The PISA Rankings and the Role of Schools in Student Performance on Standardized Tests—Posner

International comparisons are tricky, as Becker points out, but the PISA (Programme for International Student Assessment), which tests 15-year-olds for proficiency in reading, math, and science by well designed standard tests conducted in thousands of schools all over the world, is a careful and responsible program, the results of which deserve to be taken seriously. The latest results (which are for 2009) reveal among other things that although the United States spends more money per student on secondary school education than any other country except Switzerland and Austria, Americans’ performance on the PISA tests is mediocre. In the latest tests Americans ranked 17 in reading, 24 in science, and 30 in math. 15-year-old kids in East Asian nations (including Australia and New Zealand), along with Finland, Switzerland, the Netherlands, Belgium, and Canada, outperform the United States in all three subjects. Since 2000, when the PISA tests were first given, the United States has fallen in rank in reading and science, and is unchanged in math.

The rankings tend to be interpreted as measures of the quality of a nation’s pre-collegiate school system (primary and secondary education, since primary education influences performance in secondary schools). But this may be a mistake. Schooling is only one, though doubtless an important, input into performance on the PISA tests. Another is IQ. There have been some efforts to compare IQ across countries, notably by Richard Lynn and Tatu Vanhanen; see their 2006 book IQ and Global Inequality. Their results cannot be regarded as definitive, given significant limitations in the data, but they are suggestive. The authors find that the East Asian countries, which generally rank highest on the PISA tests (including reading—not just math and sciene), have the highest average IQs; the average IQ of Americans is lower because of our large black and Hispanic populations, which have lower average IQs than whites and Asians.

IQ is understood to reflect both genetic endowment and environmental factors, particularly factors operative very early in a child’s life, including prenatal care, maternal health, the educational level of the parents, family stability, and poverty (all these are correlated, and could of course reflect low IQs of parents as well as causing low IQs in their children). The case for very early intervention in children’s development, powerfully urged by the distinguished University of Chicago economist James Heckman, can be understood as an effort to lift IQs in the black and Hispanic communities and by doing so improve the educational performance of black and Hispanic children, including performance on the PISA tests. It is true that Heckman emphasizes noncognitive skills that facilitate learning, but these skills could also increase performance on IQ tests, indicating a positive effect on IQ.

The 2009 PISA test scores reveal that in American schools in which only a small percentage (no more than 10 percent) of the students receive free lunches or reduced-cost lunches, which are benefits provided to students from poor families, the PISA reading test scores are the highest in the world. But in the many American schools in which 75 percent or more of the students are from poor families, the scores are the second lowest among the 34 countries of the OECD; and the OECD includes such countries as Mexico, Turkey, Portugal, and Slovakia.

If IQ is playing a significant role in America’s mediocre showing on the PISA tests, improvements in secondary school education are unlikely to have dramatic effects. The white and Asian kids in American schools are already doing fine, for the most part; the black and Hispanic kids may not do much better until their early childhood environment is improved to the point at which black and Hispanic IQs are raised significantly.

Analysis of the PISA results has revealed some other interesting facts. One is that higher teacher salaries dominate small class size as a factor in high PISA scores. This is a reassuring finding because it suggests that secondary school education can be improved at no net increase in cost, since higher teacher salaries are offset by larger classes—if class size is raised in proportion to increases in teacher salaries, there is no net increase in the school’s cost, and there should actually be a reduction in cost in the long term because a reduction in the number of classrooms reduces the size and therefore cost of a school even if each classroom is larger. Another reassuring finding, in light of all the agitation over charter schools and voucher systems, is that private schools on average do not outperform public schools after adjusting for the quality of students upon entrance and that competition for students does not seem to improve average performance either. Of course these are generalizations across many countries and America’s individualistic culture may not fit them.

This observation is especially pertinent to another finding in the PISA report, which is that poor kids do better in a school that has mostly middle-class kids. Our education system, both public and private, tends as a practical matter to be segregated according to family income and social class. This is a reflection of economic inequality, which is great in the United States and growing.

Becker points out that despite the imperfections of its educational system, America remains preeminent in innovation. This is important but it appears to be due in part to the nation’s attractiveness to immigrants. Many of our innovators are foreign born; increasingly they are Asian. The United States, as a result of generous immigration policies in the nineteenth and early twentieth centuries, also has the largest Jewish population in the world—larger than Israel’s, and a higher percentage of American Jews are Ashkenazi—that is, descendants of European Jewish immigrants. This is significant because Ashkenazi Jews have a significantly higher average IQ than other Americans, including (though the margin is small) East Asians, and, as important, a very strong cultural orientation toward high achievement in business, science, and intellectual fields generally. From the standpoint of innovation, a wide distribution of IQs is more important than the average IQ, because most innovations will come from persons with an above-average IQ, and in scientific and other technical fields from persons with a way-above-average IQ. Moreover, because of the bell shape (normal distribution) of IQ across persons, a higher average IQ translates into a much longer upper tail—the part of the distribution that contains the highest IQs.

If as I am speculating (and I emphasize that it is speculation), IQ is a major factor in school performance, we should hesitate to place too much weight on variance in educational investments, methods, etc. in explaining differences in that performance, relative to genetic and cultural factors and also (and importantly) to economic inequality.

Comments

[EKurtz]

Posner: "IQ is understood to reflect both genetic endowment and environmental factors, particularly factors operative very early in a child’s life, including prenatal care, maternal health, the educational level of the parents, family stability, and poverty (all these are correlated, and could of course reflect low IQs of parents as well as causing low IQs in their children). The case for very early intervention in children’s development ... can be understood as an effort to lift IQs in the black and Hispanic communities and by doing so improve the educational performance of black and Hispanic children.."

But that experiment - enriching the environment for minority youth - is in effect already taking place in the homes of children of affluent minority parents. So in such children we should see a reduction of the performance gap when their test scores (eg on the SAT) are compared to whites with similar backgrounds. But..

QUOTE
...there is a major flaw in the thesis that income differences explain the racial gap. Consider these observable facts from The College Board’s 2006 data on the SAT:

. Whites from families with incomes of less than $10,000 had a mean SAT score of 993. This is 130 points higher than the national mean for all blacks.
. Whites from families with incomes below $10,000 had a mean SAT test score that was 17 points higher than blacks whose families had incomes of more than $100,000.
END QUOTE
"A Large Black-White Scoring Gap Persists on the SAT" The Journal of Blacks in Higher Education, 2006
(http://www.jbhe.com/features/53_SAT.html)

The second point made in the above quote is the most important. It means that, though exceptional blacks can make it to the middle and upper middle class, their offspring will tend to sink to lower SES levels, so that, though we will always have a (relatively small) black middle class, blacks who move into the higher SES levels will not be able to establish middle class lineages. The great, and mostly unexamined, hope of the sixties, the establishment of an enduring, hereditary black middle and upper-middle class, can now be seen to be an illusion. See also this:
http://www.eastbayexpress.com/eastbay/rich-black-flunking/Content?oid=1070459

[EKurtz]

Posner: "Moreover, because of the bell shape (normal distribution) of IQ across persons, a higher average IQ translates into a much longer upper tail—the part of the distribution that contains the highest IQs."

The shape of the IQ curve is an artefact of the way that IQ test questions are chosen. The choice of a normal distribution is primarily a matter of convenience.

Here is the Godfather of IQ testing on the subject:

QUOTE
Item difficulty [in a mental test] is completely under the test constructor's control....Hence, in constructing a test it is possible, within broad limits, to produce almost any desired form of frequency distribution of the raw scores in a given population. If we have no basis for arguing that the obtained scores have true measurement properties in addition to merely having a rank-order correlation with the latent trait that they measure (and this seems to be typically the case for psychometric test scores) the precise form of the obtained score distribution is essentially arbitrary. The very most that we can say in this case is that (within the limits of measurement error) our test scores have some monotonic relation to whatever the test really "measures." If we could truly measure whatever latent variable, such as g, accounts for the variation in the obtained scores on an absolute scale (i.e., one having a true zero and additivity of scale intervals), the form of its population distribution could turn out to be quite different from that of the test scores we have actually obtained. Certain forms of distribution are simply more useful than others, psychometrically and statistically, and it is this consideration that mainly determines the form of the distribution test constructors decide to adopt.
END QUOTE
Jensen "The g Factor," p 101, note 16.

[EKurtz]

Byron M. Roth: "In my book The Perils of Diversity: Immigration and Human Nature, I cite a study by Erling E Boe and Sujie Shin, "Is the United States Really Losing the International Horse Race in Academic Achievement?” (Phi Delta Kappen, 86,9, May 2005, 688-695, online at: http://www.lowellamrine.com/wbqust/documents/BOE.pdf. Quoting the authors:

"...If these [black and Hispanic students] were to perform at the same level of white students, the U.S. would lead all the other G7 nations (including Japan) in reading and would lead the Western G5 nations in mathematics and science, though it would still trail Japan in these subjects."

Steve Sailer has done an excellent job of analyzing the PISA results, see:

"PISA Scores Show Demography Is Destiny In Education Too—But Washington Doesn’t Want You To Know"
http://www.vdare.com/sailer/101219_pisa.htm

Scroll down to the graph; US performance, combined and broken down by race, is shown by the red lines.

It is inevitable that, as the national proportion of Hispanic students increases, the US average in international tests will decrease. This will lead to more hand-wringing among the elite and for further demands that we spend more money to "fix the schools". The underlying cause will of course never be acknowledged.

[JL]

EKurtz, Jensen spends the next pages after your quote arguing that general intelligence is indeed normally distributed in a population.

[Scathach]

It is very interesting to discover that the white population in the US scored higher than whites in European countries. Has anyone begun a study as to why this is? Perhaps higher IQ people are more willing to take risks, as in immigrating and starting a new life?

[EKurtz]

JL: "Jensen spends the next pages after your quote arguing that general intelligence is indeed normally distributed in a population."

It seemed to me that Posner was treating the normal distribution of IQ as a *discovery*, which it clearly is not, unlike, say, the normal distribution of height. Jensen's argument that g is Gaussian is not numeric; it depends primarily on the (reasonable) belief that, like height, the genetic determination of IQ involves contributions from multiple independent loci, which is known to give rise to that sort of distribution. But this is an argument from plausibility, not a scientific verification.
As Jensen says
"Most psychometricians implicitly work on the statistically advantageous assumption of normality, and no argument has yet come forth that it is theoretically implausible or adversely affects any practical uses of g-loaded tests. But the question is mainly of scientific interest, and a really satisfactory answer to it cannot come about through improved measurement techniques per se, but will become possible only as part and parcel of a comprehensive theory of the nature of g. "

There are good reasons to believe that the right hand (ie higher) portion of the IQ distribution conforms to the normal model. As Posner put it: "a higher average IQ translates into a much longer upper tail-the part of the distribution that contains the highest IQs". If mental ability (for which IQ is a proxy) is normally distributed, then a population with a higher average IQ will demonstrate greater competency in cognitively difficult areas than one with a lower average - ie we can predict the high IQ performance of a group from its average. This is clearly evident in the different levels of success of, say blacks and Ashkenazi Jews in science and technology. Ashkenazi Jews are major recipients of Nobels in science, whereas blacks are relatively absent from scientific and technical fields, except in the movies.

[Supra Shoes]

Check out the Power Lab if you want to be able to understand and deal with crisis like we're seeing in Egypt. It's an incredible & insightful experience to make you a better person. They usually have them in April.

[Alfonso Fanjul]

Posner peers into the woods and notes "white and Asians" are "already doing fine; for the most part". He could as well peer into the suburbs and find that "blacks" and Hispanics in those areas are doing pretty well too. He could then peer into the areas of substandard funding with worn out schools and aging textbooks and find "whites, blacks" and Hispanics not faring well in both urban and rural areas.

[Total Protect Home Warranty]

Well that's right that IQ is understood to reflect both genetic endowment and environmental factors, particularly factors operative very early in a child’s life, including prenatal care, maternal health, the educational level of the parents, family stability, and poverty..

[Alley]

I doubt that we can limit immigration of "unskilled" while living next door to an impoverished nation now wracked by the further corruption from the industry of supply our drug needs.

[Health Lab]

This is really interesting, You are a very skilled blogger. I've joined your feed and look forward to seeking more of your wonderful post. Also, I've shared your website in my social networks!

[Spinal Stenosis]

American schools, despite Arne Duncan’s claims, are performing well. US-educated Asians scored 541 in contrast with the Japanese, who scored 520. US-educated whites scored 525 in contrast with Germans, who scored 497. US-educated Hispanics scored 466 in contrast with Mexicans, who scored 425. No data was apparently collected in sub-Saharan countries.

[Danyme]

We are a group of volunteers and opening a new scheme in our community. Your website provided us with valuable info to work on. You have done an impressive job and our entire community will be grateful to you.

[Safe]

I loved as much as you'll receive carried out right here. The sketch is attractive, your authored subject matter stylish. nonetheless, you command get got an impatience over that you wish be delivering the following. unwell unquestionably come more formerly again as exactly the same nearly a lot often inside case you shield this increase.

[health care]

Excellent post. I was checking continuously this blog and I'm impressed! Extremely helpful information particularly the last part :) I care for such information a lot. I was looking for this particular info for a long time. Thank you and best of luck.

[Homesure Services]

Well Becker points out that despite the imperfections of its educational system, America remains preeminent in innovation. This is important but it appears to be due in part to the nation’s attractiveness to immigrants. Many of our innovators are foreign born; increasingly they are Asian..

[Appraiser]

I must say, you've got one of the best blogs Ive seen in a long time.I am waiting for it from such a long time...Thanks for posting..I like this type of blog..thanks..

[Bathroom]

Well good post..Thanks for pointing out the obsession with class size. As if whether one gets to ask one question per day in a 20 student class is really that much different than asking one question per week in a 300 student amphitheate.

[true religion outlet]

In game publishing the majority of work is either developed in-house or commissioned, and it tends to be the case that the ideas that drive that work are internally generated. Original ideas only come from the outside if their creator has a very long track record or if the game is already a proven success in some other format.

[replica watches]

SPARKLE PARTY!!!

[vuitton outlet]

The karma definitely made it's way here!! :D

[cycling jerseys]

I'm glad things are getting better for you :D

[fake watches]

Thank you, they really are!

[wow gold]

YAY! I'M SO HAPPY FOR YOU! I tried donating but it kept saying I did something wrong. I told myself to go back the next day but I totally forgot :( I'm sorry! But I'm glad everyone was able to raise that much for you! CONGRATS!! And your sons are adorable! :D

[armani watch]

Awwww, it's okay!! Thanks for trying!!