Arnold Kling  

More on IQ

PRINT
Opportunism... Tolstoy, the Clarkian...

Stephen J. Ceci writes,


Each of us gains every year approximately .3 of an IQ point (6 IQ points every twenty years), and this has been found for nearly 30 nations. It was a secret before Flynn and others made this discovery because the IQ tests were periodically re-normed and the average scores were reset to 100 even if the average person had actually scored a 106. The size of the IQ gain is smaller on tests that are more directly taught in school and home (e.g., vocabulary, arithmetic) and largest on tests that would seem unrelated to schooling (e.g., matrices, detecting similarities).

I think the first sentence is wrong. I think of an IQ test as a test given to a cohort of people, meaning people of approximately the same age. The test places you in a percentile within the cohort. Within a cohort, it is impossible for "each of us" to gain. If my percentile goes up, then someone else's percentile has to go down.

When I think of the Flynn effect, I think of differences across cohorts. Suppose that I took at test at age ten, and the 50th percentile score was 27 out of 40. Thirty years later, another group of ten-year olds takes the same test, and the 50th percentile is now at 29 out of 40. So young folks are on average smarter than old folks were when they were young, but old folks are not gaining anything.

Otherwise, I think Ceci's essay is ok.


Comments and Sharing


CATEGORIES: IQ in Economics



COMMENTS (20 to date)
Troy Camplin writes:

IQ tests aren't "normed" by age, but across the board. And they have all tended to norm upwards, suggesting that IQ is going up -- perhaps as the world becomes more complex. This would also explain the trend upwards as we get older, since we see the world as more complex the older we get. This isn't to say that some people don't become smarter -- someone has to make up for the fact that I seemed to have gained 40 pts in the past 25 years.

Buzzcut writes:

Why does Flynn think that the Flynn effect proves that IQ is bogus? Why can't it be evolution in action, just like the increase in height over time was evolution in action?

Floccina writes:

"The size of the IQ gain is smaller on tests that are more directly taught in school and home (e.g., vocabulary, arithmetic) and largest on tests that would seem unrelated to schooling (e.g., matrices, detecting similarities)."

This could be seen as a knock on schooling.

When is Brian going to be done with that book?

Bruce G Charlton writes:

Actually, I think Ceci's essay exemplifies the bizarre way in which Flynn's work has been misinterpreted as if it challenged or refuted the theory of general intelligence which Ceci summarizes as follows:

• An underlying ability (called g) is needed for all forms of cognitive performance
• g is manifest in any broad cognitive battery such as IQ
• g is related to many types of biological markers and is highly heritable
• Large individual and group differences exist in g
• Variation in g predicts differential life outcomes
• Therefore, variation in life outcomes is at least partly rooted in biological differences in g

The above is true, indeed could be stated even more strongly.

But in reality, the Flynn effect is neither surprising nor mysterious. As Buzzcut implies, the Flynn effect is _exactly_ what would be expected for any biological variable which correlates with fitness.

Indeed, precisely the same patterns of longitudinal change and difference across groups are seen for height, health, symmetry and life expectancy as for IQ.

So it would be astonishing if the Flynn effect did _not_ happen. That the FE has been taken as a challenge to IQ theory is merely evidence either of limited biological understanding, or wishful thinking.

Ceci writes of an IQ mafia, and mocks the 'genetic meritocracy syllogism'. It is not a logical syllogism but a massively tested, predictive and currently un-contradicted scientific theory.

But Ceci does not want want hereditary general intelligence to be true; and Ceci is therefore grateful to Flynn for enabling him to continue denying its truth.

The most revealing sentence of Ceci's was the last: 'Regardless of whether this hunch proves accurate, all of us own Flynn a deep debt of gratitude for complicating what had started to seem like a closed case.'

The IQ hypothesis is as near a 'closed case' as you get in biology. But of course science is never closed, rather it is now time to accept IQ theory (provisionally) and to move on. There is a huge scientific agenda of further exploration and application (while monitoring whether predictions continue to be fulfilled).

Ceci needs to recognize that the type of piecemeal post hoc argument he uses against IQ is the same as that used by creationists arguing against natural selection. It is simply un-scientific to employ double standards in ever-more-desperate attempts to deny what is a clear, comprehensible, powerfully explanatory and extremely important scientific theory.

http://www.udel.edu/educ/gottfredson/reprints/2007doublestandards.pdf

General Specific writes:

This IQ research and the interest in it just seems so--what, boring? Useless? Seriously, I worked for one of the largest and fastest growing companies in the US, interviewed and hired thousands of engineers all over the world, and did a great job at it if I don't say so myself. Yet I read all this IQ stuff and my experience says--useless.

So please tell me: what are we doing to do with this IQ stuff?

Fly Fisher writes:

General: Why the hostility toward, and fear of, IQ discussions?

What is true is already so. Owning up to it doesn't make it worse. Not being open about it doesn't make it go away. And because it's true, it is what is there to be interacted with. Anything untrue isn't there to be lived. People can stand what is true, for they are already enduring it. -- Eugene Gendlin


Intelligence is a real thing. Perhaps you think the common tests don't measurement it well. That may be so, but, in interviewing thousands of engineers, you must have noticed something behind all those eyes that made some stand out from the others. I have educated thousands of engineers, and, although intelligence may not always correlate well with grades, it is clear to me that some people have intellectual tools that set them apart from others. I believe this ability can be measured, and I find it far from boring when the ability is correlated with other characteristics. Far from boring.

Just my opinion, though.

Bruce G Charlton writes:

General Specific - nobody is obliged to find any scientific theory interesting: is natural selection interesting?, game theory?, inclusive fitness?, social intelligence? To me all these are interesting, but I don't expect everyone else to share the interest.

But there are several standard ways of dismissing a theory, and they usually go in three steps.

1. It is not true, absurd, dangerous.

2. Okay - it is true - but trivial.

3. Okay - it is true - and important - but didn't we already know this?

Different people say each of these things in criticizing IQ research. They can't all be correct - indeed, I would suggest that none of them are correct.

Ian writes:

I worked for one of the largest and fastest growing companies in the US, interviewed and hired thousands of engineers all over the world, and did a great job at it if I don't say so myself. Yet I read all this IQ stuff and my experience says--useless.

That's like saying, "I've recruited thousands of professional basketball players from college teams in my time and this hight thing -- in my experience -- is useless. What's with the height fetish?"

Ian writes:

So please tell me: what are we doing to do with this IQ stuff?

That's like asking, "What are we going to do with all of this 'socio-economic status' stuff?" On a wide variety of social outcomes, IQ has more predictive value than SES.

Ian writes:

"IQ tests aren't "normed" by age, but across the board. And they have all tended to norm upwards, suggesting that IQ is going up -- perhaps as the world becomes more complex."

How should we measure "complexity" in a way that would make the proposition falsifyable? Without a measure, the proposition is equivalent to saying, "Some X-factor is raising IQs".

Troy Camplin writes:

Complexity is mathematically measurable, and has been measured by complexity scientists. Complexity is not an "X-factor" -- we know that life in, say, Dallas, TX is much more complex than it is in rural Zambia. For measures of psychosocial complexity, I recommend you look into the Spiral Dynamics group, including Don Beck and Christopher Cowen, authors of "Spiral Dynamics." Also, if you've ever taken an IQ test, you know that 1) they are testing patterns, and 2) pattern complexity increases over the duration of the test. The section of the GRE that MENSA accepts as valid proof of high IQ does the same thing.

General Specific writes:

I still ask: what do you want to do with the IQ research? What are your goals? Please someone answer.

I'm interested in cognitive science, anthropology, etc. I'm not against measuring something called intelligence. But I still question: why are a select set of economists so obsessed with it? I think the world deserves an answer: What do you want to do with it?

Regarding interviewing thousands of engineers: true, we gave them tests--verbal tests, had them work problems on a board, as part of the interviewing process. By asking questions. In person. One human to another. I believe the best and greatest organizations are a network of people who have developed something called trust and experience with one another. No abstract test required.

Not a test. Not an IQ test. An IQ test is so orthogonal to what makes for the success of a great organization. (I didn't say IQ, I said IQ test). One might find correlations, but it would be a waste of time. By and large.

Does anyone really believe that there is a height test in basketball? No. There are people watching basketball players. Obviously there is a correlation, but can you imagine a top notch recruiter with a checksheet that has height on it? Not much of a recruiter. In fact, no need for recruiters. Just a checksheet.

I find it ironic that libertarians are obsessed with what is really a sort of Confucian bureaucratic process--a test to rank people.

So what's the interest? What are you going to do with it? What policy changes will you make? What generalizations will be derived from the results?

And finally, with all due respect, the word or concept "complexity" is not the answer to everything.

Bruce G Charlton writes:

General Specific - go to:

http://www.udel.edu/educ/gottfredson/reprints/index.html

...and read. All your questions will be answered. And you will discover that you are wrong about IQ tests and recruitment. Just plain wrong.

General Specific writes:

Bruce: I'll check the references.

Now can you tell me what you in particular want to do with IQ research?

I'm not against IQ. I'm not against studying IQ. I'm not against the idea that a successful company has employees with something that might be categorized as high IQ.

But I've always been cautious about a number applied to a person. My experience has lead me to believe that grades in school, as an example, are not a great correlation with future success. Rough, but one I would be cautious using. Grades are a factor. They are sampled in a fairly rich long term way. IQ tests--less so. What? 30 minutes? 60 minutes? How long is the IQ test that measures this person?

And my greater concern: I am against James Watson telling us he'd love it if we were all equal but "people who have to deal with black employees find this not true."

Do you know what he meant by that? "People who have to deal with black employees?"

So I'm currently against the idea that this IQ research is going to provide us with much actionable data. Until it does, I think it best a topic for psychologists and/or cognitive science. Let people figure out even what they are doing with it.

So my question, which nobody has bothered to address yet? What specific actions (or future inactions) should be based upon IQ research?

Fly Fisher writes:

James Watson believes in IQ tests.
James Watson said something stupid.
Therefore, I shouldn't believe in IQ tests?

Perhaps I'll extend this logic to convince myself that I shouldn't believe in solid-state transistors. After all, William Shockley believed that we should sterilize people with low IQs, and he received a Nobel Prize as co-inventor of the transistor. As an aside, his greatest achievement -- in my humble opinion -- was annoying his employees so much at Shockley Semiconductor that Robert Noyce and Gordon Moore left to form Fairchild and, ultimately, Intel. Thank you, William Shockley.

Back to the question:

So my question, which nobody has bothered to address yet? What specific actions (or future inactions) should be based upon IQ research?

I'll grant you this: confronting the fact that people make decisions and take actions based on perceptions about a person's intelligence is an extremely uncomfortable thing to do. Most people don't like to talk about it, and, because of that, most people don't want to answer your question.

And most people probably doubt that they have any chance of swaying your position on this, so you shouldn't assume that their silence is due to a lack of answers. That there are so many horrible things that could be done with IQ research doesn't mean that nothing good could come from it. In my opinion, if you are correct in your position that nothing useful can come from the work, then we should want more, not less, people studying and discussing the topic. You should not want to leave the field to the crack-pots.

Ian writes:

"Complexity is mathematically measurable, and has been measured by complexity scientists. Complexity is not an "X-factor" -- we know that life in, say, Dallas, TX is much more complex than it is in rural Zambia"

The ethnic fabric of the Congo is pretty complex.

Is the Washington DC school system more complex than the school system of St. Paul Minnesota?

How can we be sure we are not confusing cause with effect (i.e. that high IQs are the primary source of complexity rather than the reverse)?

Europeans imposed complex cultures on Africa, yet IQs did not rise to the challenge.

Ian writes:

"I still ask: what do you want to do with the IQ research? What are your goals?"

What is the "goal" of temperature research? What is the "goal" of height research? What is the "goal" of weight research?

IQ is a measurement. It does not imply any single agenda or policy.

If you want to ignore the predictive power of an IQ test, that is your business (or should be).

General Specific writes:

Anecdotal aside: I worked with Shockley's son years ago.

Regarding Watson: I think from his comments it's pretty clear that Watson has a low opinion of black people. Watson is not an expert in IQ research, is he? He's a geneticist. Does that make him an expert on IQ? Yet here in this blog (on Caplan's side, not Kling) we hear a grand defense of Watson and his bravery. I think a great deal of damage and ungood can be caused by irresponsible dabbling in IQ research. I think Watson was a dabbler (correct me if I'm wrong) and most of the people here are as well.

Psychologists should certainly be studying intelligence, multifaceted as it is. But I don't think anyone is brave--at this point--by bringing this research into economic theories. It's way too tenuous, and way too prone to ideological and prejudicial manipulation. I just looked at an article on the web that analyzes IQ and national economies, trying to argue around all the inconsistencies of high IQ asians and their backwards economies--until this century--by various forms of hand waving. The article was pretty much all handwaving. Yet it sold itself as "brave." Fighting against PC. I think there is a different agenda at work. Racism.

Granted: There is obviously something called intelligence that met my criteria in hiring all those engineers I mentioned earlier. But I can't quantify it. And I don't think anyone can. Not in any way that would be useful.

I would not use a confucian 60 minute (or however long) intelligence test to make many decisions. I will look at richer forms of information. And I'll look at the environment that I have. Whether it be a community, country, race, etc.

So I ask again: What do economists hope to achieve with their interest in IQ? Nobody wants to answer this. Not a soul. Honestly, I find this disconcerting. Economists should always be looking at cost benefit analysis on any matter, so they should have a benefit that will match cost of this research.

The main good, in a practical sense, that I've seen with this IQ research so far: it does tend to bring the racists out of the closet. They demonstrate their bigotry, whether Shockley or Watson, and then they can be marginalized. Society has more important matters to attend to.

Troy Camplin writes:

Beck and Cowan point out in Spiral Dynamics that you cannot impose complexity on a people -- that will do far more harm than good. The theory of psychosocial complexity they espouse (and which has been proven as much or more than any othe social theory -- not to mentioned successfully used in South Africa post-apartheid) says that psychosocial complexity increases in stages, and that you have to go through each stage. The levels are: survivalist, tribalist, egocentric, authoritative, Enlightenment/Modern, postmodern, integrationist, holistic. When the English and French went into Africa, they attempted to impose a Modern world view on top of tribalist and egocentric world views. You can't skip stages. To try to do so creates disaster. Oftentimes, it resulted in people going down a stage or two. If we want to help people move up, we have to recognize that they have to go through these stages, and then we have to figure out how to get them to do so (they have to come to realize, on some deep level, that their way of seeing the world isn't working for them). There is no way you can tell me that someone living in tribalist conditions is in any way living in as complex a society as a postmodernist living in Paris.

The West went thorugh these stages too. There were tribes. Homeric Greece was egocentric. Plato and Aristotle were authoritative thinkers -- Medieval Europe was an euthoritiative society. I don't think I have to go through the rest. The most advanced societies to date are postmodern, though there are more complex thinkers within those societies. YOu get increases in complexity of society when it becomes necessary -- due to population density, enviornmental factors, etc. People don't change unless they have to. Some of us find ourselves in more complex situations than others -- and that complexity can be increased considerably with the right education, etc. Someone who has to live in a more complex society will learn to recognize more complex patterns. Those who live simple lives do not have to develop more complex patterns of thinking.

Charlie 0893 writes:

Keeping in mind that when I talk about evolution, it is, in this sense, not an origin theory but the theory of evolution as it pertains to continuing life. People are naturally becoming smarter. This makes sense according to the theory of evolution. When two above average IQ people mate, there is most likely to be above average offspring.

Comments for this entry have been closed
Return to top