Sully says this blew his mind. It doesn’t surprise me at all:
In the study made public on Thursday, Dr. Friedman and his colleagues compiled a brief test, drawing 20 questions from the verbal sections of the Graduate Record Exam, and administering it four times to about 120 white and black test-takers during last year’s presidential campaign.
In total, 472 Americans — 84 blacks and 388 whites — took the exam. Both white and black test-takers ranged in age from 18 to 63, and their educational attainment ranged from high school dropout to Ph.D.
On the initial test last summer, whites on average correctly answered about 12 of 20 questions, compared with about 8.5 correct answers for blacks, Dr. Friedman said. But on the tests administered immediately after Mr. Obama’s nomination acceptance speech, and just after his election victory, black performance improved, rendering the white-black gap “statistically nonsignificant,” he said.
The reason this doesn’t surprise me is that it has been well-documented that African-Americans do worse on tests if they are asked to identify themselves by race before their test, so clearly people’s preconceived ideas “race” and “intelligence” influence their performance on these things . And, unlike some, I’ve never been one for half-baked white supremacist ideas about IQ.
To some extent all of this stuff indicates that the notion of using standardized tests to measure “intelligence” (to the extent that such a thing exists and is quanitfiable in the first place) is complete horseshit. And it also goes to something that everyone should just admit: we live in a society that makes blacks feel bad about themselves in a way helps cause them to do poorly on the bogus aptitude exams that we administer in order to convince ourselves that we’re a meritocracy.
All of that said, if there really is an “Obama effect” on African-American performance on standardized tests (the study hasn’t even been peer-reviewed yet, so it’s much too early to say), it’s great news for our society.
Sully’s mind got blown because he thinks IQ is a really valid measure of human ability and aptitude. If he were not a damn fool on the subject, he would not have occasion to be so astonished.
They’ve similarly found that women who are told that women are poor at math and science do worse on math/science aptitude tests than women who are told that women lack in this area due to lack of encouragement. Just don’t tell Larry Summers.
Gex, what Summers said was misinformed but I’m not ready to lump him in with the true IQ nuts yet.
Wasn’t Sully the editor of TNR when they pushed that book called something like "The Bell Curve" that all but said blacks were inferior?
Lest we forget and excepting the recent Forbes article, Sullivan really IS still a conservative. Just ask him whether or not people should get foreclosure assistance if you ever doubt that central fact. That he should be shocked by something like this really just fits into who he is.
From Sully’s link:
@DougJ: why not?
Frankly, Summers’ sexism was reason to bring his IQ into doubt in my book. Not because of the sexism itself, but because he went in front of a group of top scientists and suggested that maybe intelligence is a secondary sex characteristic. It was asinine to say something that was guaranteed to piss off all the women present because it was sexist; it was flat out unintelligent to think that a room full of biochemists wouldn’t have, you know, maybe checked into the validity of that already.
What did he expect? That someone would stand up and say "wow! you mean females might be inherently mentally inferior to males? Why, nobody had ever considered that – it runs so completely contrary to millennia of establishment thinking, we none of us even thought to investigate such a radical possibility! Thank you, thank you, Dr. Summers, for helping us to think outside the box, and so on."
Because he made a few off-hand comments about women’s possible inferior abilities. He didn’t write *an entire book* on the subject.
Anyone can say something stupid off-hand. I know I have. It’s why I would I wouldn’t make a good president of a university.
But there’s miles of difference between what Summers did and what the authors of the Bell Curve did…or what Sully did in promoting it.
I think the honkies marking the tests were just afraid to fail their new Nubian Overlords.
Shrug, to this day, I have never figured out the big deal about the damned book. The main conclusion I remember from it was so mundane that it was offensive- within group differences are much larger than between groups differences.
I guess I never paid enough attention to the rest of the controversy around it and should just shut up.
ok, I can see that. Summers’ chauvinism is just as stupid, but he didn’t actively and enthusiastically pursue it over 350 pages of text.
Not sure I buy into it that it was the saying and not the underlying belief that the saying revealed that was the problem with Summers though. I don’t have a problem with him saying something dumb, I have a problem with the sexism that he was dumb enough to reveal.
This is great news for Joe The Plumber.
@John Cole: The point of the scientific studies in the book needs to be distinguished from the point fof publishing and promoting the book. The point of publishing the book was to give a scientific gloss on a racist political argument against affirmative action.
Look to what The Bell Curve is cited as saying, rather than what it actually says.
@jenniebee: Again, the problem is not what Summers said, but the context in which he said it. Summers was excusing the extremely sexist results of decades of tenure decisions, which at Harvard are made within a process that makes it very unlikely for women to get tenure.
Making a half-baked semi-scientific argument is not a big deal (the man is an Economist, after all), but making such an argument in that debate was unforgivable.
Correct me if I am wrong, but didn’t Larry Summers actually point to some kind of study being done in MIchigan on the genetic differences between men and women as the justification for what he was saying? And didn’t he also say something like "I hope to be proven wrong" right after offering up that study as a possible reason for the lack of women at the top of science?
I heard about the reaction to his comments long before I ever found out what he actually said, but after I did hear what he said I honestly didn’t think it rose to the level that a lot of people took it.
J. Michael Neal
@jenniebee: One thing to keep in mind is that the only reason Summers got in trouble is because he was devoting time and effort to promote science education among girls. He may have an underlying sexist view of women’s intelligence; I don’t know. His actions, though, indicate that he is working to counter any such belief.
There are decent reasons to oppose Summers as treasury Secretary. This isn’t one of them.
“intelligence” (to the extent that such a thing exists and is quanitfiable in the first place)
I think the last eight years are proof enough that there IS such a thing as intelligence and it IS a good thing to ask for a reasonable amount of it in your leaders.
[email protected] Futzinfarb
The study has not yet completed peer review, BUT…
Take a look at the plot that is shown as evidence here. (Vanderbilt University press release on Ray Friedman research.)
Then review the critique of graphical analysis here. (Taibbi critique of Thomas Friedman.)
Finally consider the rich irony that both plotters are Friedmans.
I propose a new phenomenon: "Friedman napkin" – a plot of arbitrarily selected parameters with either real or imaginary data used to reach a wildly inappropriate conclusion.
(Note: Both "Friedman plot" and "Friedman graph" already have specific meanings in math and science, hence the slightly more pejorative term "Friedman napkin." On the other hand, the number of syllables is nicely consistent with "Friedman unit.")
You heard it here first.
Brick Oven Bill
DougJ; I am glad that this paper acknowledges historical differences, and am happy to know that the election of President Obama has the power to raise the performance standards of all Americans to the same high level. Think of it.
This makes sense because Almighty God has provided each and every one of us with an identical blank slate mind, upon which our cultures form our being and behavior. And now that we are a multicultural America, perhaps we can get rid of all of the birthright privileges and programs. This would be nice.
Wait, wait: so what you’re trying to convince me is that when you take a race and give them relentless and overwhelming negative feedback about their place in society and tell them they’re worthless and stupid, they start to behave like they have a lesser place in society and start to act as though they’re worthless and stupid?
You silly liberals! Next thing you’ll try to convince us that up is up and down is down.
If the Friedman Napkin effect did exist, then the white folk would score even better than their pre-election peers because they aren’t racists anymore.
This study has nothing to do with Obama, but the conclusion is relevant to this discussion:
Honestly, I didn’t think much of it until I reached this line in the article:
I mean, it’s not a huge revelation to suggest that anti-social "gangsta" posturing is a self defense mechanism to hide insecurity, but it’s interesting to see the underlying rationale (or irrationale) revealed, and it gives a sense of how much transformative potential Obama’s election has for the black community, which I didn’t really believe until now. But we’ll still have to wait and see.
I’m not sure I buy that. Bush was willfully ignorant. I have no idea how he would do on an IQ test. He could turn out to be another Richard Rossner for all we know. That guy wouldn’t make a good president either.
@DougJ: Fair enough. But somehow I see a continuation of a pattern in which privileged white men look at why women and minorities don’t achieve as much as them and suspect that there’s something lesser about women and minorities.
@[email protected] Futzinfarb:
Pardon my ignorance, but what is the meaning of the numbers up the side of the graph in the first link? 12 Whatses?
Brick Oven Bill
But I have a nagging feeling that the birthright privileges will stick around because this ‘test’ sounds a little shaky. There seem to have been 84 targeted test takers split into four groups of, presumably, 21 test takers each. The authors seem to have been able to select from a wide range of test takers, from high school dropouts, to doctors.
Let us meet those who selected the groups:
1. Professor Friedman’s ‘research interests’ are conflict resolution, negotiation, and diversity.
2. Dr. Ko adds a different specialty as her interests are diversity, stereotyping, and power.
3. And lastly, we have Professor Marx, who has a fancier line of work. We will quote him:
"One of my primary lines of research centers on defining and refining stereotype threat: highlighting the distinctions between stereotype threat and priming effects, targets’ and non- targets’ emotional reactions during the threat experience, and how social identity activation differentially affects targets’ and non-targets’ test performance."
I am sure that these are all honorable people and that, when this paper is “peer reviewed”, it will be found that none of those outwardly obvious variables that can help predict human performance were taken into account when selecting the different groups, in an effort to raise the scores of some of those groups.
If I was designing a test like this, I think I would keep the same people and change the questions. But that’s just me.
@J. Michael Neal:
I think this misses Summers’s bigger problem, which was that he already had a bad relationship with the Harvard faculty. There were plenty of faculty members who were looking for publicly acceptable reasons to get rid of him, and his comments provided a pretext. Had he been popular with the faculty, they would have ignored his comments or even defended them as protected by academic freedom.
The origins of ‘intelligence’ as a psychological construct encompassing innate mental ability, and intelligence testing (not the same as aptitude testing, BTW) as a way to measure traits thereof, are shocking by current standards. Weeding out ‘defectives,’ from schools, the military, or society at large, was the suggested pragmatic thrust. Still is, in weird and subtle ways (yup, ‘The Bell Curve’ and affirmative action). New ideas and theories of what intelligence is and isn’t, how best to measure its multiple forms, and better ways to use the resulting information abound in psychology and related fields – it’s now so much richer, more humane, and potentially genuinely pragmatic without being exclusionary or plain hateful.
Which does nothing to explain reliance on dated, culturally-constipated concepts and measures by so many institutions – from public primary schools to ritzy private universities, in healthcare settings, in workplaces, and in government. Blame capitalism: the commonly-used tests are entrenched by nothing as lofty as scientific validity/reliability, or even simple tradition. They endure because a handful of publishing companies make a hell of a lot of money off them, enough to hire lobbyists and to grease the wheels of ethically-challenged researchers. It’s a shame, and a failure on too many levels to itemize in a (already way too long) blog comment.
Oh, and don’t get me started on ‘peer review’… yeesh. Same problems, different topic.
[email protected] Futzinfarb
I don’t know for certain what the numbers on the vertical axis mean but from the context of the reports about Ray Friedman’s research, they are consistent with being the raw (mean ?) number of questions answered correctly.
People have no idea how hard it is to dig out from this. I’m right this minute waiting for part of my admissions model to finish running. I have 3 different ones. There’s the honest one that focuses solely on variables that we have proven correlate strongly to how students perform, we have the one the faculty want because they think that SAT scores mean something other than mom and dad have enough money to send me to Kaplan 11 times, and the one inbetween that I can sometimes get people to pay attention to.
It doesn’t matter how many PhDs I run my numbers in front of, they simply cannot accept that honest model. They agree with it when I’m forcing them to see it directly, but as soon as we talk about policy, they slide right back to test scores.
For the curious, in my discipline, high test scores mean squat. Low test scores tell us something, but once you get a little above the mean score, better scores mean nothing. It basically all come down to how many courses you take in high school and how well you do, weighted more heavily for specific relevant topics. If we eliminated SATs from the model altogether, it wouldn’t suffer much. I should add… if we could interview students, that’d by far be the better means, but it’s just impossible given the numbers for a public school like ours.
And don’t rule out change on the testing issue. UC is pretty close to ditching them altogether, and a few other schools are just waiting for that to happen to follow suit.
Mind the perspective a PhD sociologist?
As John says, the Bell Curve actually only drew conclusions about central tendencies (means or averages), the spread of performance on one intelligence measure (the one used by the military), and the fact that there were lots of intelligent blacks, just not as many as white. So it never said that whites are smarter than blacks, just that a higher proportion of the population is. Now, that work has been heavily critiqued, and properly in my opinion, so I’m not defending it. Also, as someone else pointed out above, it wasn’t the results of that book so much as it was the conclusions that people wanted to draw from it. Murray still claims he’s surprised that racists glommed on to his work, but what the hell did he really expect.
Regarding the study cited by DougJ in the post, the sample sizes are too small, making it less likely to find statistically significant results. So there may have been race differences, just not large enough to be 95% certain it wasn’t due to sampling error. I’d also like to know how they ensured that all groups samples were matched on education and social class. That said, it wouldn’t surprise me one bit if Obama’s election had a positive effect on minority testing outcomes. Another poster noted the effects of stereotypes, another the effects of labeling, and there is bound to be an effect of blacks simply paying more attention to things like politics and economics which could improve testing outcomes.
Summers, now that dude is just lame. There is data showing differences in male-female aptitude for science/math, but nothing "genetic" or "biological", it is all correlation. Summer’s mistake, again noted above, was to say something he knew would piss people off when the results he cited were not robust and demonstrably spurious.
@Martin: I’m a little surprised Martin, I find that SAT and GRE predict college GPA better than any other factor, accounting for the largest percentage of the variance. My data shows it predicts college GPA better than high school GPA does. But I would add that the percentage of the variance explained (r-squared) is not that large, so obviously there is a lot more going on than cold analytical test-taking ability. I agree that we get too hung-up on tests, but for kids that didn’t try in high school because they were bored, the SAT/ACT/GRE is a great way to demonstrate aptitude.
Personally I think if we can get a reliable and valid measure of "effort", we’d have our best indicator of potential performance.
Peer reviews should not be anonymous and they should be published with the papers. Right now there is no transparency or accountability in the process.
Brick Oven Bill
“We were skeptical that we would find any effect, but our results surprised us.”
I hope this is true, because, in my world experience, things usually boil down to money. And, based on this result, think of all of the work there is to do if we can make everybody equal through the process of education or inspiration. Think of the grants and think of the prestige. And then, think of the grants again. And who will fund them. Hey, I know, let’s call it the ‘Obama Effect’.
Now who can we get to do the work?
My bias is that I was a C student in high school, did well on the tests, and did very well in college. I have also witnessed the practical effects of placing unqualified minorities into programs and jobs with unyielding standards. Our society is not doing them any favors. It is breaking good men.
Martin is right that the sample size for Black’s is small (particularly when they were split in the middle into "those who watched and those who didn’t". That said, I can’t imagine they’d have sent it in to a journal without a sig test.
I’m also a little surprised they got a stereotype threat effect given the large range in educational backgrounds of their test takers. Typically you only get them when the people are taking a very difficult test in an area that they are skilled at, and strongly personally invested in. Hm. Weird.
I’m also a little surprised they picked JESP… They tend to be pretty persnickety about really tight designs… plus the finding of stereotype thereat here isn’t very novel, the only cool thing about the data is the funky real world Obama manipulation. I bet it gets gonged and they shop it to a more niche outlet :)
Oh, and DougJ, I think you’re probably right that Bush ain’t outright stoopid
@Brick Oven Bill:
You can’t break good men (or women). What you are observing is people with mid-level aptitude being given the job over other people with mid-level aptitude, accept everyone expects the darkie to be a wunderkind, and are therefore disappointed when he is just like the dopey white boy and therefore proof that affirmative action produces inferior results.
The biggest affirmative action program of all time? The "Greek" system in college.
Brick Oven Bill
I had a guy who had run a big air compressor without motor oil, destroying it, as an E-5. He was promoted to E-7 by the time we met. He was known throughout his professional community as unqualified and people made jokes about him. The CO directed me to establish a paper trail after something bad happened.
I assigned him books to read, we met and reviewed his Plan of Action and Milestones on a regular basis, and nothing changed. This was not a bad guy, but he was just in over his head. He was a strong person, but I witnessed him crying once. Eventually he did something really bad, and the CO went to his boss to explain why this individual was a danger to the command. My CO came back with his head down.
Several months after I had transferred, a friend told me how this individual almost killed the entire unit and was finally fired. It sounds like there might be some people who are decision-makers who might read this. Please set admissions criteria based solely on merit. This guy was a disaster for the perceptions of his ethnic community, if that is what is important.
Ripley FTW. The "intelligence quotient" tests were invented, administered, and refined, explicitly and implicitly, by a turn-of-the-20th-century cadre of male WASPs and WASP-wannabes looking for a shiny new scienterrific reason to ensure that upper-class WASP males would continue to be the measure of human worth. Trying to "fix" these tests to measure the intellectual abilities — much less the potential for success or inventiveness — of a student generation heterogenous beyond anything Stanford or Binet could imagine in their wildest nightmares is like trying to retrofit a steam-powered Daimler horseless carriage so it can replace the Challenger on the next space shuttle mission.
Speaking of the "Obama effect" check out these photos of the lovely couple. I don’t know about you but I almost choke up just looking at them. Grace, maturity, dignity, the loveliness of their affection for each other.
Beats Pickles and the Pretzel Choker all to hell. Thank GOD, we’re past those two idiots.
Intelligence tests are undoubtedly biased in a number of ways. But there should be no doubt that those who do perform well on those tests statistically are much more likely to perform effectively on the job or in graduate school.
Specifically, overall GMAT scores explain 25% of variance in 1st year graduate GPA. Other intelligence tests explain 26% of the variance in job performance in non-technical jobs, and more in managerial (34%) and engineering jobs (31%). Clearly, these tests are not complete horseshit.
So, while pointing out the problems of using intelligence, such as racist exclusion or cultural bias, are correct, there has to be some basis for selecting individuals to attend graduate school and fill jobs. Intelligence is one of those, and it is a powerful predictor that shouldn’t be discarded. It should be improved – either by modifying the various tests to reduce bias, or by providing better lifelong nutrition, parenting support, and education.
As you noted, it’s the within group variance that was by far the largest. Psychologists and sociologists know at least some of the reasons why blacks score less than white (though are rarely willing to discuss it, even when tenured in academia), and it is essentially economic in origin – though there is obviously still some argument about that (as in, blaming the culture). Either way, if we fix economic problems that affect the African American population as well as those at the edges of the middle class and below with substantial investments in the social safety net, education, nutrition, and other substantive issues, I suspect that the group differences of one standard deviation on intelligence tests will go away.
Person of Choler
We’ll see if the effects of this miracle continue.
Shortly after the Inauguration the sea level at the marina near my home dropped about 10 feet and I thought for sure that He had stopped the ocean’s rise as He said He would. But later in the day the water reverted to its earlier level.
The same thing might happen to test scores.
Both of you are correct, by degree. These tests explain a quarter of the variance. What explains the rest?
No one will be surprised to know that the US ranks first among the industrialized nations in social class stability. The correlation between social class birth and social class outcome is the highest among the descendents of the greco-roman-judeo-christian nation states. What do we make of this?
Affordable universities might explain some of the low correlation.
a) Social safety net in general
b) Low birth rates. Native-born people have high parental attention and expectations
c) This explains the variance between individuals, but general curiosity. People who have a willingness to learn and find things out will have the drive to learn and succeed. Incurious people may do a lot of reading, and reading on a high level, and do well on IQ tests from that alone. But they may not be flexible enough to do well in certain kinds of jobs.
I have been that guy. I score huge on standardized tests, but I just don’t do all that well at day-to-day stuff. But for a long time I didn’t know that. As a result I’ve gotten in way over my head at a few jobs and gotten in situations where I’ve broken equipment worth tens of thousands of dollars. As a result, merely a year out of college I had a huge drinking problem, an eviction, bill collectors calling me, and was sleeping on a friend’s couch. It’s now 3 years later and I’m still trying to get my life back together.
I live in America and I’m white so people don’t point to me as an indictment of my race, but still, being in over your head is a miserable place to be.
sorry I botched the blockquoting there.
How odd that among all these mostly clever people discussing what Summers said, not one shows signs of having actually read it.
Jenniebee and others will no doubt take comfort now in learning that he did not say that "females might be inherently mentally inferior to males". What he said was that according to some evidence (which he did not at all claim was conclusive) females were less likely to be at the extremes of intelligence or stupidity. It is an achievement in misreading, whether deliberate or not, to see that as a claim that women are (or might be) mentally inferior.
No sensible person will take my word for this. Rather, any such person will either look it up or just walk away from the question as not worth the effort of fact-checking. The Internet is your friend.
Of course it was silly of him to say it. In a public forum, on an inherently touchy issue, with many enemies who, as has been mentioned, were gunning for him, to advance this idea without especially strong evidence to back it up was such a dumb-ass move that it pretty well proved his disqualification for his post.
@Bootlegger: In particular, women’s aptitude in math/science has improved dramatically since Title IX. Guess women must be evolving very quickly thanks to legislation!
Summers did, however, go on and make the further claim that the intelligence gap (along with "personal choices") explained underrepresentation in the sciences.
What was frustrating about it from my perspective was that it seemed like it was the first time he’d thought about the issue at all, because he seemed to be completely unaware that a personal choice to continue with a career in science is not one that is made in a vacuum, and one that is sensitive to things like institutional discrimination as well as overt sexism. To think that "maybe women just aren’t as smart" is a live hypothesis for the leaky pipeline seems to assume all of that away (as well as assume that success in an academic career tracks smarts.)
What evidence do you have about “on the job” performance? (I’m curious because I haven’t heard this before.)
One of the many reasons I quit reading Sullivan was his rampant racism. The guy honestly believes that Blacks have a lower IQ than Caucasians. Fucking retard.
@DougJ: Grad school reference: Oh, I., Schmidt, F. L., Shaffer, J. A., & Le, H. (2008). The Graduate Management Admission Test (GMAT) is even more valid that we thought: A new development in meta-analysis and its implications for the validity of the GMAT. Academy of Management Learning & Education, 7, 563-570.
Job references: There are a ton of them. A few that are easy for me to reference include:
Hunter, J. E., & Schmidt, F. L. (1996). Intelligence and job performance: Economic and social implications. Psychology, Public Policy, and Law, 2, 447-472.
Hunter, J. E., Schmidt, F. L., & Judiesch, M. K. (1990). Individual differences in output variability as a function of job complexity. Journal of Applied Psychology, 75, 28-42.
Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262-274.
That should give you plenty to chew on if you really want to understand the role of general mental ability in job performance. The first article addresses the issues of minority hiring rates in particular as a function of the general mental ability testing.
It should be noted that the inappropriate use of GMA tests to discriminate has been a problem in the past. Legal challenges have settled the issue by requiring that GMA tests (as well as other employment selection and testing procedures) be job relevant. That requires validation studies, or cross-validation from similar jobs.
@Bootlegger: I don’t you are giving these numbers sufficient credit as far as their predictive validity. Specifically, considering that on a stand-alone basis, unstructured interviews account for 14% of resulting job performance, it seems that improving that number to 25% on the basis of using only one pencil and paper test is pretty impressive. In performance terms and productivity terms, 11% is nothing to discount. In turnover terms, again, it is nothing to sneer at.
When combined with other methods, such as integrity tests or personality tests, the numbers keep climbing. The use of integrity tests with general mental ability tests push the selection-performance validity to 42%, just with those two pencil and paper tests.
There is a limit on how much explanatory power can be obtained with any test (or with any measurement at all). Also, newer methods not yet employed when these studies were first published would have pushed these numbers even higher. There is still variance in performance that cannot be explained with a selection test (such as supervisor interactions with the future employee), though a significant portion of it is measurement error, but that is not the point. If psychologists could completely predict behavior, then you would only increase the amount of testing required to obtain jobs, because the other predictors could be measured as well. Depending on your point of view, that means either that a) people would be perfectly fit into their job, and would stay in a job and perform very effectively because the job fit them very well; or, b) people would be excluded from jobs that they could do on the basis of some test when they should be able to do the job. That is a value judgment. I would point out that as much bull as corporations put out there of that they think hinder their competitive ability, this is an area which has been shown to directly impact individual performance in the workplace, and we should want effective performance by employees, if only to ensure that our industry is healthy, and especially when they have to have some method to choose between many applicants.
As far as class-outcome issues, there is no denying that there is a strong link. My interpretation of this is that we need to invest in our people – all aspects of education, early childhood nutrition, parental support of all kinds. Those types of interventions will undoubtedly expand the abilities of Americans and increase American competitiveness – if we progressives can get it across that it isn’t a give-away, but a long-term project to increase American competitiveness in a global economy.
Statistics indicate consistent underperformance by blacks and Mexicans in the San Diego school system. "Liberals are at a loss to explain this".
As a black, I must say this is the most ridiculous nonsense. One year later, and, Bingo, the smarts have landed into these black kids’ heads. Notice that no percentage of improvement is given, just "black performance improved." This kind of crappy "research" is of no advantage to these kids at all.
And just what does "half-baked" ideas about IQ mean? Since scientists and other researchers are not allowed to explore the subject of IQ group differences, where do you get off dismissing a subject that we know nothing about?
So? Just because "whites are smarter on average than blacks" isn’t as strong a statement as "every white is smarter than every black" doesn’t mean it’s not also inflammatory.
The entire idea of a single intelligence quotient number is itself at best half-baked. It’s probably more like a quarter-baked.
Failure to reject the null hypothesis doesn’t prove it, especially in a study with a small sample size. Also, it’s my understanding that releasing the results of a study prior to peer review is generally regarded as a red flag. The real test is whether the CHANGE in the gap is statistically significant. They could have reported the same results if they only tested one black guy and he happened to score above average.
Notice that this study (or even an abstract) doesn’t seem to be web accessible.
1. Note that they give the scores pre-Obama, but not post-Obama. So there was still a gap, but perhaps only at a 90% significance level rather than a 95% level.
2. Sample size of students is tiny, as you noted.
3. The number of questions is very small.
4. Self-selection effects (it’s an internet study, not a representative sample).
5. Different times of year may catch students in or out of school, different people who may have heard about the study (including from the experimenters, who may have sent flyers or emails to different places), etc.
"What evidence do you have about “on the job” performance? (I’m curious because I haven’t heard this before.)"
The evidence of intelligence (or more specifically, ‘g’ or general ability) on job performance is documented by Professor Linda Gottfredson. For instance, the army uses psychometric tests for the basic reason that they predict a persons ability to learn new information.
You can access her papers from her faculty page:
"To some extent all of this stuff indicates that the notion of using standardized tests to measure “intelligence” (to the extent that such a thing exists and is quanitfiable in the first place) is complete horseshit. "
If that is the case, why are psychometric tests so effective in predicting academic performance?
Also, researchers have uncovered biological correlates to general ability. See Behavioral and Brain Sciences (2007), 30:135-154 Cambridge University Press:
"Overall, we conclude that modern neuroimaging techniques are beginning to articulate a biology of intelligence. We propose that the P-FIT provides a parsimonious account for many of the empirical observations, to date, which relate individual differences in intelligence test scores to variations in brain structure and function."
Also, EEG studies show increased neural efficiency in the cortex of brighter individuals:
"In the field of physiological study of human intelligence, strong evidence of a more efficient operation (i.e., less activation) of the brain in brighter individuals (the neural efficiency hypothesis) can be found"
Also, there has been some interesting recent research by Ian Deary investigating why more intelligent people tend to have greater longevity. Aside from the obvious (they avoid risks/eat healthier food etc) Deary suggests there could be some overall fitness factor, as basic reaction time predicts longevity even better than IQ tests. Gottfredson and Miller also suggest this idea of an overall fitness factor with their recent paper showing a strong correlation between intelligence and semen quality.