrounds (even where that is not the case). And yet the academic gap in any quantified STEM disciplines never changes.
And I have given you the example of wealthy black children underperforming poor white children on SATs. What, exactly, would it take to persuade you that opportunity has been equalized?
I believe your recalicitrance to admit that this gap cannot be eliminated is based on a good heart, but you are putting what you want to be above what is, and there is not a single nation or culture, regardless of their history, which has been successful in eliminating the gap. I admire your heart but not your objectivity.
I’m not foolish enough to think that there is an obvious or easy explanation for a phenomenon that’s influenced by so many complex factors. I’m just objecting to the widespread practice of jumping to the conclusion that it must reflect some innate inferiority on the part of blacks.
Fortunately, nobody (at least nobody here) is arguing for such a position. I’m not claiming in any way that there’s any kind of “conspiracy” to keep black test scores low. I’m just noting that there are a number of factors other than the hypothetical black intellectual inferiority that could be involved in it.
A society without significant racial prejudice against blacks. It baffles me how you can imagine that just because some black children are from wealthy families, therefore there must not be any social or cultural inequalities that might affect their academic performance, and therefore their underperformance must be the result of their innate inferiority.
It’s not that I’m objecting to the very concept of hypothesizing possible innate race-based inferiority. To say “Persistent race-linked outcome differentials are a significant phenomenon, and we should study them without assuming a priori that there can’t possibly be any innate race-linked ability differentials” is a perfectly reasonable position.
But to say “Persistent race-linked outcome differentials can only be accounted for by innate race-linked ability differentials, and anybody who disagrees is a reality-denying (although good-hearted) PC dogmatist” is pure horseshit.
We simply don’t have the ability to test the phenomenon rigorously enough to infer any such definitive conclusions about it. That you insist on drawing such conclusions anyway does not speak well for either your objectivity or your heart.
That’s probably a difference in cultural values. East Asian immigrants are usually obsessed with academic acheivement and will push their children to study. Also many time their income levels will not correspond with their education levels. Many college educated Asians will come to the US and start from scratch, like running a corner grocery store or laundry. (It’s a stereotype because it’s commonly true.)
You can’t really compare the two minority groups. They have completely different historical backgrounds when it comes to their experiences in the US.
I do not want to speak for everyone else, but institutional to me means the effect of inertia over time of giving 1 group every advantage over another. This institutionalized racism has made its way into almost every facet of society. To think that several hundred years of being told that your intelligence is no different from a common animal disappears when “given the same educational opportunity” puzzles me. You can give every statistic in the world, but realize hiding behind present statistics completely strips psychological and historical factors from “intelligence”. I believe your error is in thinking we have actually leveled the playing field where we can accurately compare the groups. It seems you think two people living in the same community going to the same schools should produce the same results if they are the same genetically, regardless of their race.
From here.
From here.
From here.
How does your theory explain these results? To me, there is an obvious psychological aspect of intelligence that is no doubt influenced by our environment. By no means am I claiming genetics is not the predominant factor in intelligence. I really have no idea. This is not my field. However, I do know your repeated NBA example and cherry picking of test results as some kind of proof that intelligence is genetic is lacking.
I have not seen anyone claim there is a conscious institutional conspiracy to keep blacks from performing in a certain way. What I have seen are claims that certain institutionalized forces (lack of education, being told you are not intelligent, lack of economic opportunity, etc) have combined in ways that keep one group under-preforming compared to another. Your arguments tend to reject these institutions and claim people have “the same educational opportunity”. Wealthy blacks face different psychological drivers than poor whites. You can type until your fingers are tired that these do not matter, but that does not make it true. If the stereotype threat is accurate, that would very much explain why Asians outperform blacks - because they are expected to.
Simply looking at test scores and the present state of society is only picking the fruit of a rotten tree.
It isn’t. Group differences start to show up at age 3. psychology.uwo.ca/faculty/rushtonpdfs/PPPL1.pdf -
In any case, a recent meta-analysis also showed publication bias - that tests showing no stereotype threat didn’t get published (ie. as you saw with Climategate). The overall conclusion is that stereotype threat does not exist. The unpublished and published studies were compared on many indices of quality, including sample size, and the only variable predicting publication was whether a significant effect of stereotype threat was found.
http://www.isironline.org/meeting/pdfs/program2009.pdf#page=68
A statement and a link to an entire paper without a page number is not helpful.
Do you have access to this paper that the general public does not have? Because it appears to still be under review and the only place I found discussion on it was here (which incidentally contains an exact quote of what you typed). I will wait for the paper and the academic response to it, as opposed to
Regardless, to disprove a theory, not by testing that theory itself, but reading other papers, is certainly an ambitious task. If you have any research actually testing the theory, I would be grateful.
My observation is that we will never have the ability to normalize rigorously enough to satisfy those who want to find environmental influences as the principal accounting for differences among human groups. They will simply continue raising the bar for “proof” while the world goes on and every effort to ameliorate innate differences continues to be unsucessful.
For the most part, that reflects a good heart, and perhaps that’s why I find myself uncomfortable arguing for the opposite position. But I am often reminded of the old joke where the donkey-stealing neighbor protests his innocence while the donkey brays in his backyard. Upon being confronted, his defense is that it is unfair to take the donkey’s word against his.
The donkey is braying, Kimstu. Look around your own or any institution of higher education and persuade me that the cards are stacked against black students; that the admission committee is biased against blacks; that the faculty wants to see them underperform; that they are bereft of financial resources unfairly; that they are so obsessed with their history of slavery and institutionalized racism they cannot study properly; that they are undermotivated to become Doctors and Scientists and Engineers; that they have no encouragement…perhaps you would like to look beyond our borders here in the US and point to any other nation and culture with opposite results to ours as proof that the problem is local, and therefore environmental.
What you will find when you look at success is that in non-STEM fields there have been remarkable improvements in degrees awarded and job penetration. Why, other than a diminished ability to perform in quantifiable fields, would there be less success in other areas? In medicine (my field), particularly when Howard, Meharry and Morehouse are removed, the under-representation of blacks is remarkable, and the desperation of medical schools to find qualified black students is equally remarkable.
I don’t mind you questioning my heart. I expect it, and in fact I consider the motivation to do so noble. But the scientist in me says that if it were not for the social overlay, we would not question so vigorously the potential reasons for the outcomes we continue to see. I do not believe for one moment that the average institution of higher education in the US has STEM departments which do anything except actively recruit, and encourage, and diminish barriers for black students. Where is their success?
If we can get to a world where we don’t care about race, that’s great. In the interim, if we just care about people, we need to be careful about how readily we make assumptions. If we decide that equal opportunity creates equal outcome because all groups are equal, we risk shifting away from race-based preference quotas to opportunity-based preference quotas. Under such a shift, we will diminish further the pool of black applicants into fields where they are already desperately under-represented.
We owe it to our society to vigorously stamp out race-based injustice–institutional or otherwise–and a priori assumptions of ability for any given individual. We owe it to ourselves to amend past wrongs. We owe it to ourselves to not only equalize opportunity but to provide an additional helping hand to anyone dealt an unfair card, be it nurture or nature. Such positions do not require that we overlook the cold reality of mother nature’s inequality, or pretend we cannot hear the donkey bray. Recognizing group differences in ability does not require us to become assholes.
Right there is the massive gaping flaw in your argument. I am not, in fact, questioning any POTENTIAL reasons for race-linked outcome differentials at all.
On the contrary, as I have repeatedly stated, I think it’s perfectly justifiable to consider innate race-linked ability differentials as a POTENTIAL reason, among many other plausible potential reasons, for race-linked outcome differentials.
What I am questioning—because it is, in fact, highly questionable—is the eagerness of many proponents of theories like yours to jump to the conclusion that innate race-linked ability differentials are the ACTUAL reason for race-linked outcome differences.
The problem is not that you insist on considering innate racial differences in ability as a possibly real phenomenon. The problem is that you insist on describing that possibility as though it were a demonstrated fact, and that in so doing you are persistently talking out your ass.
Whether I could persuade you of that conclusion, and whether I could persuade a reasonable person who isn’t already immovably biased against it, are two different questions.
A reasonable non-biased person would readily recognize that although there are many ways in which higher education doesn’t stack the deck against black students, and even some ways in which it tries to stack the deck for them, there is still a lot of pervasive racism doing a lot of negative deck-stacking.
A reasonable person would also note that pervasive negative cultural and social stereotypes about individuals’ ability have a significant negative impact on their performance. This is the well-known phenomenon of “stereotype threat”, examined (for instance) in several studies where students performed differently on tests depending on which stereotypes about their identity they were confronted with. For example, when female Asian-American students took a math test preceded by a survey that reminded them that Asians are considered good at math, they performed better than the control group, but when similar female Asian-American students took the same test preceded by a survey that reminded them that women are considered bad at math, they performed worse than the control group.
No reasonable person would simply dismiss such factors as automatically negligible just because we now have other social factors that attempt to compensate for them.
Group differences show up around age three. http://psychology.uwo.ca/faculty/rushtonpdfs/PPPL1.pdf
Also, a recent meta-analysis of stereotype threat found a large number of unpublished studies. The unpublished and published studies were compared on many indices of quality, including sample size, and the only variable predicting publication was whether a significant effect of stereotype threat was found.
[QUOTE]
Numerous laboratory experiments have been conducted to show that African Americans’ cognitive test performance suffers under stereotype threat, i.e., the fear of confirming negative stereotypes concerning one’s group. A meta-analysis of 55 published and unpublished studies of this effect shows clear signs of publication bias.
The effect varies widely across studies, and is generally small. Although elite university undergraduates may underperform on cognitive tests due to stereotype threat, this effect does not generalize to non-adapted **standardized tests, high-stakes settings, and less academically gifted test-takers. Stereotype threat cannot explain the difference in mean cognitive test performance between African Americans and European Americans.[/**QUOTE]
http://www.isironline.org/meeting/pdfs/program2009.pdf#page=68
Don’t forget gender. Campuses want a roughly 50/50 balance, so male students are getting their own affirmative action - there are so many more female applicants.
It’s my personal opinion that everyone should be evaluated for individual ability and achievement. Probably will never happen though.
The cite you provided for this claim was merely an abstract in the conference program of a meeting of the “International Society for Intelligence Research”, and did not actually cite any research publications. Moreover, the claims in that abstract do not in any way invalidate the hypothesis that the “stereotype threat” phenomenon contributes to minority underperformance in test-taking; they only argue that stereotype threat alone is not sufficient to account for the difference in results, which AFAIK nobody’s disputing.
So did I get your stereotype study correct? :
***There is no stereotype effect for groups that excel at something (in your study, math scores over 700).
For mixed-gender groups that are good, but not great (650-700 scores), women are much more suggestible and have a presumption of inferiority at math that consumes their cognitive ability and prevents them from converting a word problem to a math problem (p.68). Therefore if you add a sentence to a word-problem math test saying it’s gender fair, math scores of good-but-not great students will equalize by gender.
Specifically, in the math problems given this test group, if you didn’t add a sentence that the test was gender-fair, then women failed to formulate a strategy 14% of the time, where men failed to formulate a strategy 2% of the time. If you added a sentence to the instructions mentioning that the test was gender-fair, women failed to formulate a strategy 4% of the time and men were now unable to formulate a strategy 9% of the time (more than a four-fold worsening of that measure).
And when faced with new or challenging situations, stereotype threat may undermine women’s performance (bottom, p. 68).***
Well…OK. It’s a bit of sniff-test failure for me to find that adding a comment to test instructions that the test is gender equal quadruples a male’s inability to formulate a successful problem-solving strategy, but perhaps we can agree that your cite does suggest that:
-
High-performers are unaffected by sterotype threat. Perhaps this is a nice way of saying that really smart people aren’t dumb enough to buy into stereotypes.
-
Stereotype threat is notoriously difficult to assess indirectly, and typical results from typical studies such as this are sort of all over the place when you really get under the covers.
-
Men and women are different, perform differently under different circumstances, and are affected by different things. Must be something innate, huh? Else how did these differences arise that give us need to try and create circumstances to equalize performance metrics.
Perhaps a better way to assess stereotype threat is not to infer it, or assume it exists, but administer a quantifiable test getting at how that individual sees themselves and their ability with reference to the rest of society. Another sniff-test failure for me for this particular study is the underlying assumption that there is good evidence that women carry around some sort of cognitive burden that men excel at math and they just can’t compete. As a math ignoramus (yeah; ignoramus there also…) with mostly female math teachers through the years, I beg to differ.
But hey, I am all for the “good job!” back pats and all that, just so you know. And for many years I wore a cute little blue ribbon that said “I can tie my own shoes!” on my lab coat. I just think the whole excuse-finding stereotype thing is a titchy bit overrated, particularly when my ass is getting kicked by someone who is simply too bright to buy into that b.s.
And why, to the point of the main discussion, wouldn’t it be a successful strategy for higher-education (all?) teachers to simply make it clear at the beginning of all classes that girls are just as good as boys, and all ethnic/racial groups are just as good as each other. Why wouldn’t that solve the problem?
No. You are a priori committed to believing that the “stereotype threat” phenomenon it documents isn’t real, so you made up a completely vague and subjective measure that you called a “sniff-test”, and decided that according to your “sniff-test”, the study wasn’t credible.
Not very persuasive reasoning, but not very surprising either.
Do you have specific rebuttals to my summary of the study findings? Which findings did I get incorrect? I only had a limited amount of time to look it over and I don’t want to summarize it incorrectly.
I did use your study as a cite to support the notion that the “stereotype” problem is not an issue for the smartest tier…I did not mean to discount the issue entirely.
I do believe there are large environmental influences, but it’s true that for me many of these sorts of smaller studies never seem to pan out into actually changing the outcomes we see. I leave it to others to decide if the reason for my “sniff test” failing here is “vague and subjective” (i.e. in your study, telling men that a math test was gender-neutral quadrupled their inability to find a solution).
For one thing, your characterization of the statistics was rather inept (which is par for the course for non-mathematician doctors, IME as a math professor who sometimes uses medical studies to help explain statistics concepts in quantitative-literacy courses). In particular, this:
Describing a change of 7 percentage points as a “quadrupling” of the measured effect, just because the 7-point change happens to occur in the interval between 2% and 9%, is silly. If the same 7-point change had happened to occur in the interval from 4% to 11%, it would have had a similar interpretation statistically, but according to your “creative mathematics” it would have represented a vastly different impact: less than tripling the effect instead of more than quadrupling it. This is not a useful way to think about statistical results.
Then I’m not sure I see what we’re arguing about here. As I’ve said many times, I am not objecting to the hypothesis that there might be innate genetic differences in certain kinds of intelligence or ability linked to racial categories. I don’t think it sounds very likely, given that qualities like intelligence are so genetically and physiologically complex while basic racial categories are so genetically vague and broad, but AFAIK there’s nothing that makes it actually impossible. Nor do I feel that it should be illegitimate to study such a hypothesis scientifically, or to make policy decisions based on it IF we find demonstrably decisive evidence in favor of it.
All I’m objecting to is the unjustifiable haste with which many proponents of such hypotheses attempt to claim that the hypotheses are already soundly established on a scientific basis. Face it, they aren’t. And given the complexity of all the factors involved, they may never be.
I know it’s frustrating for you to feel that hypotheses and evidence which seem patently plausible and convincing to you are just constantly being dismissed by others as inconclusive. And your frustration naturally leads you to suspect that your opponents have already made up their minds and will never accept these conclusions, no matter how good the evidence may become in the future.
But frankly, that’s your problem, not mine. I’m not going to be buffaloed into accepting premature (at best) broad-brush conclusions about an extremely complicated issue involving myriad factors in genetics, psychology and sociology, just because your frustration at my resistance inspires you to sneer at me as a PC dogmatist.
Variation in human intelligence is a fricking difficult phenomenon to test because many of its factors are damn near impossible, if not actually impossible, to control for in a meaningful way. Sure, we should keep studying it and hope that continued social changes will eliminate or reduce some of the confounding influences so that we can get conclusive results. In the meantime, though, we have no valid scientific basis for decisively rejecting the perfectly reasonable null hypothesis that there are no significant differences in human intelligence across the socially-constructed categories known as races.
But if genes occur in different frequencies across groups then why would you expect equal outcomes? For instance, the RR variant of ACTN3, a gene that affects fast generation of muscular force and correlates with excellence at speed and power sports. The opposite variant of the gene is called XX. Tests indicate that the ratio of people with RR to people with XX is 1 to 1 among Asians, 2 to 1 among European whites, and more than 4 to 1 among African-Americans.
And it’s been found that genetic change has increased over the past 10-15 thousand or so years. Lahn got pressured into moving onto other projects, but clearly some changes relate to neurological function.
http://query.nytimes.com/gst/fullpage.html?res=9B01E1DE1331F93AA3575AC0A9639C8B63&sec=health
As the cost of sequencing drops more findings will come out. Other examples of recent changes.
http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1371%
http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.0030090
It’s a reasonable null hypotheses, sure. But still the donkey brays when I look around. And the places I look–higher education–are the places which have extended the firmest hand and the greatest effort in normalizing those confounding influences. Since there is enormous pressure to not only equalize opportunity, but to go the extra mile in lending a helping hand to the underperforming, let’s see how far we can get. We can certainly agree on the need to do that.
To date the outcomes of such efforts have not been encouraging, here or anywhere else in the world.
Perhaps you will be able to revisit the topic in a few years and claim victory for your null hypothesis. If not, I am sure there will be calls to extend the period of time needed to equalize opportunity. Perhaps you will still be looking at advanced math studies with a paucity of representation from groups you had hoped would have long since been proven equal in aptitude. But then again, maybe the Inuit will have come to dominate the NBA and you will still get the last laugh on the Pedant and his silly theories that “confounding influences” are not all that difficult to control for.
I personally rather suspect that in (non-human) animal husbandry, where there is no social overlay, genetic science will have moved on and accepted the obvious: we are our genes.