Slashdot Mirror
What Does IQ Really Measure?
sciencehabit writes "Kids who score higher on IQ tests will, on average, go on to do better in conventional measures of success in life: academic achievement, economic success, even greater health, and longevity. Is that because they are more intelligent? Not necessarily. New research concludes that IQ scores are partly a measure of how motivated a child is to do well on the test. And harnessing that motivation might be as important to later success as so-called native intelligence."
I recommend Steven Jay Gould's "The Mismeasure of Man" for a thorough look at IQ tests over the ages and how that 99% of the time they are bogus.
I've always felt that the score from an IQ test was actually the real test. Reason being is that some people get a big score, think they're all that and a bag of chips, and let life beat them into the ground because they thought success was predestined. Other people get a low score, think they are stupid, and let life beat them into the ground because they thought failure was predestined. The most successful people, in my experience, see the score from an IQ test, say, "hmm, that's interesting," and then continue to try to do their best at whatever it is they want to do with their lives.
In other words, I feel that IQ tests are largely curiosities that are frequently harmful and only rarely actually useful.
Lets not even get started on the blatant testing demographic bias (target vs actual demographic/etc) that makes the scores skewed against people based on background.
Lemme be an iconoclast here for a moment.
So IQ doesn't measure intelligence. So what? If IQ score is, as claimed, highly correlated with success in life, and if it's measuring motivation and determination rather than intelligence, and if it's motivation that determines success in life, doesn't that make the IQ test pretty damned useful?
Who even knows what "native intelligence" means, anyway? If I've got a test that tells me whether someone understands problems, can find solutions to them, and is motivated enough to carry through, isn't that as useful a definition of "intelligence" as any?
Or to put it bluntly: of what use to anyone is a brilliant mind who doesn't give a shit?
Granted this distinction may be useful, since the remedies (if any) for lack of motivation vs. lack "native intelligence" may be different - or maybe not. I suppose the assumption is that native intelligence is more genetically determined, whereas motivation is more determined by environment, but I find that questionable. Some people have exceptional drive and energy throughout life, even despite circumstances, and most of us don't.
I also take issue with the article:
Why? If IQ scores measure motivation as well as intelligence, then admissions based on IQ already do favor those who want to do the work.
On the upside, you got to post on Slashdot how really smart you are.
The world's burning. Moped Jesus spotted on I50. Details at 11.
This is a subject I've studied before. IQ means different things to different people. Looking over some major tests, I found several schools of thought:
1) Mental quickness and flexibility
2) Factual knowledge
3) Ability to do problem solving
4) Spatial recognition.
IQ is *supposed* to be a general measure of how "smart" someone is (general intelligence), but while it does seem true that general intelligence does exist (doctors can pick up new knowledge in unrelated fields faster than people in some low-level fields), generally the tests just measure specific intelligence.
For example, when trying to test for mental quickness, they might give a kid a jigsaw puzzle to solve (this is what they did on my test in 2nd grade, actually - I spent half my time trying to put it together in unusual ways). But a kid can be "smart" and still be bad at jigsaw puzzles. Since its a timed event, there's also a certain amount of luck involved in how well a kid scores. The difference between "gifted" and "normal" might just be the time span it takes an unlucky kid to try the wrong pieces before he randomly pulls the right piece.
Factual knowledge is also a very difficult to assess subject. I looked over the Titan test (http://www.eskimo.com/~miyaguch/), which is supposed to identify the top 0.0000....01% most intelligent people on the planet. Ok, cool. But one of the answers was an analogy involving Kuru, the prion brain disease contracted by cannibals in Papua New Guinea. I think the test only allowed you to miss a few questions (out of 45) before it ruled you out of the cool kids club. But my objection is, how does knowing what Kuru is make you a smart person? You might just be a trivia buff. And how can you rule someone out for not knowing it? The potential knowledge space for humanity is so impossibly large, that the probability of knowing individual random tidbits of knowledge like that is correspondingly low. How do you differentiate between smart, super-smart, and super-duper-smart? I don't think that any IQ test can provide that level of resolution, really.
More unanswered questions:
Another problem is, of the four categories above, and others people have thought of, which do you assess on an IQ test, and how do you average them together?
Why do we assume that IQ follows a Gaussian distribution?
What role does linguistic fluency and creativity play into the assessment?
I'm not saying that IQ tests are bullshit, but I think people assign them too much value. When you can have the same person take five different IQ tests and get scored between 150 and 230 (http://en.wikipedia.org/wiki/Marilyn_vos_Savant#Rise_to_fame_and_IQ_score), I think we could agree the person is "smart", but beyond that, I don't think tests really mean that much.
Oh yeah? I'm fairly certain my high IQ score is the cause of my intelligence.
The problem with those kinds of tests is they are designed to aim straight for the middle of the bell curve
That's why if you're really smart, they make you take further IQ tests that are aimed progressively higher up. Answering interesting questions can eventually become an exercise in tedium though, so they have to spread it out.
Hah, I got 100.
Perfect score, bitches.
Yes, I'm sure you only scored a paltry 156 because the test was so easy that you got bored.
Ugh, the indignity of taking something designed for normals.
Oh yeah? I'm fairly certain my high IQ score is the cause of my intelligence.
I'm fairly certain your intelligence is the cause of your high IQ score.
And all this time I thought my intelligence was the cause of my high IQ score.
Flawless double whoosh.
Well if I show you a drawing depicting a house, and the house is green, and the question asks "how many sides of the house are green?" and you answer 4 (assuming a box shaped house insofar as you can see), and you learn on a test exam that your answer is wrong, and you should have answered "at least three", you learn something about the nature of the test (i.e. make no assumptions). That knowledge will teach you to take IQ tests smarter, and you'd have done better than someone who went in without that learning. Certainly you can say this is a bad question, but in practice, your score depends on your answer to good and bad questions (just like any exam). The more practiced you are and the more you have learned how to think about common problems, the better you are likely going to do.
I think it's probably pretty hard to develop a test with excellent questions, which are also original and have been verified to be "good" by the standards of the IQ judging process. And thus you end up with a test that doesn't measure what we think we want it to measure. That in itself isn't really a bad thing, you can easily argue that the results speak for themselves (those who score high achieve high on other metrics), but you have to be careful. People who do less well all get lumped together, and some of those people may not have been achievers at that point in their life but might change later for a number of reasons. But they're grouped in with people that have ACTUAL mental, emotional or other disorders, as well as people who are brought up poorly and have no actual hope for a variety of reasons. The net result confirms itself: those who were once good performers, on average perform better than the group of people who were not.
For that reason IQ tests should stay as they are, an academic attempt to measure something we can't really define very well in an effort to understand ourselves. They should not be used for any other purpose, particularly education or employment.
IQ is highly correlated to conventional measures of success in life. My father's a psychologist and he says that IQ tests are instrumental in identifying learning problems (e.g. if you score high on an IQ test, but have poor grades, this can be an indicator that there's a deficiency that needs to be investigated) among other things.
I think the main problem is what it's called. "Intelligence Quotient" is an unfortunate vestige of the bygone era in which its standard testing methodology was devised. The average Joe (like the AC above) assumes that IQ is treated as a comprehensive, innate label of the inner workings of your brain and that's just not how it's treated today.