Arachnotron

Today's exam is nothing like the first exam.

Nightsd

drusba, that was quite interesting and heavily entertaining, thanks for the history! And thanks for confirming my huge suspicion that the SAT was created by white-supremacist nazis

freakchild12

children, no need to get angry =) LOL jk...

zoe09

Haha thanks everyone. I see that this is a very heated issue for some of us . I ponder the correlation accuracy mostly because: One of my family members knows her IQ. Out of curiosity, this person utilized multiple SAT to IQ charts, all of which she found via Google. On all of them, the SAT puts the IQ about 10 to 30 points higher than it actually is. I can understand how one could score lower on the SAT than his/her IQ allows, but I do not see how the opposite could occur if the SAT accurately measures IQ. The SAT and IQ scores are, in this case, as much as two IQ standard deviations apart - the correlation is terrible. However, my relation does have ADHD (whatever that may mean - so many people "have" it). Thus, IQ factors typically lower in "ADHD individuals" were lower (surprise!). Additionally, she is allowed 50% more time on the SAT. I suppose that the "ADHD" factor skews the results in this situation. Nevertheless, I do not think that one can accurately convert the SAT scores into IQ scores.
However, I do not know if it is possible to convert accurately SAT RANGES to IQ RANGES.
If anyone knows of a chart that converts SAT RANGES (not scores!!!!! Those are terribly erroneous) into IQ scores, I would be delighted to know.
gratia tibi!

tokenadult

"Measure" is not the most fitting word for any mental test. The long story about why you should prefer the industry-standard word "estimate" can be found in

Amazon.com: Measurement in Psychology: A Critical History of a Methodological Concept (Ideas in Context): Joel Michell: Books

Any mental test has error in estimation. There are no exceptions. Each day's test score is a snapshot based on that day's sample of learned behavior. The person who scores highest of all people on earth on some brand of IQ test might not be the person who scores the highest on the SAT--and the other way around.

zoe09

Alright sorry for my misuse of "measure." You get my point. So the huge disparity between the IQ score and the SAT's estimate (sorry if I again employ the wrong vocabulary) is reasonable and within normal error?

Arachnotron

I'd say the disparity would be present moreso if the writing score were taken into account, as it's the least g-loaded section of the exam.

zoe09

^ Agreed. However, a huge disparity remains even when one ignores the W score in the situation that I previously detailed.

tokenadult

Even tests that are very strongly correlated (= for the most part sort test-takers into the same rank order) can still show HUGE differences in scores for some individual test-takers.

zoe09

Well I suppose that that ^ is as good of an answer as exists. Thanks for your time!