This indicates to me that Harvard a) admits people on the basis of their hard work rather than their overblown SAT scores, b) isn’t any tougher to graduate than Podunk U. once you get admitted, and c) there are lots of people at Harvard with low SAT scores, which, depending on your opinion of Harvard, either confirms that Harvard is overrated or gives you reason to admire their graduates.
Based on that, would you care to revise your statement, december?
In any event, the SAT confirms that you have learned some of the things that are necessary to graduate college, but I would still say that as a measure of intelligence I find it lacking. There are plenty of people that would do well on the SATs, if only they had learned the answers, which in many cases is out of their control.
Furthermore, success on an IQ test and/or the SAT does not indicate success in the future. So basing intelligence solely on test results is silly.
Dodging your question once again, the University of California produced a report on the utility of the SAT test http://www.ucop.edu/sas/research/researchandplanning/pdf/sat_study.pdf that basically concluded that SAT was, compared to other tools at their disposal, not very useful in predicting college performance.
Some interesting snippets:
So, without arguing why the SAT I is not a good predictor (the SAT II is an achievement based test), I will defer to the UC system’s analysis to show that it just is.
Personally, I am not a believer in g and so my gut reaction to tests such as SAT I and IQ tests is that they paint an incomplete picture. Of course, an incomplete picture is often better than pure randomness, but just how much better is up for debate. I saw one site claim “Validity estimates for general mental ability (i.e., intelligence or general cognitive ability) indicate that (after correction for attenuation and restriction of range) g accounts for 20% to 25% of the variance in performance, leaving between 75% and 80% unexplained (Jensen, 1998).”
As for my “extraordinary” claim that that many Mensa members are not successful, I base my opinion on extrapolation of the above and pure hearsay from what Mensa members are like (I once considered joining them).
Well, I firmly believe that SAT scores and IQ scores are valid general assessors of certain types of intelligence, but are hardly valid measurements for intelligence as a whole. IQ tests seems to be really good for people who are good at figuring out patterns. The only time I’ve taken a serious IQ test was when I was in 8th grade, and I scored 180. Nowadays, I score anywhere from 140-180 on the online tests. I’m not bragging, there’s a point to come…
HOWEVER, I do not think this makes me at all smarter than someone with an IQ of 120 or 140. I feel outwitted by the majority of posters here at the SDMB, yet from reading the post-your-IQ threads (which I’ve never posted to,) it seems the average score on this board is 130 or so. Like I said, IQ tests seems to test your ability for a very particular type of thinking: pattern recognition (which requires spatial, logical and recall abilities.) I think once you get past a 120 IQ, you may as well lump everyone in together as “above average intelligence.” I don’t think distinctions within this group have any meaning.
Then there’s Marylin vos Savant. Yet more proof that IQ means jack shit.
I don’t believe anyone ever made the blanket assertion that IQ and SAT are unrelated to intelligence. It’s just that you cannot say, with any degree of confidence, that a person with an SAT of 1600 is more intelligent than a person with an SAT of 1550, or, for that matter, a person with an SAT of 1300. Same with IQs. My SAT score was 1300, putting me in a quite different percentile than my IQ score. Obviously, these tests measure quite different things, and are only good for only the broadest of generalizations.
However, if we rounded up a random sample with a large enough sample size of people who had a SAT of 800 and a SAT of 1200, I am willing to be that there would be a statistically significant difference in any measure you care to define ofr “intelligence”. Income, marital stability, happiness, beauty of their spouses, criminal sentences, “creativity” (as measured by standard tests).
Maybe, you could pick out a few things that have a negtavie correlation, but, on the whole, I am willing to bet that most things which are commonly considered strong indicators of intelligence will also be reasonably correlated with SAT/IQ scores. Sure, you could pick out hundreds of examples where a person with a SAT of 800 goes to start a million dollar firm, but the general trend surely runs.
This is just from personal experience BTW, could somebody post a reputable longitudinal study on IQ. I know there were quite a few around.
BTW: When you get to either end of the scale, the picture gets a bit fuzzier. Im not debating that genius and idiot Savant IQs are disproportionate to intelligence, Im generally working in the 80 - 12 range
What the heck do happiness, marital stability, beauty of spouses and creativity have to do with intelligence? Happiness is an attribute of personality and psychology, and I’d reckon you’d find as many (if not more) unhappy people in the higher SAT range. Marital stability relates to interpersonal skills, perhaps “EQ” (emotional quotient) as they call it. Beauty of spouses? My sociology classes seem to find the correlation that people generally marry people of the same attractiveness. IOW, a perfect 10 generally marries another perfect 10. A 7 marries a 7, etc. (Gross generalization, but that’s what we’re talking here.) An exception can be made for the income factor, I suppose. Powerful, rich men and their trophy wives, for instance. (Not to be sexist, but I don’t really know any examples of the reverse.) Creativity? Well, that’s a whole new can of worms. First you have to define that. Anecdotally, I know plenty of people who are very sub-par on standardized tests, yet miles more creative than the average student I saw at university.
An IQ of 160 would be quite high for a genius. After all, the cutoff is either 132 or 140, depending which test you’re using, and the great majority of people above those cutoffs will be below 160. So if Einstein had a 160 IQ, he would have been above average even as geniuses go.
Humungous IQs are tossed around here so often that I don’t think people realize how rare these high IQs actually are. A 160 IQ is REALLY high, extraordinarily high - about one in ten thousand people. When you get up to 180, you’re one in a million, one of the five or ten thousand smartest human beings on Earth.
Most people will never in their entire lives know a person with an IQ that is legitimately 180 or higher. The odds against it are astronomically high.
There is no such thing as a “reputable” way to measure your IQ score on-line. An actual IQ test has to be given by certified test-giver (usually a psychologist) and is under strictly controlled conditions. I had the opportunity to take an IQ test a few months ago when a friend was going for her master’s in psychology and as part of her program took classes to become certified to give IQ tests. At the end of her class she needed a “Test subject” to give a test to, and I volunteered. Technically, my score couldn’t even be considered valid, since I knew the test giver and she might have read the questions differently, or something. I wasn’t supposed to be told my score, but of course I got it out of her. I did well, but not nearly as well as on some on-line tests I had taken. These kinds of tests are where I think many of the impossibly high scores come from.
I am confused as to how many people think the IQ test only measures the ability to solve puzzles. Unless it has changed significantly, this is not the case at all, IMO. The test took over 3 hours to take, and was very comprehensive. I had to apply logic, reasoning, math, english, visual and audio comprehension skills, and even story comprehension, or “finish this story”, “what would this person do” type questions. There were also some “define this word” questions. Only one section was the “puzzle solving” variety, and one for sequence problems. Any tests online I have seen appear to only be one or a few sections of an IQ test, and are not given correctly.
In addition to the different sections, there were several different methods of taking the test. Sometimes I was timed, with and without knowledge of being timed, sometimes I had to read the section myself, other times she read questions to me, sometimes I wrote answers and sometimes gave them orally, there was even some working with blocks and shapes. This was by far the most comprehensive test I have ever taken.
There is no such thing as a “reputable” way to measure your IQ score on-line. An actual IQ test has to be given by certified test-giver (usually a psychologist) and is under strictly controlled conditions. I had the opportunity to take an IQ test a few months ago when a friend was going for her master’s in psychology and as part of her program took classes to become certified to give IQ tests. At the end of her class she needed a “Test subject” to give a test to, and I volunteered. Technically, my score couldn’t even be considered valid, since I knew the test giver and she might have read the questions differently, or something. (I wasn’t supposed to be told my score, but of course I got it out of her;) I did not do nearly as well as on some on-line tests I had taken. Those kinds of tests are where I think many of the impossibly high scores come from.
I am confused as to how many people think the IQ test only measures the ability to solve puzzles. Maybe this used to be the case, but now this is not the case at all, IMO. The test took over 3 hours to take, and was very comprehensive. I had to apply logic, reasoning, math, english, visual and audio comprehension skills, and even story comprehension, or “finish this story”, “what would this person do” type questions. There were also some “define this word” questions. Only one section was the “puzzle solving” variety, and one for sequence problems. Any tests online I have seen appear to only be one or a few sections of an IQ test, and are not given correctly. During some sections, the test-giver had to evaluate my answers and 0-3 points could be scored, depending on how close I was. For other sections there was more than one answer. There were also control questions not used for scoring.
In addition to the different sections, there were several different methods of taking the test. Sometimes I was timed, with and without knowledge of being timed, sometimes I had to read the section myself, other times she read questions to me, sometimes I wrote answers and sometimes gave them orally, there was even some working with blocks and shapes, and deciphering pictures. This was by far the most comprehensive test I have ever taken.
Whether this is a good indication of intelligence, I don’t know, but I think it is better than an SAT test. The biggest difference is that to do well on the SAT you need to be good at taking multiple choice tests, and it is more dependent on the classes you have taken, (when I took the ACT’s I had not yet taken Advanced Algebra, and my math section score suffered because of it), but I think the IQ test measures “basic intelligence” (whatever that is) better…that is, I didn’t need to rely on “book smarts” to take it, just basic principles. It is not possible to study for, and I doubt taking it more than once would affect your score much.
“Define omen”
“A portent of things to come”
She leans forward a bit and I happen to see her scribbling “important things to come”
“I said ‘portent’ not important”
“portent?”
“yes, portent”
“is that P-O-R-T-E-N-T?” (she asked, with a look of confusion, clearly not recognizing the word)
“yes”
(she tilts her pad so I can no longer see what she writes)
“Place these cards in order” (IIRC, One is a man driving a car, one is a woman walking with a hat and a package, another is of the man and woman driving together with the hat flying off and the package in the backseat, the last is a woman walking without the hat and without the package).
“Two combinations were equally likely. Since the woman lost her hat in this picture, the picture of her without the hat must come after, and the picture of her with the hat must come before. Since she also doesn’t have the package, she must have left it in the car, which means that the picture of man driving alone must come before they were driving together since there is no package in the car. However, nothing in any of these pictures suggests the order of the man driving alone and the woman walking with the package.”
“Just pick one”
“There is no basis for me to decide. I suspect the ‘right’ one is the man first, but that’s just giving into a societal gender bias. My answer is that the first two occur simultaneously.”
“What is wrong with this picture” (She shows me a horse with an English saddle that doesn’t have stirrups.)
“Well, I’m a surprised that you are still using that test. Although the stirrups are missing, isn’t that a bit biased to people in this country who exclusively use a western saddle? I remember reading a disproportionate amount of mid-westerners answered ‘pommel.’ Shouldn’t an IQ test be independent of what part of the country you are from?”
“Who wrote Faust?”
“Goethe” (Although I’m not sure, my guess is that my pronunciation of the name in German threw her).
Velma is right. While the usefulness of IQ tests is open to debate, this is only the case when we are talking about one of the major IQ tests (Stanford-Binet, WISC, etc.) administered by a trained professional (usually someone with at least a master’s degree in psychology). The process takes several hours, and unless you are participating in some sort of study then you will probably have to pay for it.
If the test and tester are not up to these standards then the score produced as a result is truly useless and not worth discussing at all. Sadly, I suspect that many of the high IQ scores you see people bragging about actually fall into this category.
It can be confusing to quote an IQ without identifying the scale used. Although the Wechsler and Stanford-Binet give broadly similar results, the Cattell scale used by Mensa exaggerates the high and low ends of the curve.
A Cattell score of 171 only rates 147 on Stanford-Binet.
In the Test the Nation on BBC a few months ago the highest score of the online participants was 148 - (only the online participants were subjected to strict time limits on each question) - and they indicated on the programme that the “genius” level for their scale was on the order of 130.
This site agrees with the estimate of “only” 160 for Einstein on the Stanford-Binet scale, and classes that as Nobel Prize level. It also estimates Goethe as 210.
Of course not. Upon re-reading, I realise I wasn’t clear about what I was trying to state. My point was that happiness, as a psychological or even physiological state, is not necessarily dependent on “intelligence.” Basically, being happy doesn’t mean you’re more likely to be smart and less-than-intelligent people aren’t more likely to be miserable. It’s apples and oranges for me.
Lamia - I really don’t think any IQ score is really worth discussing. What’s the point, really? I don’t know what IQ test I took, but it took about 3 hours, was NOT adminstered by a professional (just a teacher) and completely consisted of symbolic, numerical and alphabetical tests. Most in the form of “complete the sequence.” There were no definition sections, no complete the story, nothing that seemed culturaly biased. Like I said, it was almost completely pattern-recognition and had a few items which included common words and tested the ability for your brain to come up with a sequence of letters which ends the first word and begins the second word. (E.G. MAL ____ ITION. Answer: IGN, malign, ignition.)
It was pretty much a logic test. Why would anyone pay for such a test? What does it help them accomplish? Suppose one of us took one, and we really did score in the 160s. Then what? Does this provide us with anything useful? If you’re intelligent and you learn things quickly, you probably know that already. Why measure it with a yardstick? I’m willing to wager we have a number of 140+s on this board. Does this make them any more successful than the 110s or 120s? Are they richer? Are they really smarter? I highly doubt it.
