I.Q. tests

[quiet bear]

Does anyone know where one can take a fairly decent I.Q test on line?

I've taken a handful of different ones, but I doubt their accuracy, even thought my scores have been fairly consistent. I just doubt how valid they are.

Can anyone suggest something legitimate?

[godfry n. glad]

Read The Mismeasure of Man, by Stephen Jay Gould.

[quiet bear]

I just realized I made the op I.O. tests. This does not bode well.

[Dingfod]

My I.O. is about 3500, at least according to one test.

[quiet bear]

Well, this last one I took said my I.Q. was 143, which meant it was somewhere between 133 - 143. It also said the average I.Q. for someone my age (42), with my education (HS diploma), is between 120 - 130.

So you know what that means, right?

I'm smarter than the average bear. Nyuk nyuk nyuk.

[Sorrel]

I've found that an online IQ test which gives partial results (you pay a fee for the full results) tends to give you a value 5-20 points higher than you'd get from an offline test with no fee structure attatched.

For reliability, you can pick up a paperback book of tests from a bookstore and work through those. On the two occasions I've done that, their answers have been within 2 points of the IQ stated following a supervised test.

The actual numbers depend on the test. Most IQ scales use the number 100 as an average. People with scores between there and 115 would be likely to go into further education. Above 115 (top 10 percent roughly) are likely to go to univeristy/higher education. 130 puts you in the top 2% of the population, and 132 the top 1%.

But I've never come across an online one which doesn't exaggerate.

I just went and took one.

"A person whose IQ score falls in this range is considered to be "gifted". Enter your email address and click a payment button below to buy."



So for validity I'd say do a book of tests - publisher already has your cash so it isn't to their advantage to boost your score to make you want to cough up to hear more about how brilliant you are.

[Julie]

I'm very good at taking IQ tests. I score 160ish on the on-line ones and I score 152 on paper tests.

[Psykottik]

I was in gifted classes since the fourth grade. I took a few IQ tests. In the fourth grade I scored 120. In the eighth grade I scored 139.
(I had the highest IQ at the school, and due to a slight screw-up, I was allowed to see the scores. Shortly thereafter, my head swelled and I began telling teachers that I couldn't possibly be wrong. This landed me in summer school. Thus the downward spiral began....lol...)

Then I took a test online somewhere.....scored 149 and paid for the results.

Hey Julie, I think the range of 150-170 is considered "super-genius." That's pretty decent....

[Moxy]

Originally Posted by quiet bear

It also said the average I.Q. for someone my age (42), with my education (HS diploma), is between 120 - 130.

Aren't they supposed to be adjusted for age? The average should always be 100...unless they're using some scale they made up.

[Legs]

quiet bear, you know I love you big guy ~ but I have to wonder about the IQ of someone who can't find IQ tests online.

show spoiler

[quiet bear]

Har har.

I.Q. smarts doesn't always equal googling smarts.

That was pretty funny, though.

[The Jesus Lawyer]

muscle beach for nerds

that said, i have twice had the test administered to me by a psychologist and done the online test...seems i'm smarter in the digital domain.

[Dingfod]

Same here: IQ test by Psychiatrist (age 14): 126; online (age 48): 146 - 153. I have asked the question before, is it possible to get more intelligent with age? Or does my Jedi memory trick trick IQ tests too?

[quiet bear]

You're certainly not getting prettier. And red lipstick? At your age?

Harlot!

[Godwhacker]

I.Q. tests have quite a history, and I don't feel like writing a book, but, briefly, they were developed primarily for school age children early last century. The original "I.Q." was a ratio of the age the child was "operating" at intellectually over their true age. They took a bunch of children at various ages, gave them the IQ test, found the average and used that as the benchmark.

Thus, if an 8 year old scored at the level of an average 10 year old, their IQ would be roughly 10/8 = 1.25 or 125.

Obviously, however, this method does not work well with late teens and adults, whose intellectual development has pretty much stopped. This, and other factors, led to IQ tests that were just based on norm groups. Thus, the IQ test would be given to a bunch of people of all ages and education levels. An individual's test result would be compared to the average score of the norm group of people the same age level. Average was set at 100 for each group, with a standard deviation of 15 or so. Thus, a person who scored two standard deviations above the mean of their age matching norm group would have an IQ score of 130.

The distribution of IQ scores follows the usual bell curve (in actuality, it doesn't quite at the lower end, but thats another story). Thus, two standard deviations above the mean (IQ score of 130) is better than roughly 98% of all people at that age. Also, real IQ tests usually do not go beyond 3 standard deviations above the mean. Thus, the highest IQ score possible is usually around 145. Anyone claiming to get a score much higher than this on a professionally administred IQ test is BSing you.

I am talking here, of course, about professionally administered IQ tests. These tests are the ones that are used in schools and such. They are extensively normed and researched, and one does need a good amount of training and experience to really know how to administer and interpret and thus they cost $$$ (in fact, some states expressly forbid such tests to be used by anyone other than a trained psychologist).

IQ is often overused and abused. First and foremost, IQ tests tend to only test academic style knowledge, and are heavily weighted towards verbal abilities. Social, musical, and other "knowledges" are not measured by IQ testing.

Second, IQ tests were developed from a Western socio-cultural perspective. Minorities and foreigners don't score as well as a result. For example, many of the questions on IQ tests test knowledge of US/European history and culture (e.g., the WAIS knowledge subtest). Obviously people out of the mainstream or from another part of the world are not going to be as saavy on this kind of test as a mainstream US or european citizen.

Third, a person's mental state at the time of testing affects their performance. A person who is feeling anxious or depressed is not going to score as well as they would if their mood was baseline.

Like any other tool, IQ tests have their uses, we should keep in mind their limitations. Would close by also pointing out that IQ correlates with success only up to about the +2 standard deviation level. Above that, there is little more benefit. In fact, there have been numerous papers done on persons who score very high or near "perfect" on IQ tests and the fact that the majority of these persons tend to have very average lives and jobs. In fact, several were noted to be garbagemen.

[seebs]

I am under the impression that the real range of the tests is about 4 standard deviations for the widely-used ones; so, 160-170 maybe you could measure, but past 170, it's just noise.

[Nightson]

I got a result of 154 on the one I was administered a couple years ago.

[Godwhacker]

Originally Posted by seebs

I am under the impression that the real range of the tests is about 4 standard deviations for the widely-used ones; so, 160-170 maybe you could measure, but past 170, it's just noise.

Not for the WAIS and its kin (the most widely used IQ tests administred by psychologists). +/- 3 standard deviations or so is pretty much it. The main reason for this is that if you go above/below 3 s.d.'s you are either above the 99th percentile, or below the 1st. Usually, psychologists in my area (geropsychology) are not interested in those that score well on WAIS, but rather those that score poorly (i.e., suggesting dementia). Thus, scoring -3 s.d.'s puts you at or below the 1st percentile. 1st percentile is considered "severely impaired" performance. Thus, once a person hits this point, you can't really get any worse, at least not in the classification system we utilize.

In addition to the above, you are correct in that the tests become much less sensitive the farther they get from the average. One test question right or wrong could make the difference between a 140 and a 130, versus five or ten test questions between 125-130 (this is not a direct example, just illustrative). Thus, once you get above 99th or below 1st, you are probably getting lots of error, which is another reason we don't bother going beyond +/- 3 sd's.

Heres a pretty good website to explain the WAIS to non-psychologists:

http://www.psychologicaltesting.com/iqtest.htm

[seebs]

Haven't seen the WAIS. I think when I was a kid it was mostly Stanford-Binnet, which I think goes up to 160. There was something the US military used which they felt was sorta meaningful up to 170, but really past 145-160, it's all sorta made up numbers.