« Cross-pollination | Gene Expression Front Page | Culture matters »
August 03, 2003

IQ comparison site

The IQ comparison site is worth a look.

Godless comments:

The IQ-SAT conversion tables may be of some utility, but stuff like the Cox guesstimates of IQ are utterly subjective and pseudoscientific. Which is not to say it's not interesting, but it's on par with astrology :)

Posted by razib at 07:01 PM

Too bad there's no GMAT to IQ comparison.

Posted by: Gordon Gekko at August 3, 2003 07:56 PM

Does anyone know if there is any benefit to being a member of a high IQ society? Seems to me that that they are a bunch of losers who have nothing else to be proud of except their high IQs. The high IQ people who are successful are too busy making money or winning nobel prizes or whatever they do to join such a society.

Mensa seems to be the only society that has a big membership, but 98th percentile that's a big joke. You can be pretty dumb and still be a member of Mensa. George W Bush's 1210 SAT score almost qualifies him.

I guess that the REAL high IQ society is your Ivy League alumni organization.

Posted by: Gordon Gekko at August 3, 2003 08:05 PM

Interesting. My SAT score(s) would have predicted a higher IQ than what I actually usually score. If I agree w/ Gordon and lend more credence to the SAT itself I'll have to revise my estimate of where I stand (percentile-wise) upward -

Posted by: bbartlog at August 3, 2003 08:18 PM

98th percentile means what exactly in IQ terms? 120?

Posted by: Johnny Rotten at August 3, 2003 10:19 PM

That SAT-IQ conversion table put me in a ridiculously high percentile... 99.949?!? I don't think the SAT is nearly difficult enough to determine IQs at such high levels.

Posted by: Oleg at August 3, 2003 11:53 PM

The GRE-IQ conversion way overestimates IQ, too - it puts me at 99.992 percentile - and trust me, I just ain't that smart...

Posted by: jimbo at August 4, 2003 05:40 AM

Interesting stuff... Is there any contemporary study similar to Roe's 1952 study of the most eminent US-born scientist ?

I agree with Mr. Gekko, it would be far more interesting to have the actual elite (businessmen, attorneys, physicians, scientists, artists...) tested for their I.Q., rather than college students!
I wonder if there is such a strong correlation between high I.Q. and professional achievement, like it was questioned in a previous thread...

Posted by: eufrenio at August 4, 2003 06:02 AM

98th percentile is IQ 131

Posted by: michael vassar at August 4, 2003 07:03 AM

As I recall the last time I took the SAT there are not enough *really hard* questions to do a good job of intelligence sorting at the high end - so past a certain point the test measures consistency and 'breadth of knowledge' more than raw intelligence. But then again that was after 1995 so maybe they made it easier overall as well. BTW, it is possible to miss a question (a few, even) and still get a 1600 score, at least on today's SAT. For example I believe you can miss one math question and up to three verbal questions and still get 800s in each section. Also, while the SAT may be a fair general intelligence test when administered to 18-year olds it is not as general as a good IQ test; if you give the SAT to a 25-yr old he will have an advantage.
For the curious, I scored as follows on the SAT:
12 yrs old, 1981: 1280 [700M/580V]
15 yrs old, 1984: 1520 [780M/740V]
26 yrs old, 1995: 1600 (one error in verbal)

I retook it in 1995 because I had notions of reentering university and the previous score was of questionable validity to the admissions guys due to being >10 yrs old. It seemed easier than before, but as I said being older gives an unfair edge.

Posted by: bbartlog at August 4, 2003 07:12 AM

Oh, FYI ... MENSA says that SAT scores taken after 1995 can't be used as a reliable indicator of IQ. FWIW.

I tend to agree with that though.

How reliable do people think those internet-based timed IQ tests are? For example, there's one at emode.com and another at iqtest.com. IQ experts?

Posted by: Johnny Rotten at August 4, 2003 09:06 AM

the emode one doesn't have a time factor from what i recall, i think that's moronic. on the other hand-mebee i'm a dumbshit, because i score around the same on it that i normally score on IQ tests, +/- 5 points.

Posted by: razib at August 4, 2003 09:20 AM

Does anyone know how the IQs for GRE, SAT, GMAT, LSAT, etc test-takers can be measured?
The SAT isn't reliable past about 1500 because it isn't intended to be. It's not intended to measure g, but to measure expected success, and it does this somewhat better than IQ tests do (though neither are very reliable at high levels). High IQ produces high SAT scores predictably enough that Cal Tech gets the highest average SAT scores even though they don't consider SAT scores as an admission criterion.

Posted by: Michael Vassar at August 4, 2003 11:32 AM

I joined Mensa to meet men to date. It worked.

Posted by: Jacqueline at August 4, 2003 12:40 PM

The SAT and other college and graduate school entrance exams are the best standardized tests ever created in the history of mankind.

What people think of as pure IQ tests have not gone through the rigorous testing on very large sample sizes that the College Board has access to.

Grading on the tests are now done using Item Response Theory (IRT). IRT is in fact required for grading computer adaptive tests. The GMAT and GRE are both now computer adaptive tests, which theoretically means that they are accurate for testing the entire range of ability scores.

Are these tests IQ tests, or tests of "g"? Yes and no. I would say that the SAT measures in descending order of importance g, math ability, and verbal ability. If the college board really wanted to be fancy, I bet they could do a three factor item response theory grading of the entire test and extract a very good approximation of "g". Better than you would get from Wechsler or Ravens or any other supposedly pure IQ test.

It has clearly been shown that coaching has little impact on SAT scores. Coaching generally raises one's score by less than half a standard deviation (100 points). Because all the test takers have probably done some kind of pre-test preparation, this becomes a very small factor.

If, as someone else said, that the SAT predicts college grades better than IQ tests (I don't know if that's true), it's probably because the SAT measures "g" more accurately than the IQ tests. Of course, the math component is important to math based programs like engineering or science. "g" alone wouldn't tell you if a person is cut out to be an engineer.

Posted by: Gordon Gekko at August 4, 2003 01:46 PM

I tutor SATs, and it's possible to teach someone how to boost their score by MUCH more than 100pts (more like 300pts.) but as far as I know it can't be done well in a group setting (Too personalized, given time and $ I re-teach math from ground zero, basing math on elementary physics (which we have brainware for) rather than vice-versa and teaching speed-reading and process-of-elimination techniques, among other things).
Anyway, does anyone know the score range for e-mode and IQ.com tests?

Posted by: michael vassar at August 5, 2003 12:37 PM

Ask an employer--or a supervisor of grad students--what he or she'd rather have. Moderate ability, good work ethic--or genius screw-off.

Posted by: RJS at August 29, 2003 10:22 AM

Anyone out there with an IQ: 140 and SAT (math & verbal combined): 596... yes, 596!If so, please pass on your thoughts, experiences, resources.

Posted by: Mary joppru at September 22, 2003 02:47 AM

What is intelligence?

The sat does measure quickness but it can be alterred by studying for it or taking a class. The SAT like all standarized timed tests measures the ability to handless stress. This is little mentioned. I feel that one reason that I do well on tests is that I do not get nervous like others. i got a 660 verbal, 730 math in 1996. The classes do make a difference not because you are just practicing but becuase they teach you little tricks and you practice. the tricks ofcourse dont make you a more knowledgeable to smarter or quicker person but help you do well on the test. So it does give an advantage to those that are able to pay for an expensive course such as the princeton review that i took. true, for 50 dollars, I could have found the princeton review book and studied myself if I had the discipline.

The SATs were created with a very egalitarian notion that we should not punish those smart kids from poorer backgrounds that do not have the knowledge and education of the upper class. unfortunately because of the reasons mentioned, it is not a playing field.

Either way, doing well on the SAT is not a true indication of whether one will do well in college. I think a better sampling of that would be high school grades (except for the people that had a life changing experience and decided to start to work hard). getting good grades has nothing to do with quickness in basic math. It has to do with time management, discipline and consistency...and not pissing off the teacher.

there is also the elite few people with low paying jobs that immerse themselves with books and degrees so they can brag about them but are they really happy or have they attained their goals?...well, i guess they have if their goals were to get a lot of degrees and look like a star in academia.

The IQ tests are probably a little like the SATs, but what do they really tell us? You could have someone with a 99% score in the IQ lose all their money in the stock market becuase they were stupid enough to believe all of the fake hype? then they didnt get out becaues they were afraid to lose a little, so they lost everything. Or they could intially support a war like vietnam though evidence existed that it was clearly the wrong war to fight. Or you could have a person with a low score make an invention that no one had actually done successfully before and become rich. Would the 99% IQ person even think of it? Im not sure. Most great ideas dont need a PHD. Im good at that kind of stuff (having ideas) and studied mechanical engineering but just last week I bought a new can opener and for several days couldnt figure out how to work the damn thing. Then my 18 year old friend with not very high SAT scores or life experience came and she quickly figure it out...then she laughed at me. :)

Also, I had a technical and letters and science degree from a top school. I feel that I understand politics very well. I think I know how peace could be brought to the middle east. I have learned a lot about people and life in my travels around the world. I have lots of huge ideas for what inventions would do well making someone rich if taken to fruition with some capital...yeah, im a true genius...except for the fact that Im sitting on my butt right, wearing nothing but my boxers, writing this in the middle of the day, unemployed and broke trying to pick a direction with my life.

It seems that the more I learn about how the world works, the more I am annoyed with the state of humanity....Ignorance is Bliss.

We should be less worried about book smarts and more worried about what traits are needed to define life goals and accomplish them. And these goals should be centerred around bringing happiness to ourselves and others. very little time was given to that topic in the schools I went to...

feel free to email comments on my rambling post.

Posted by: Kaveh at October 13, 2003 09:59 AM